Project

General

Profile

Activity

From 07/27/2014 to 08/25/2014

08/25/2014

09:47 PM Revision fe204e81 (ceph): Merge pull request #107 from dachary/wip-9209-ec-workload
erasure-code: add jerasure k=3 m=1 workload Sage Weil
09:47 PM Revision f7a31191 (ceph): Merge pull request #2319 from dachary/wip-9209-round-up-to
common: ROUND_UP_TO accepts any rounding factor
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:16 PM Bug #9223: ObjectStore/StoreTest.BigRGWObjectName/1 failure on KeyValueStore, firefly
https://github.com/ceph/ceph/pull/2326 Haomai Wang
09:00 AM Bug #9223 (Resolved): ObjectStore/StoreTest.BigRGWObjectName/1 failure on KeyValueStore, firefly
ubuntu@teuthology:/a/sage-2014-08-23_16:02:03-rados-firefly-testing-basic-multi/444749... Sage Weil
08:50 PM Bug #9217: wrong value of osd::op_w_latency
https://github.com/ceph/ceph/pull/2325
Xinxin Shu
08:49 PM Bug #9217 (Fix Under Review): wrong value of osd::op_w_latency
Xinxin Shu
09:36 AM Bug #9217: wrong value of osd::op_w_latency
I think it's a bug:... Sage Weil
02:40 AM Bug #9217 (Resolved): wrong value of osd::op_w_latency
when i use fio to test rbd performance (4kb random write) , i found that osd::op_w_latency which i got from perf coun... Xinxin Shu
07:56 PM Revision 9449520b (ceph): common: ROUND_UP_TO accepts any rounding factor
The ROUND_UP_TO function was limited to rounding factors that are powers
of two. This saves a modulo but it is not us...
Loic Dachary
06:59 PM devops Bug #9233 (Rejected): packaging changes without control directives to allow upgrade
Changes to Debian packaging that move files from one package to another need to be accompanied with changes to debian... Dan Mick
06:53 PM Revision 9720019d (ceph): Merge pull request #2323 from lpabon/fix-quick-guide
Fix syntax error in rst documentation
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh
06:44 PM Revision df3b6fc4 (ceph): Fix syntax error in rst documentation
Signed-off-by: Luis Pabón <lpabon@redhat.com> Luis Pabon
06:08 PM devops Bug #9232 (Closed): disk zap doesnt remove the dmcrypt settings on disk
Well, am really not sure how this is supposed to behave.
deployed a cluster using ceph-deploy and enabled dmcrypt ...
Tamilarasi muthamizhan
06:04 PM Revision cef12993 (ceph): doc: Added links. Clarified namespace comments. Added limitations.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:03 PM Revision d817a6a3 (ceph): doc: Minor changes.
Removed arch-related comments.
Added links to relevant docs.
Moved limitations to user management.
Signed-off-by: Jo...
John Wilkins
06:03 PM Revision 16d946bf (ceph): qa/workunits/hadoop/wordcount: remove bogus rm -r
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:02 PM Revision 5db51d6f (ceph): doc: Added a few comments and links to other relevant docs.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
05:56 PM Feature #7619 (Resolved): objecter: replace client_lock with fine grained read-write locks
Merged, commit:4dbc4c412b5b86b1694b47fff02661e332ef4b3c Yehuda Sadeh
05:49 PM Bug #9211 (In Progress): osdmap blacklist encoding order is nondeterministic
... Sage Weil
05:38 PM Revision 751b3e26 (ceph): rgw: fix test to identify whether object has tail
Fixes: #9226
Reported-by: Sylvain Munaut <s.munaut@whatever-company.com>
Backport: firefly
We need to identify wheth...
Yehuda Sadeh
05:37 PM Bug #9221 (Resolved): LibRadosTwoPoolsPP.PromoteOn2ndRead
Sage Weil
10:16 AM Bug #9221 (Fix Under Review): LibRadosTwoPoolsPP.PromoteOn2ndRead
Sage Weil
09:41 AM Bug #9221: LibRadosTwoPoolsPP.PromoteOn2ndRead
... Sage Weil
07:05 AM Bug #9221 (Resolved): LibRadosTwoPoolsPP.PromoteOn2ndRead
... Sage Weil
05:19 PM Revision c7e1b9e1 (ceph): ceph_test_rados_api_tier: make PromoteOn2ndRead test tolerate retries
If there is an ill-timed connection reset our read could get sent twice.
Weaken our assertion if the read was slow to...
Sage Weil
05:17 PM Revision 8d50abce (ceph): Merge pull request #2311 from onlyjob/master
Fix FTBFS on alpha due to incorrect check on BLKGETSIZE
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
05:15 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.

#9054 already existed for the !old_value.deleted issue specifically, renaming this ticket for the issues found by t...
John Spray
04:24 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.
Seen on next:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-24_02:30:02-rados-next-testing-basic-multi/4467...
John Spray
05:07 PM Revision 31df59f2 (ceph): erasure-code: add jerasure k=3 m=1 workload
It validates that http://tracker.ceph.com/issues/9209 is fixed.
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
05:07 PM CephFS Bug #8878 (Resolved): mds lock cycle (wip-objecter)
Sage Weil
04:28 PM Bug #9218: FAILED assert(peering_queue.empty()) during OSD shutdown

Seen on next:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-24_02:30:02-rados-next-testing-basic-multi/44...
John Spray
03:26 AM Bug #9218 (Resolved): FAILED assert(peering_queue.empty()) during OSD shutdown

http://pulpito.front.sepia.ceph.com/john-2014-08-25_02:37:45-rados-wip-objecter-testing-basic-multi/448053/
<p...
John Spray
04:25 PM Revision 4bff6f34 (ceph): mon: fix occasional message leak after session reset
Consider:
- we get a message, put it on a wait list
- the client session resets
- we go back to process the messa...
Sage Weil
03:58 PM rgw Bug #9227: test.functional.tests.TestFile fail on dumpling on VPS
this appears to be a vps thing; suite passes on bare metal. Sage Weil
03:55 PM rgw Bug #9227: test.functional.tests.TestFile fail on dumpling on VPS
The run Sage was referring to is - http://pulpito.ceph.com/teuthology-2014-08-25_08:45:03-upgrade:dumpling-dumpling-d... Yuri Weinstein
10:45 AM rgw Bug #9227 (Resolved): test.functional.tests.TestFile fail on dumpling on VPS
... Sage Weil
03:56 PM CephFS Bug #9216: mds may regard active clients as stale due to slow pg recovery
I haven't got that far yet, but if I had to guess I'd say it is not about caps, since when this happens, all existing... Alexandre Oliva
10:26 AM CephFS Bug #9216: mds may regard active clients as stale due to slow pg recovery
Interesting. Did you establish the mechanism by which the clients are being stale? Do they have a renew caps request ... Greg Farnum
01:59 AM CephFS Bug #9216 (New): mds may regard active clients as stale due to slow pg recovery
I occasionally get fuse and ceph.ko mounts into weird states, and I can generally track them down to the mds's decidi... Alexandre Oliva
03:52 PM Revision fc5caa07 (ceph): thrashosds: increase osd revive timeout (75s -> 150s)
This is needed when running valgrind.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 12a391ea...
Sage Weil
03:52 PM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
... Loïc Dachary
03:52 PM Revision 12a391ea (ceph): thrashosds: increase osd revive timeout (75s -> 150s)
This is needed when running valgrind.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
03:51 PM Bug #9207 (Resolved): osdc/Objecter.cc: 1074: FAILED assert(op->get_nref() > 1)
Didn't recur with this fix:... John Spray
03:32 PM Revision 3df93026 (ceph): Merge pull request #2318 from dachary/wip-osd-config
documentation: update recovery config options defaults Sage Weil
02:44 PM Bug #9209 (Pending Backport): osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
Sage Weil
10:06 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
"associated ceph-qa-suite workload":https://github.com/ceph/ceph-qa-suite/pull/107 Loïc Dachary
09:22 AM Bug #9209 (Fix Under Review): osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
With "the proposed patch":https://github.com/ceph/ceph/pull/2319 the above workload passes. An inspection of the OSD ... Loïc Dachary
08:07 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
Teuthology must have changed the default recovery chunk for the OSDs at runtime because... Loïc Dachary
07:38 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
"ROUND_UP_TO":https://github.com/ceph/ceph/blob/master/src/include/intarith.h#L31 only works with powers of 2.... Loïc Dachary
07:14 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
In the logs "RecoveryOp::IDLE":https://github.com/ceph/ceph/blob/master/src/osd/ECBackend.cc#L494 shows... Loïc Dachary
06:24 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
... Loïc Dachary
06:07 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
The stripe width for k=6,m=2 is 4224 instead of the 4096 default. It probably breaks a requirement somewhere. ... Loïc Dachary
01:21 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
The teuthology job re-creating the problem is running on teuthology.front.sepia.com in *screen -x -r 17865.loic* Loïc Dachary
01:14 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
The job above with k=2,m=1 passes... Loïc Dachary
01:52 PM Revision 73733ddb (ceph): documentation: update recovery config options defaults
So they match the values from config_opts.h
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
01:29 PM Revision 97b1916e (ceph): CollectionIndex: Collection name is added to the access_lock name
The CollectionIndex constructor is changed to accept the coll_t
so that the collection name can be used to form acces...
Somnath Roy
01:25 PM Revision 7513b9f7 (ceph): Merge pull request #2275 from ceph/wip-9079
msg/Accepter: do not unlearn_addr on bind()
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil
12:13 PM Revision 3e85041a (ceph): rgw: admin ops create user API can not determine existing user
Fixes: #8583
Signed-off-by: Ray Lv <raylv@yahoo-inc.com>
Ray Lv
12:10 PM Revision 469f5ba7 (ceph): Merge pull request #2317 from ashishchandra1/patch-1
Update python.rst
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Loïc Dachary
12:06 PM Revision f7ca1bf1 (ceph): Update python.rst
Signed-off-by:ashish.a.chandra@ril.com Ashish Chandra
10:29 AM CephFS Bug #9212: mon election delays mds beacon
Sage Weil
10:21 AM CephFS Bug #9212: mon election delays mds beacon
Did we identify why it was blacklisted? I don't think we have any tests that should make it that slow or whatever. Greg Farnum
09:30 AM CephFS Bug #9212 (Rejected): mon election delays mds beacon
EBLACKLISTED Sage Weil
10:24 AM Bug #9188 (Rejected): make check fails for setmaxosd
Greg Farnum
12:52 AM Bug #9188: make check fails for setmaxosd
make check works after rebase. False alarm. Anand Bhat
10:13 AM Bug #9140 (Duplicate): [ FAILED ] LibRadosTwoPoolsPP.PromoteOn2ndRead (9913 ms)
Samuel Just
10:06 AM Revision d7fc9a73 (ceph): Merge pull request #2316 from dachary/wip-warning
rbd: fix signess warning
Reviewed-by: John Spray <john.spray@redhat.com>
John Spray
09:43 AM Cleanup #9225 (Closed): check that ROUND_UP_TO is not used with improper rounding values
Will make ROUND_UP_TO generic instead. Loïc Dachary
09:25 AM Cleanup #9225 (Closed): check that ROUND_UP_TO is not used with improper rounding values
The "ROUND_UP_TO":https://github.com/ceph/ceph/blob/master/src/include/intarith.h#L31 function requires the rounding ... Loïc Dachary
09:40 AM rgw Bug #9226 (Resolved): rgw: crash when copying specific objects
as reported on the mailing list... Yehuda Sadeh
09:24 AM Bug #9176 (Resolved): mon: leaked MMonGetVersion
Sage Weil
09:04 AM Bug #9224 (Can't reproduce): osd: segv in dlopen
... Sage Weil
08:54 AM Feature #9222 (New): annotate config options
Implement http://wiki.ceph.com/Planning/Blueprints/Giant/annotate_config_options Loïc Dachary
08:41 AM Bug #8333: ceph_test_rados_delete_pools_parallel: Received fewer notifies than expected: 0 < 1
ubuntu@teuthology:/a/sage-2014-08-23_16:02:03-rados-firefly-testing-basic-multi/445025 Sage Weil
08:32 AM devops Feature #9118 (Resolved): ceph-deploy: Add pre-generated keys to a Monitor
merged commit b00d1fb into ceph:master Alfredo Deza
07:01 AM Bug #9220: objecter doesn't reconnect watch on interval change w/ same primary
ubuntu@teuthology:/a/teuthology-2014-08-24_02:30:02-rados-next-testing-basic-multi/446755 too ? Sage Weil
06:59 AM Bug #9220 (Resolved): objecter doesn't reconnect watch on interval change w/ same primary
ubuntu@teuthology:/a/teuthology-2014-08-24_02:30:02-rados-next-testing-basic-multi/446803... Sage Weil
06:51 AM Bug #9219 (Resolved): lost_unfound test got ENOENT: i don't have pgid 1.e
... Sage Weil
06:24 AM Bug #8777: osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == log.rbegin())
ubuntu@teuthology:/a/teuthology-2014-08-24_02:30:02-rados-next-testing-basic-multi/446938 Sage Weil
06:22 AM Bug #9079 (Pending Backport): osd: bad learned_addr during send_boot
Sage Weil
05:13 AM rgw Bug #8583: rgw: admin ops create user API can not determine existing user
Added new commit per Yehuda's comments, please check it at: https://github.com/ceph/ceph/pull/1950 Xiangyu Lv
04:18 AM Revision f651f93b (ceph): msg/Accepter: do not unlearn_addr on bind()
It is dangerous to set need_addr = true as it means someone may set the
addr to something else (specifically the port...
Sage Weil
01:43 AM Bug #9215 (Resolved): Ceph Firefly 0.80.5 : OSD flapping too frequently
I have not performed any changes to my cluster yet OSD's has started flapping too frequently ( within seconds ) , the... karan singh
12:45 AM Revision a67421a5 (ceph): osd: update handle_osd_map call
I had changed the implementation in Objecter
to avoid a spurious get/put cycle in "osdc/Objecter: fix resource
manage...
John Spray
12:34 AM Revision 5d1b295b (ceph): test: fix prematurely released lock in RadosModel
Should hold this lock while touching context->snaps. Don't
think this has anything to do with #9203 though.
Signed-...
John Spray
12:34 AM Revision 28110ff3 (ceph): osdc/Objecter: hold session ref longer in resend
This is mostly cosmetic: in fact we are getting an extra
ref in _map_session and holding the session lock, so
it's sa...
John Spray
12:34 AM Revision 10efe178 (ceph): osdc/Objecter: fix op_cancel on homeless session
Wrote this block without realizing that op_cancel
takes write lock on session lock, and that operation
is undefined w...
John Spray
12:34 AM Revision ef442928 (ceph): common/Timer: fix deadlock in RWTimer::shutdown
Fixes: #9213
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision e4e7a180 (ceph): librados: prefix msg handler with handle_
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 69612e75 (ceph): librados: refactor watch/notify; return notify error code
Get rid of a level of intermediate classes with confusing names and put
the notify and notify finish logic in a singl...
Sage Weil
12:34 AM Revision 6f3bb55a (ceph): librbd: handle blacklisting during shutdown
This was failing the test_lock_fence.sh test due to
a crash during shutdown when the rbd client was blacklisted.
We ...
John Spray
12:34 AM Revision e7845862 (ceph): librados: separate ::notify return values
There is a return code from objecter for committing
the notify linger op, and then later a code in the
CEPH_MSG_WATCH...
John Spray
12:34 AM Revision f356d0bb (ceph): librados: fix uninitialized values in notify
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
12:34 AM Revision 73e5bb0a (ceph): librados: verbose logging in ::notify
One is interested in knowning specifically which stage
of the operation we are at.
Signed-off-by: John Spray <john.s...
John Spray
12:34 AM Revision 2b944085 (ceph): mon/MonClient: ensure !cur_con == cur_mon.empty()
Keep these in sync.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 6baf36f9 (ceph): librados: shut down monclient after objecter
The objecter ms_handle_reset may call into monc; shut objecter down first.
Fixes: #9130
Signed-off-by: Sage Weil <sa...
Sage Weil
12:34 AM Revision 1e1ee480 (ceph): client: handle (and ignore) MOSDMap
To avoid spamming the log.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 213f6524 (ceph): librados: handle (and ignore) MOSDMap; do not leak
We return true, so we should drop the MOSDMap ref.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 9a5651b3 (ceph): osdc/Objecter: fix resource management
The refactor introduced various reference leaks, and
lacked cleanup in shutdown.
Things done here:
* Reinstate _rec...
John Spray
12:34 AM Revision 8a0382be (ceph): common/RWLock: allow disable lockdep on calls
To support callers with special cases where they
know they can safely take two locks with the
same name at the same t...
John Spray
12:34 AM Revision 2463be5b (ceph): osdc/Objecter: disable lockdep for double lock
There is a special case in _recalc_linger_op_target
where we lock two sessions at once to transfer an op
between them...
John Spray
12:34 AM Revision ad193e1e (ceph): osdc/Objecter: avoid locking two sessions at once
... in close_session
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision e7e7fdb9 (ceph): osdc/Objecter: fix ref counts in *Map_Latest
These contexts were putting before continuing to
use the CommandOp/LingerOp. I added an assert to
reflect that, whic...
John Spray
12:34 AM Revision 10a08b82 (ceph): osdc/Journaler: remove unused waitfor_trim
There was no code path for ever adding contexts to this list.
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision e1e28667 (ceph): osdc/Journaler: use finisher for public callbacks
This is needed because of occasional lock cycles with
external callers doing e.g. write_head.
We do get some weird-l...
John Spray
12:34 AM Revision c2728559 (ceph): mds: convert IO contexts
As of this change, the only thing in the MDS inheriting
directly from Context is MDSContext.
The only files touching...
John Spray
12:34 AM Revision 3487f21c (ceph): osdc/Journaler: fix C_RetryRead locking
Not sure this was ever actually getting used, but
it would have had a problem if it was.
Signed-off-by: John Spray <...
John Spray
12:34 AM Revision a0fd5e0c (ceph): mds/Server: fix wait_for_osd_map calls
These needed a C_OnFinisher and C_IO_Wrapper because they
are called back from the Objecter like an I/O completion.
...
John Spray
12:34 AM Revision 13e6c5f2 (ceph): osdc: Add lock to Filer::Probe
This is necessary now that Objecter can call back
from multiple OSD op completions in parallel: otherwise
we get mult...
John Spray
12:34 AM Revision ed1f8078 (ceph): test: add some missing modelines
Just happened to be editing these and they had
emacs modelines but no vim modelines.
Signed-off-by: John Spray <john...
John Spray
12:34 AM Revision 13fbefad (ceph): osdc: remove unused Blinker.h
Not used anyway, not touched since 2007.
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision 63731120 (ceph): common: fix Finisher perf counter
This was declared as a long running average
but then .dec() was called on it, which
triggers an assertion.
You only ...
John Spray
12:34 AM Revision 28e5c03d (ceph): tools: update for Journaler/Objecter interfaces
Journaler now requires a Finisher: construct one in
MDSUtility.
Objecter now requires separate calls to init() and s...
John Spray
12:34 AM Revision 545baf85 (ceph): librados: update Objecter shutdown
Previously checking for CONNECTED was equivalent to
checking the objecter had been initialized, but since
the separat...
John Spray
12:34 AM Revision 52b0ddcf (ceph): librados: fix race on osdmap initialization
This would cause occasional failures where calls
to lookup_pool immediately after connect() would
fail to find any po...
John Spray
12:34 AM Revision 611d0f77 (ceph): librados: avoid unnecessary locks
Revise wait_for_osdmap to be called outside of RadosClient::lock
and only take the lock if it has to wait for a map.
...
John Spray
12:34 AM Revision a786d2d0 (ceph): messages/MWatchNotify: include an error code in the message
Document the fields, while we are here.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 7a2ec05c (ceph): osd: include ETIMEDOUT in notify reply on timeout
If a notify operation times out (all watchers to not ACK in time), include
an ETIMEDOUT in the final error message ba...
Sage Weil
12:34 AM Revision 59a97850 (ceph): librados: rename watch/notify callback register functions
Make it clear these are for watch OR notify, not just watch. I was
confused.
Signed-off-by: Sage Weil <sage@redhat....
Sage Weil
12:34 AM Revision 1e0387a3 (ceph): client: remove unused C_SnapFlush
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
12:34 AM Revision 6b55c42a (ceph): client: add missing lock in sync_write_commit
This would otherwise assert when entering
the objectcacher.
Also add assertions for all the other Contexts
in Client...
John Spray
12:34 AM Revision 04d01eb3 (ceph): common: assert non-null ptrs in OnFinisher
Callers should not be constructing C_OnFinisher
if they have no context to call back.
Signed-off-by: John Spray <joh...
John Spray
12:34 AM Revision dcdcfd30 (ceph): include: generalize C_Contexts and C_Gather*
... by making them templates so that they can be
used in places that expect a subclass of Context.
The existing name...
John Spray
12:34 AM Revision 0860f1a6 (ceph): mds: add MDSContext subclasses
These allow contexts within the MDS to identify themselves
as either 'internal' contexts (expecting to be called with...
John Spray
12:34 AM Revision 4f3b8032 (ceph): mds: Switch to new context types
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
12:34 AM Revision 85fbddc9 (ceph): mds: rename IO completions with _IO_
If they are IO operations and need to take mds_lock on finish, label them.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 27aed2d2 (ceph): osdc/Objecter: make homeless_session a pointer
Have a non-pointer member that's a RefCountedObject
was awkward, e.g. tripping nref==0 assertion during
destruction. ...
John Spray
12:34 AM Revision 45b355ea (ceph): mds/MDLog: del stale ref to C_MDS_WroteImportMap
(class no longer exists)
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision 29533f3d (ceph): mds: fix C_MDC_TruncateFinish locking
This one needed to take mds_lock, and be invoked
from a finisher.
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision 68d14976 (ceph): mds: fix Server::C_MarkEvent locking
... it should acquire mds_lock.
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
12:34 AM Revision d8939380 (ceph): mds: update mds_lock handling in Locker contexts
For some contexts, we expect to be called back from the objecter/filer
on an I/O completion, so we must take mds_lock...
John Spray
12:34 AM Revision 0cfaeb91 (ceph): mds/MDLog: enable Journaler finisher
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
12:34 AM Revision 0a54f005 (ceph): librados: wait for map on create_ioctx failure
Ensure we have a map so we don't simply complain that a pool doesn't
exists. Only take the lock and wait if we fail ...
Sage Weil
12:34 AM Revision f6cd5fa8 (ceph): mds: use finisher for BootStart contexts
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
12:34 AM Revision 619ea1fd (ceph): mds: update MDLog._replay_thread for locking
... no longer use mds_lock as a guard for condition variable
while waiting for journaler contexts.
Signed-off-by: Jo...
John Spray
12:34 AM Revision 2be623b3 (ceph): mds: fix calls to Objecter::wait_for_map
These were wrong in the earlier commit:
"mds: use lock-safe OSDMap accessors; adjust Objecter wait_for_map call"
Rat...
John Spray
12:34 AM Revision db7369bb (ceph): osdc: locking for Journaler
also remove the lock pointers that JournalPointer load/save
used to require in order to use the objecter.
Signed-off...
John Spray
12:34 AM Revision 61a69c80 (ceph): osdc: Use a finisher from Journaler
Completions from I/O operations (i.e. the objecter) hop
through the finisher twice, because of the three layers of
lo...
John Spray
12:34 AM Revision 8c860f94 (ceph): osdc/Objecter: make OSDMap private
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision c7b7bdd9 (ceph): osdc/Objecter: take over ownership of OSDMap
Instead of taking a pointer to an existing OSDMap in our constructor,
allocate our own, so that we completely own it....
Sage Weil
12:34 AM Revision 641364e7 (ceph): mds: mark objecter completions with _IO_, take mds_lock
For any completion we pass directly to Objecter, make sure we take the
mds_lock in finish(), and mark the class with ...
Sage Weil
12:34 AM Revision 2e308123 (ceph): mds: push objecter completions to a Finisher
Most/all of the MDS completions need to be reentrant (and potentially
call back into the Objecter). Shove them all o...
Sage Weil
12:34 AM Revision 9cb2c815 (ceph): include/Context: add generic C_Lock
Wrap a callback in a Mutex.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision d3b87fb6 (ceph): client: add a Finisher for the Objecter
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 7262da3a (ceph): client: let Objecter dispatch directly
Add Objecter as a direct dispatcher. Drop all of the callbacks and
messages we were passing along. Wrap the IO comp...
Sage Weil
12:34 AM Revision 65f90804 (ceph): librados dispatch fix
Sage Weil
12:34 AM Revision a30f7c63 (ceph): mon/MonClient: fix shutdown typo
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 56689419 (ceph): osdc/Objecter: ignore messenger events before or after initalized
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 83d983f4 (ceph): mds: allow Objecter to dispatch its own messages
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 62365217 (ceph): librados/RadosClient: remove unused get_pool_name()
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision a37fcd85 (ceph): osdc/Objecter: add name-base pool_delete
Avoid a lookup then delete race.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 1848e990 (ceph): osd/OSDMap: make lookup_pg_pool_name const
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 1145b2cb (ceph): librados: drop osdmap pool name cache
We now have a rwlock for Objecter; no more need for a separate cache here.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision af15f9e5 (ceph): osd/OSDMap: return const string from get_pool_name
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision f6981129 (ceph): librados: use {get,put}_osdmap_read() accessors
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 5ec56c32 (ceph): client: use Objecter::{get,put}_osdmap_read() interface
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 9f37f3da (ceph): osdc/Objecter: make wait_for_osd_map() locking safe
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision fd179ead (ceph): test_mon_workloadgen: use Objecter::wait_for_osd_map
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 5328f0ae (ceph): client/SyntheticClient: use lock safe OSDMap accessors
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 85bffa30 (ceph): mds: use lock safe OSDMap accessors for TMAP2OMAP check
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision fad36411 (ceph): mds: use lock-safe OSDMap accessors; adjust Objecter wait_for_map call
We need to handle a race between when we inspect the OSDMap and when we
try to wait for the next map.
Signed-off-by:...
Sage Weil
12:34 AM Revision 0dfe6d6f (ceph): mds/SnapServer: use lock safe OSDMap accessors
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 659909e3 (ceph): osdc/Filer: use lock-safe OSDMap accessors
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision e76cfcc6 (ceph): objecter: don't serialize responses if there's no object name
This implicitly fixes an issue with the list_objects() being reentrant,
and triggers a lock dependency issue. The bet...
Yehuda Sadeh
12:34 AM Revision b4405970 (ceph): objecter: Objecter::Op is refcounted
Cancellation of lingerin ops collided with ops that we sent osdmap on,
needed to add a refcount for correct object de...
Yehuda Sadeh
12:34 AM Revision 9b811a33 (ceph): objecter: split objecter initialization
Separate objecter initialization to non cluster related work (e.g.,
internal data structures, other registrations), a...
Yehuda Sadeh
12:34 AM Revision 795cba94 (ceph): objecter: fix locking
ms_handle_reset() takes the session lock, needed to be able to call
_kick_requetss() with it held.
Signed-off-by: Ye...
Yehuda Sadeh
12:34 AM Revision 8147f9fa (ceph): rwlock: assert on unlock if not locked
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:34 AM Revision a6a88407 (ceph): objecter: take write lock in ms_handle_reset()
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:34 AM Revision 2c2e38e7 (ceph): mds: add finisher
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 09a85438 (ceph): osdc/Objecter: make Objecter a Dispatcher
Note that it's not actually doing it yet, though!
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision 31b0ead9 (ceph): osdc/Objecter: wire up native Dispatcher fast dispatch
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision 0c7dd662 (ceph): osd: let Objecter dispatch directly
No need for our ObjecterDispatcher wrapper, now!
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision c86d2b92 (ceph): librados: let Objecter dispatch its own messages
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:34 AM Revision d83d0f48 (ceph): osdc/Objecter: add OSDMap accessors that are lock-safe
Everybody needs to use these to inspect the OSDMap.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:34 AM Revision d0d16056 (ceph): osdc/Objecter: protect pool ops with rwlock
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:33 AM Revision 8f5cb22f (ceph): librados: cancel operation
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 452dc9e6 (ceph): timer: fix RWTimer shutdown
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 9e6aedf9 (ceph): objecter: a major refactoring
Fixes: #7619
Removed the client_lock (that used to pass in as a param) and replaced
it with a read-write lock (compl...
Yehuda Sadeh
12:33 AM Revision cdb8899a (ceph): objecter: shard completion_lock
Object ops responses are sharded, lock hashed by object name. This
guarantees ordering on the same object. Cross obje...
Yehuda Sadeh
12:33 AM Revision 94119c46 (ceph): test_mon_workloadgen: fix for new objecter interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 1d07ef2f (ceph): rgw: abort early from logging function
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 09af405d (ceph): osd: adapt to new Objecter interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision a2f94d2a (ceph): librados: remove warning
Seems that the current locking scheme is ok. Objecter handles its
locking internally, librados needs to hold its lock...
Yehuda Sadeh
12:33 AM Revision c9d4cdeb (ceph): librados: push snap related functions down to objecter
Need to take appropriate lock.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh
12:33 AM Revision dc8494d9 (ceph): librados: call objecter->handle_osd_map() under lock
The lock is not shared with the objecter anymore so we need to call
it under this lock, otherwise we're going to race...
Yehuda Sadeh
12:33 AM Revision 517f10d0 (ceph): librados: adapt to new objecter interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision e2fa0cf2 (ceph): librados: don't hold client lock in certain operations
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision aac22428 (ceph): librados: don't hold lock in dispatch()
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision bcc69edb (ceph): mds: adapt to new objecter interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 94e18093 (ceph): time: create RWTimer
a timer implementation that uses RWLock
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh
12:33 AM Revision 4c6eb6e7 (ceph): client: adapt to new objecter interface
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 92615ea9 (ceph): common/RWLock: track read/write locks via atomics for assertion checks
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 8f9a9dbb (ceph): common/RWLock: check return values
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 3a399ea8 (ceph): common/RWLock: add get()
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 91c938f5 (ceph): common/RWLock: add explicit unlock() to *Lockers
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh
12:33 AM Revision 33ecf3d1 (ceph): common/RWLock: add lock context
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Yehuda Sadeh

08/24/2014

11:40 PM Revision 74ce4f26 (ceph): rbd: fix signess warning
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
10:50 PM Revision d848a400 (ceph): Merge pull request #2138 from xiaoxichen/fix_crush_ruleset
mon/OSDMonitor : Use user provided ruleset for replicated pool
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Loïc Dachary
07:49 PM Revision f7c0001a (ceph): common: remove spurious uint32_t in buffer.c
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
04:10 PM Feature #8343 (Closed): please enable data integrity checking (by default) / silent data corruption
Loïc Dachary
04:06 PM Bug #8349 (Resolved): env-vs-args unittest is racy
Fixed by https://github.com/ceph/ceph/commit/3230060f07c738383cc1034a99d60d2ad369560f Loïc Dachary
03:32 PM Support #8462: related to integrity of objects
Loïc Dachary
03:12 PM Feature #7238 (Fix Under Review): erasure code : implement LRC plugin
The rados test work (no thrashing). Loïc Dachary
02:57 PM Support #8310 (Closed): Most pgs stuck stale, no osds reporting them, repair ineffective
Loïc Dachary
09:25 AM CephFS Bug #9212 (Won't Fix): mon election delays mds beacon
ubuntu@teuthology:/a/teuthology-2014-08-22_23:04:01-fs-master-testing-basic-multi/444359... Sage Weil
08:36 AM Bug #9211 (Resolved): osdmap blacklist encoding order is nondeterministic
... Sage Weil

08/23/2014

05:00 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.
fwiw the reproducer hits a crash on firefly, but not emperor or dumpling. A fair bit changed in ceph_test_rados for ... John Spray
03:13 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.

So it turns out that ceph_test_rados is also crashy on master, as I found when I took my reproducer for this issue ...
John Spray
03:53 PM rbd Bug #9210 (Resolved): osdc/ObjectCacher.cc: 529: FAILED assert(i->empty()) on fencing test shutdown
... Sage Weil
12:41 PM Revision 6ad8e61a (ceph): Fix FTBFS on alpha due to incorrect check on BLKGETSIZE
Ceph FTBFS on Alpha with:
~~~~
libtool: compile: g++ -DHAVE_CONFIG_H -I. -D__CEPH__ -D_FILE_OFFSET_BITS=64 -D_REENT...
Dmitry Smirnov
11:50 AM Feature #7238: erasure code : implement LRC plugin
Loïc Dachary
11:25 AM Feature #7238 (Fix Under Review): erasure code : implement LRC plugin
Although thrashing tests using an LRC pool fail, I believe this is due to the size of the pool rather than the plugin... Loïc Dachary
11:29 AM Bug #9209: osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
The same YAML file run against firefly 0.80.5-171-gca3ac90-1trusty instead of master succeeds. Loïc Dachary
11:23 AM Bug #9209 (Resolved): osd/ECUtil.h: 66: FAILED assert(offset % stripe_width == 0)
Using ... Loïc Dachary
12:39 AM Revision 0ec5bd1a (ceph): mplement ceph.created_pool
ceph.created_pool allows the user (via yaml lines) to add pools
that the ceph_manager knows about.
Fixes: 9091
Signe...
Warren Usui

08/22/2014

11:52 PM Revision 25542430 (ceph): mds/Server: rename perfcounters
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:31 PM Revision b0cc869a (ceph): mds: rename a bunch of metrics
Also clean out a few that were unused and some dead code.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
11:14 PM Revision 31ef1a97 (ceph): mds: set l_mds_req on client request
This is one input for the load balancer.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
10:28 PM Revision 9e539723 (ceph): set boto timeout, too, for s3tests
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
10:28 PM Revision 9d466aa1 (ceph): tasks/s3tests: push boto config with idle_timeout setting
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:26 PM rgw Bug #9208 (Resolved): rgw: civetweb does not drain request buffer correctly
When radosgw returns an early error without reading the request content, we need civetweb to drain the buffer so that... Yehuda Sadeh
06:10 PM Revision 8d406001 (ceph): Merge branch 'wip-weak-refs'
Sage Weil
06:08 PM Revision 06682c4b (ceph): vstart.sh: debug rgw = 20 on -d
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:08 PM Revision 8a212170 (ceph): Merge pull request #2292 from lpabon/rgw-doc
vstart.sh to enable support for simple Rados
Tested-by: GWabhishek.lekshmanan@gmail.com
Sage Weil
05:24 PM Subtask #6478 (Rejected): ErasureCode : XOR plugin
This has been obsoleted by the work on the ISA plugin. Loïc Dachary
05:22 PM Feature #7238: erasure code : implement LRC plugin
Fixed a bug that made the plugin incorrectly claiming it could not recover when the last OSD was out, running tests a... Loïc Dachary
05:05 PM Revision 850242ca (ceph): rgw: separate civetweb log from rgw log
The civetweb log now is independent from the rgw log.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh
05:05 PM Revision 00c677b4 (ceph): rgw: use a separate callback for civetweb access log
Access log is separate from the regular civetweb logging. Also, changed
the log level for the regular logging as it's...
Yehuda Sadeh
04:17 PM Revision f246b569 (ceph): common/shared_cache: dump weak refs on shutdown
Before asserting, dump what the refs are along with the counts.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
04:13 PM Revision 6f15a4ca (ceph): Merge pull request #2304 from majianpeng/fix1
small bug fix for journal
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:04 PM Revision 6cf583c4 (ceph): common/shared_cache: take a cct
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
03:46 PM Revision 997f766a (ceph): Merge pull request #2291 from ceph/wip-fingerprint
mon: add a unique clsuter fingerprint
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil
03:17 PM Revision d6aa7833 (ceph): Merge pull request #2307 from dmsimard/crushdoc
Doc: Add config option to turn off crush update
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Loïc Dachary
03:09 PM Bug #9207 (Resolved): osdc/Objecter.cc: 1074: FAILED assert(op->get_nref() > 1)
ubuntu@teuthology:/var/lib/teuthworker/archive/john-2014-08-22_10:24:47-rados-wip-objecter-testing-basic-multi/441988... Sage Weil
03:04 PM rgw Bug #9206 (Resolved): rgw: cross rgw message headers filtered by apache 2.4
apache 2.4 filters out header fields that have underscores in them. Need to convert underscores into dashes. Yehuda Sadeh
03:00 PM Revision 2ba081d4 (ceph): Merge pull request #2288 from xiaoxichen/fix_add_simple_ruleset
CrushWrapper: pick a ruleset same as rule_id
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Loïc Dachary
02:55 PM Revision 2c274851 (ceph): Doc: Add config option to turn off crush update
Signed-off-by: David Moreau Simard <dmsimard@iweb.com> David Moreau Simard
02:52 PM Bug #9205 (Resolved): osd: notify ops reordered
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-21_11:40:02-upgrade:dumpling-x:stress-split-master... Yuri Weinstein
01:23 PM devops Feature #9136 (Resolved): ceph-deploy: use pre-existing ceph.conf
merged commit 2781538 into ceph:master Alfredo Deza
01:22 PM Revision 78e84f34 (ceph): CrushWrapper: pick a ruleset same as rule_id
Originally in the add_simple_ruleset funtion, the ruleset_id
is not reused but rule_id is reused. So after some add/r...
Xiaoxi Chen
12:44 PM devops Feature #9118 (Fix Under Review): ceph-deploy: Add pre-generated keys to a Monitor
Pull request opened https://github.com/ceph/ceph-deploy/pull/235 Alfredo Deza
12:02 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.
Does not reproduce very often, but eventually caught in the act with debug turned up.
The oid in the asserting ope...
John Spray
06:39 AM Bug #9203 (Resolved): ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limi...

http://pulpito.front.sepia.ceph.com/john-2014-08-22_02:21:21-rados-wip-objecter-testing-basic-multi/440722/
http:/...
John Spray
11:28 AM Bug #7995: osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
added patches to master that will dump the weak_refs on shutdown Sage Weil
06:32 AM Bug #7995: osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
http://pulpito.front.sepia.ceph.com/john-2014-08-22_02:21:21-rados-wip-objecter-testing-basic-multi/440850/
http://p...
John Spray
06:24 AM Bug #7995 (New): osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
This is happening again:
http://pulpito.front.sepia.ceph.com/john-2014-08-22_02:21:21-rados-wip-objecter-testing-b...
John Spray
11:15 AM Bug #8736: thrash and scrub combination lead to error
This needs to be prioritized.
Confirmed, logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-21_11:...
Yuri Weinstein
10:30 AM Revision 18357782 (ceph): needs network or it will not start if enabled
example error message: "unable to find any IP address in networks: 10.2.27.0/25", so the service wont start Marco Garcês
10:19 AM Bug #8985: "[WRN] map e9 wrongly marked me down" in upgrade:dumpling-x-firefly---basic-vps suite
Yuri Weinstein
06:36 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
The stack trace created by the minimal script is different from the one reported above, but it fails at the same poin... Loïc Dachary
05:51 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
The problem does not show if waiting after the object is inserted. It is a race condition.... Loïc Dachary
05:25 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
For the problem to show the file being removed has to be the primary. Loïc Dachary
05:06 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
Even simpler and does not require root privileges... Loïc Dachary
04:56 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
The following reproduces it reliably on my laptop:... Loïc Dachary
03:47 AM Fix #8914 (In Progress): osd crashed at assert ReplicatedBackend::build_push_op
Thanks for the update, will try again :-) Loïc Dachary
05:33 AM Revision 8c69054b (ceph): osd: make coll_t::META static to each file
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
02:57 AM CephFS Bug #4545: error creating empty object store. Invalid argument.
i maybe found the problem.
before you mkcephfs,you should ensure the dir(/var/lib/ceph/osd/ceph-0) empty.
once i wr...
cache china
02:32 AM Bug #9202 (Can't reproduce): Performance degradation during recovering and backfilling
From recent test and analysis, we find slow requests mainly happen at 2 patterns during recovering and backfilling.
...
Zhi Zhang
01:05 AM Revision 5234abfa (ceph): Merge pull request #2273 from dachary/wip-mailmap
mailmap updates Loïc Dachary
12:48 AM Revision 5f1e1e0e (ceph): Merge remote-tracking branch 'gh/wip-filejournal'
Reviewed-by: Sage Weil <sage@redhat.com>
Tested-by: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
Sage Weil

08/21/2014

11:39 PM Revision 825212a5 (ceph): Merge pull request #104 from dachary/wip-8933-rgw-erasure-code-profile
rgw: add erasure_code_profile configuration Tamilarasi muthamizhan
11:27 PM Revision e5c5bcf9 (ceph): rgw: add erasure_code_profile configuration
If erasure_code_profile is present at the same leve as ec-data-pool, it
is used to override the default hard coded pr...
Loic Dachary
11:12 PM rgw Feature #8911: RGW doesn't return 'x-timestamp' in header which is used by 'View Details' of Open...
Thanks Luis... actually its a new feature request not a bug. Since we want one to one headers mapping between Swift a... Ashish Chandra
10:24 PM Revision 493577bf (ceph): mds: fix crash killing sessions without conn
Bug was introduced in 54eca56695 (mds: use Connection::mark_down()).
Used to call SimpleMessenger::mark_down, which ...
John Spray
10:22 PM Revision dcf8c037 (ceph): mds: logging in SessionMap
(move definitions to .cc to get it)
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
10:09 PM Revision 5c29ae6b (ceph): tasks/ceph: add ceph.stop task
So that we can explicitly stop daemons on demand. Useful
for MDS tool tests that want the MDS daemons not to be runn...
John Spray
10:09 PM Revision 1e7bfb84 (ceph): tasks/workunit: fix log message
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
10:09 PM Revision 8211d83d (ceph): tasks/ceph_fuse: enable umounting from config
This is for any test config that needs to run
some workunit with clients unmounted. It allows
you to go toggle the m...
John Spray
10:09 PM Revision bb52a973 (ceph): tasks/mds_client_recovery: network freeze test
This is about testing the CephFS client's handling
of losing connectivity to the MDS.
Fixes: #7810
Signed-off-by: J...
John Spray
10:09 PM Revision d001cc27 (ceph): tasks/mds_client_recovery: use existing clients
This will enable using .yaml changes to switch this
guy over to use kcephfs client once the teuthology
code around it...
John Spray
10:09 PM Revision 1855e094 (ceph): suites/fs: add client recovery
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
10:09 PM Revision 83a745d2 (ceph): gitignore: ignore vim temp files
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
10:09 PM Revision 0d37b8e4 (ceph): suites/fs: add cephfs_journal_tool test
Fixes: #4886
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
10:06 PM Revision 82ac61fc (ceph): Merge remote-tracking branch 'gh/wip-lttng'
Sage Weil
10:05 PM Revision 01ce2497 (ceph): Revert "Merge pull request #2253 from adamcrume/wip-lttng"
This reverts commit 5b9ade5058f72116eb3dc920f442361c2a33b46b, reversing
changes made to b0aa846b3f81225a779de00100e15...
Sage Weil
09:58 PM Revision 0b1a4328 (ceph): Merge pull request #2301 from ceph/wip-9176
mon: fix occasional memory leak; clean up dispatch return codes
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil
09:11 PM rgw Bug #9201 (Resolved): rgw: bad object with different pool alignment
http://qa-proxy.ceph.com/teuthology/sage-2014-08-21_17:03:27-rgw-master-testing-basic-multi/440046/teuthology.log
...
Yehuda Sadeh
08:43 PM Revision 7e4b525e (ceph): Merge pull request #2274 from ceph/wip-9144
filestore: fix force_sync race with mount + umount + mount
Reviewed-by: David Zafman <dzafman@redhat.com>
David Zafman
08:14 PM Revision 5b9ade50 (ceph): Merge pull request #2253 from adamcrume/wip-lttng
RBD-replay
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
08:10 PM Revision 3ced97b2 (ceph): mon: make dispatch(), _ms_dispatch() void
They always return true.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:07 PM Revision 610f4be5 (ceph): mon: always process the message in dispatch
Nobody should be chained after teh mon, so we can safely drop any message
we don't understand.
Signed-off-by: Sage W...
Sage Weil
08:05 PM Revision 19df386b (ceph): mon: fix occasional message leak after session reset
Consider:
- we get a message, put it on a wait list
- the client session resets
- we go back to process the messa...
Sage Weil
06:14 PM Revision 675b0042 (ceph): mon: add a cluster fingerprint
Generate it on cluster creations with the initial monmap. Include it in
the report. Provide no way for this uuid to...
Sage Weil
06:14 PM Revision b245d600 (ceph): mon: generate cluster_fingerprint if null
This triggers after an upgrade of a legacy cluster that has no fingerprint.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:57 PM Revision 07ab36f9 (ceph): lttng: Remove tracing-specfic local variables when lttng disabled
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 9673e2f3 (ceph): lttng: Split --with-lttng and --with-babeltrace options
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 76160925 (ceph): rbd-replay: Add unit test for Ser
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 552c4b44 (ceph): rbd_replay: Add unit test for batch_unreachable_from
This requires a fair amount of code reorganization, since the types in
the function signature were previously not in ...
Adam Crume
05:57 PM Revision 7ff6ed11 (ceph): lttng: Only require lttng for Fedora >= 20 and RHEL == 6 for RPM distros
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 861e6a39 (ceph): lttng: Update packaging
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision a35b5d2c (ceph): lttng: Don't build trace headers when lttng is disabled
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 42d633f3 (ceph): lttng: Remove -lboost_regex from unit test
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision d1064bff (ceph): lttng: Fix build error on 32-bit systems
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e419ba60 (ceph): lttng: Don't build trace libraries when lttng is disabled
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 84bfe910 (ceph): lttng: Test for sane lttng-gen-tp
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 9c388f08 (ceph): lttng: Check that the main babeltrace header is present
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 141f37a6 (ceph): lttng: Check for babeltrace when autoconfiguring lttng
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 230c5b8e (ceph): lttng: Only install rbd-replay-prep when enabled
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 3f1917c0 (ceph): lttng: Add informational messages to configure.ac
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e18748ed (ceph): rbd-replay: Add man page, and add binaries to Debian packaging
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 0f052f8c (ceph): rbd-replay: Convert prep-for-replay.py to rbd-replay-prep.cc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 0cd2651c (ceph): rbd-replay: Add man page and package binaries for rbd-replay-prep
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e555867e (ceph): rbd-replay: Move tracepoints from internal.cc to librbd.cc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 18c081e5 (ceph): rbd-replay: Remove extent tracepoints and inline extents
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 660f1d68 (ceph): rbd-replay: Fix bug in rbd-replay-prep, Thread::issued_io with wrong IO
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 8d71b6c1 (ceph): lttng: Remove tracing from libcommon
This is a short-term fix for issues caused by tracepoints in libcommon.
Code crashes at runtime if the same tracepoin...
Adam Crume
05:57 PM Revision d4186fee (ceph): lttng: Support --with-lttng=check
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision edddb3fa (ceph): lttng: Fix "make distcheck"
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision fed22f49 (ceph): lttng: Add distro packaging
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e22f59f7 (ceph): rbd-replay: Support replaying partial traces
Tracing may start after the application is started, and image open calls
may missed. To support replaying these trace...
Adam Crume
05:57 PM Revision 684643a3 (ceph): lttng: Fix ceph_ctf_stringp
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision c40ac911 (ceph): rbd-replay: Use standard image@snap instead of image/snap
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 612e85b0 (ceph): rbd-replay: Add rbd_replay::rbd_loc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 6f79f132 (ceph): rbd-replay: Switch ImageNameMap to using rbd_loc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision c78f994e (ceph): rbd-replay: Switch ImageNameMap from regex to plain string matching
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision f64776ac (ceph): rbd-replay: Fix compiler warning in unit tests
Was getting:
test/test_rbd_replay.cc:44:3: warning: converting ‘false’ to pointer type for argument 1 of ‘char testin...
Adam Crume
05:57 PM Revision 8b9f8878 (ceph): rbd-replay: Remove boost_regex from LIBADD and LDADD
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 63273a28 (ceph): lttng: Replace Boost dependencies with custom string code
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision f5272d57 (ceph): rbd-replay: Remove outdated message that says regexes are used
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 4109879f (ceph): lttng: Default to not storing buffer content
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 5365a3e6 (ceph): rbd-replay: Add assertions
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 2a66e09a (ceph): rbd-replay: Add --pool option
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 2eb6ea33 (ceph): rbd-replay: Add --map-image option for image name mapping
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision c888a898 (ceph): lttng: Fix 'make check' by setting up fake HOME
If the HOME variable is not set, LTTng complains, and the extra output
fails to match what the tests expect, so they ...
Adam Crume
05:57 PM Revision 2a11a5cc (ceph): lttng: Disable LTTng by default, add --with-lttng configure option
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision c4939f9a (ceph): lttng: Preload liblttng-ust-fork.so in TESTS_ENVIRONMENT
This adds LD_PRELOAD=liblttng-ust-fork.so to TESTS_ENVIRONMENT.
This prevents lttng from complaining when processes a...
Adam Crume
05:57 PM Revision 8a67647a (ceph): rbd-replay: Add unit tests
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 0d822225 (ceph): rbd-replay: Use non-zero fake data
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision c651bf20 (ceph): rbd-replay: Support writing Actions to ostreams
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision dd8ad8e8 (ceph): rbd-replay: Check that return values are non-negative
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 2db768f0 (ceph): rbd-replay: Fix memory leak in PendingIO
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision d4697bca (ceph): doc: Add usage comment to librbd::RBD::AioCompletion
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision ffc9d05b (ceph): rbd-replay: Add --latency-multiplier option to usage statement
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 25d3d422 (ceph): rbd-replay: Hash completions across multiple mutexes to reduce contention
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e7bf657e (ceph): rbd-replay: Clean up prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 4c360ce9 (ceph): rbd-replay: Fix examples/rbd-replay/replay
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 3dc8443c (ceph): rbd-replay: Add --read-only flag
Removes ignoreWrites flag in prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
05:57 PM Revision 1540f2fd (ceph): rbd-replay: Fix async writes in prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 0421b692 (ceph): rbd-replay: Add command-line arguments to prep-for-replay.py
Includes:
--print-on-read
--print-on-write
--window
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
05:57 PM Revision e9b39d91 (ceph): tracing: instrument filestore
This adds objectstore tracepoints for the filestore. It'd be nice to add
these to the objectstore interface some how ...
Noah Watkins
05:57 PM Revision f6257758 (ceph): lttng: Add more tracing to librbd
Includes:
- aio_complete
- aio_discard
- aio_flush
- aio_get_return_value
- aio_is_complete
- aio_read
- aio_wait_for...
Adam Crume
05:57 PM Revision 4c4289c2 (ceph): lttng: Add rbd-replay
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 16b75623 (ceph): rbd-replay: Switch logging from cout to dout
To enable logs, we also have to use global_init to parse our
command-line args, so we now have other standard Ceph go...
Adam Crume
05:57 PM Revision 4eac5187 (ceph): Add assertf macro
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 9452423f (ceph): rbd-replay: Add rbd_replay subsystem
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision b99f8199 (ceph): rbd-replay: Bugfix: ensure PendingIOs are added before they're removed
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 80110160 (ceph): rbd-replay: Printing pending IOs when worker threads waiting to die
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision f2228c0c (ceph): rbd-replay: Change Worker::m_pending_ios from vector to map
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision eec93823 (ceph): rbd-replay: Print "Still waiting for" more judiciously
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e312be61 (ceph): lttng: Trace ReplicatedPG::do_osd_ops
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision fcf49257 (ceph): lttng: Add tracing-common.h
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 10a476ad (ceph): examples: Add example for the librados C API
Simply copied and modified the C++ API example.
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
05:57 PM Revision 16cd0744 (ceph): lttng: Trace librados C API
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 012ff295 (ceph): lttng: Prevent creation of incorrect .o files when generating from .tp
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 00b2ec6f (ceph): lttng: Prevent deletion of tracing-common.h when cleaning
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision e1e157fb (ceph): lttng: Split up libtracepoints
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 772148e2 (ceph): lttng: Remove 'ver' from trace in code for CEPH_OSD_OP_NOTIFY
'ver' is obsolete and variable exists only for proper deserialization
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
05:57 PM Revision d0834bb8 (ceph): lttng: Basic tracing of librbd
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 115cfb3a (ceph): tracing: bootstrap lttng-ust with mutex events
See src/tracing/README.md
Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
Noah Watkins
05:57 PM Revision 7fa513ea (ceph): lttng: trace mutex::unlock
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:57 PM Revision 3ac99e3f (ceph): lttng: add pg and osd tracepoints
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:57 PM Revision 915a6a64 (ceph): lttng: Check for lttng/tracepoint.h when configuring
Adam Crume
05:57 PM Revision 33b87f92 (ceph): tracing: automake-ify tracepoint generation
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
05:57 PM Revision ae599464 (ceph): lttng: Trace OpRequest
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:57 PM Revision 1802bc25 (ceph): lttng: Add rmw_flags to tracepoint in PG::queue_op
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
05:28 PM Bug #9153 (Resolved): erasure-code: jerasure_matrix_dotprod segmentation fault due to package upg...
Loïc Dachary
05:14 PM Revision ca3ac907 (ceph): Merge pull request #2298 from dachary/wip-9153-jerasure-upgrade-firefly
erasure-code: preload the jerasure plugin variant (sse4,sse3,generic)
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:55 PM Feature #8147 (Resolved): osd: make split automatically trigger scrub
Sage Weil
04:49 PM Bug #8998 (Resolved): osd: SEGV in OSD::heartbeat()
no backport needed; this happened bc update_osd_stats() was in OSDService but still using hte other dout macro, but f... Sage Weil
04:49 PM rgw Feature #9200 (Resolved): rgw: log civetweb access
Apache has an access log, civetweb has one too, however we need to incorporate it into our logging system. Yehuda Sadeh
04:44 PM CephFS Bug #5762 (Resolved): teuthology: Failed MPI runs lead to a hung test instead of a failure
Sage Weil
03:38 PM Revision 50fd1fcf (ceph): use 'mon create-initial' always
But don't error if it fails, as this would mean that the monitors
are just taking longer to form quorum. Go and try t...
Alfredo Deza
03:38 PM Revision 182f22fe (ceph): ignore errors on informational service status
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 5b946e1a6df8b7f80001f5176435eb36b12...
Alfredo Deza
03:29 PM Feature #8639: mon: dispatch messages while blocked waiting for IO
Sage Weil
03:29 PM Feature #7516 (Resolved): mon: reweight-by-pg
Sage Weil
03:27 PM Fix #9199 (Resolved): librados: watch linger pings need to verify pg mapping hasn't changed
at the same time, osds might want to push osdmap incrementals to client sessions with watchers to expedite things ... Sage Weil
03:22 PM Feature #9198 (Resolved): librados: notify callback includes gid of notifier
Sage Weil
03:21 PM Feature #9197 (Resolved): librados/osd: notify reply payload
Sage Weil
03:21 PM Revision e870fd09 (ceph): os/FileJournal: For journal-aio-mode, don't use aio when closing journal.
For jouranl-aio-mode when closing journal, the write_finish_thread_entry may exit before
write_thread_entry. This cau...
Ma Jianpeng
03:21 PM Fix #9196 (Resolved): librados: watch_check() to synchronous verify we haven't missed notifies
Sage Weil
03:21 PM Fix #9195 (Resolved): librados: issue watch callback on (possibly) missed notifies
Sage Weil
03:20 PM Fix #9194 (Resolved): librados/osd: watch reconnect needs to be exclusive to detect possibly miss...
Sage Weil
03:18 PM Linux kernel client Bug #8806: libceph: must use new tid when watch is resent
the watch resend needs to use a new tid to avoid the dup op detection in the osd. this is how librbd avoids this pro... Sage Weil
03:16 PM Revision 5765fde1 (ceph): Merge pull request #102 from ceph/9171
ignore errors on informational service status Zack Cerza
03:15 PM Revision 54870dcc (ceph): Merge pull request #103 from ceph/9172
use 'mon create-initial' always Zack Cerza
02:55 PM Bug #9176 (Pending Backport): mon: leaked MMonGetVersion
Sage Weil
01:08 PM Bug #9176 (Fix Under Review): mon: leaked MMonGetVersion
https://github.com/ceph/ceph/pull/2301 Sage Weil
02:49 PM rgw Bug #9160: rgw failures with 'NoneType' object has no attribute 'get_contents_as_string'
http://pulpito.front.sepia.ceph.com/sage-2014-08-19_15:19:41-rgw-master-testing-basic-multi/435812/
http://pulpito.f...
John Spray
02:43 PM rgw Bug #9160: rgw failures with 'NoneType' object has no attribute 'get_contents_as_string'
http://pulpito.front.sepia.ceph.com/john-2014-08-20_19:21:46-rgw-wip-objecter-testing-basic-plana/438545/ John Spray
02:03 PM Revision 70a1f18a (ceph): use 'mon create-initial' always
But don't error if it fails, as this would mean that the monitors
are just taking longer to form quorum. Go and try t...
Alfredo Deza
01:56 PM Bug #9144 (Pending Backport): filestore: commit triggered during journal replay
Sage Weil
01:44 PM Revision 5b946e1a (ceph): ignore errors on informational service status
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
01:21 PM Bug #9193: notify does not return an error code on timeout
https://github.com/ceph/ceph/pull/2302 Sage Weil
01:20 PM Bug #9193 (Resolved): notify does not return an error code on timeout
commit:7c7bf5fee7be397ef141b947f532a2a0b3567b42
There is simply no error code passed back to the caller; the API c...
Sage Weil
01:10 PM Bug #9150: osd/ECBackend.cc: 529: FAILED assert(pop.data.length() == sinfo.aligned_logical_offset...
suspect this and #9135 to be a ghost due to misbehaving underlying fs Sage Weil
01:09 PM Bug #9145 (Resolved): recursive lock of CollectionIndex::access_lock (52)
Sage Weil
01:07 PM Revision a66a4931 (ceph): os/FileJournal: Only using aio then alloc the related resources.
If define HAVE_LIBAIO, it alloc related resouces. But itt don't check whether
using aio mode. Only using aio it alloc...
Ma Jianpeng
12:51 PM Bug #9182 (Need More Info): osd deadlock after ms_handle_reset
Sage Weil
12:50 PM Bug #9181 (Need More Info): Osd: segv in OpTracker::unregister_inflight_op
no log, core isn't giving me good info :( Sage Weil
12:41 PM Revision 8d7e77b9 (ceph): erasure-code: preload the jerasure plugin variant (sse4,sse3,generic)
The preloading of the jerasure plugin ldopen the plugin that is in
charge of selecting the variant optimized for the
...
Loic Dachary
12:34 PM Bug #8885 (Can't reproduce): SIGABRT in TrackedOp::dump() via dump_ops_in_flight()
Sage Weil
12:09 PM devops Feature #9136 (Fix Under Review): ceph-deploy: use pre-existing ceph.conf
Pull request opened https://github.com/ceph/ceph-deploy/pull/234 Alfredo Deza
12:07 PM devops Bug #9185: incorrect Centos 6.5 fastcgi package
ok, the idle timeout is working fine.. i can pause the radosgw process (kill -STOP) and curl will block for well over... Sage Weil
10:27 AM devops Bug #9185 (In Progress): incorrect Centos 6.5 fastcgi package
Sage Weil
09:52 AM devops Bug #9185: incorrect Centos 6.5 fastcgi package
(09:51:57 AM) sagehm@newdream.net/montreal: mod_fastcgi-2.4.7-1.ceph.el6.x86_64
(09:52:15 AM) sagehm@newdream.net/mo...
Sage Weil
11:43 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Does fio complete eventually? Are there any other hung tasks in dmesg? A task blocking for more than 120 seconds is... Ilya Dryomov
11:38 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
I apply http://gitbuilder.ceph.com/kernel-deb-precise-x86_64-basic/ref/wip-request-fn/linux-image-3.16.0-ceph-00037-g... German Anders
11:37 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Ok, I've applied the "..." with Kernel 3.16.0 and the error continues:
...
Aug 21 14:38:45 mail02-old kernel: [ 7...
German Anders
10:19 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Eric is correct, the fix isn't in 3.16 stable yet, and unfortunately won't be in 3.15 at all - Linus pulled it into h... Ilya Dryomov
10:10 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
The fix looks like it made it into 3.17rc1. I have been testing this kernel since Sunday, and have not triggered the ... Eric Eastman
09:31 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Upgrade to kernel: 3.16.0 and got the same problem:
...
[ 70.858716] Key type ceph registered
[ 70.858800] l...
German Anders
11:18 AM Linux kernel client Bug #9192 (New): krbd: poor read (about 10%) vs write performance
We started testing the 3.17rc1 kernel over the weekend, as it is the only Linus
released kernel that has the fix fo...
Eric Eastman
10:05 AM devops Feature #5773 (In Progress): ceph-deploy: should add more tests to ceph-deploy task
Tamilarasi muthamizhan
09:55 AM CephFS Bug #9152 (In Progress): mds: beacon needs to not take mds_lock
wip-9152 John Spray
09:50 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
The compiler is spitting out a warning about getcwd -- no evidence that that's what it's actually hitting in this ins... John Spray
08:53 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-20_23:04:01-fs-next-testing-basic-multi/439228/ Greg Farnum
08:29 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
How did you track it down to getcwd? If that is the issue there are a bunch of avenues of attack here, and we should ... Greg Farnum
06:31 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
mdtest has a getcwd call into an unzeroed buffer that it doesn't check the error of. If fuse is failing the getcwd f... John Spray
09:46 AM devops Bug #9190 (Resolved): idle times out do not work on ubuntu precise
This maybe similar to #9185
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-21_08:29:18-upgrade...
Yuri Weinstein
08:26 AM Bug #9188: make check fails for setmaxosd
"make check" is passing on our gitbuilders (http://ceph.com/gitbuilder.cgi). Try updating and running it again? If th... Greg Farnum
02:28 AM Bug #9188 (Rejected): make check fails for setmaxosd
make check fails for setmaxosd. This is after a recent change in setmaxosd behavior to disallow shrinking of OSDs. He... Anand Bhat
07:49 AM Revision c8e2b89c (ceph): os/FileJournal: Tune the judge logic for read_header.
When reading journal-header, it should firstly check the result of
pread and then do decoce operation.
Signed-off-by...
Ma Jianpeng
06:56 AM CephFS Bug #9151 (In Progress): mds should log/error/warn when segments are NOT getting trimmed
John Spray
05:56 AM CephFS Feature #9189 (Resolved): Expose client identifying metadata to MDS, e.g. hostname

Currently, when doing e.g. a "session ls" on an MDS's admin socket, we get client IDs and IP addresses. It would b...
John Spray
05:35 AM CephFS Bug #9173 (Fix Under Review): Crash in Server::_session_logged

https://github.com/ceph/ceph/pull/2297
John Spray
05:27 AM Revision 389ad614 (ceph): added a test with erasure_code_profile parameters
Signed-off-by: tamil <tamil.muthamizhan@inktank.com> Tamilarasi muthamizhan
03:50 AM Revision fdbab468 (ceph): Fix set_alloc_hint op cause KeyValueStore crash problem
Now KeyValueStore doesn't support set_alloc_hit op, the implementation of
_do_transaction need to consider decoding t...
Haomai Wang
03:28 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
Missed a step to mention.
before i did a repair on the primary osd; i aslo did a scrub
#:/build/ceph-firefly84/sr...
Dhiraj Kamble
03:17 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
Hi Loic,
please find below the steps to reproduce the issue.
@*#:/build/ceph-firefly84/src# ./ceph -v
*** DEVE...
Dhiraj Kamble
01:09 AM rgw Bug #9155: Swift Subuser - 403 Forbidden - during upload/post
made a comment on your proposed fix. Dhiraj Kamble

08/20/2014

10:09 PM Revision b0aa846b (ceph): Merge pull request #2295 from dachary/wip-9153-jerasure-upgrade
erasure-code: do not preload the isa plugin
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:10 PM Revision 3ed8c682 (ceph): erasure-code: do not preload the isa plugin
Because it's not built for all architectures and distributions.
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
09:02 PM devops Bug #9187 (Resolved): osds down after fresh deploy in master branch of ceph
Sage Weil
09:02 PM devops Bug #9187: osds down after fresh deploy in master branch of ceph
thsi si fixed later today. it was the isa preload thing:
2014-08-20 21:04:58.845739 7f7369af2780 -1 load: jerasur...
Sage Weil
04:37 PM devops Bug #9187 (Resolved): osds down after fresh deploy in master branch of ceph
ceph version 0.84-367-gf71c889
test setup: mira023
ceph-deploy version: 1.5.11
created 4 osds, with a combi...
Tamilarasi muthamizhan
08:48 PM Bug #9180 (Resolved): keyvaluestore: bad op 2563
done, commit:fdbab46852e74d405b5c747da98564a5866ec8a7 . thanks!! Sage Weil
08:07 PM Bug #9180: keyvaluestore: bad op 2563
We need to backport commit c08adbc98ff5f380ecd215f8bd9cf3cab214913c(https://github.com/ceph/ceph/commit/c08adbc98ff5f... Haomai Wang
10:39 AM Bug #9180 (Resolved): keyvaluestore: bad op 2563
... Sage Weil
08:11 PM Revision 4c2ae691 (ceph): add pom.xml so to deploy the libcephfs to maven repository.
to build a jar, version=0.80-rc1-2008-gf71c889 mvn package -Dversioin=;
and mvn deploy command will deploy the jar to...
Huamin Chen
08:08 PM Revision 54aa5d92 (ceph): Merge pull request #2219 from somnathr/wip-optracker-lock
TrackedOp: Removed redundant lock in OpTracker::_mark_event()
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
06:41 PM Revision f680a24b (ceph): TrackedOp: Removed redundant lock in OpTracker::_mark_event()
ops_in_flight_lock seems redundant in OpTracker::_mark_event()
and this lock is highly contended for. Removing the sa...
Pavan Rallabhandi
06:41 PM Revision e45f5c2c (ceph): TrackedOp:_dump_op_descriptor is renamed to _dump_op_descriptor_unlocked
Caller don't need to hold lock before calling _dump_op_descriptor(),so,
to reflect this it is renamed to _dump_op_des...
Pavan Rallabhandi
06:23 PM Revision 5026a898 (ceph): Rebased civetweb version to master's
Luis Pabon
06:17 PM Revision e665e627 (ceph): Rebased and changed debug option
Signed-off-by: Luis Pabón <lpabon@redhat.com> Luis Pabon
06:15 PM Revision 4fa4553d (ceph): Merge branch 'master' into rgw-doc
Conflicts:
src/vstart.sh
Luis Pabon
05:33 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Plugging one of the 520s into a 3Gbit sata port makes no difference either. Mark Kirkwood
04:58 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Updated the bios on the work machine. No difference. Mark Kirkwood
05:27 PM Revision bfe31bec (ceph): Add centos/rhel7 to distros 'all' folder.
Signed-off-by: Sandon Van Ness <sandon@inktank.com> Sandon Van Ness
05:10 PM Revision 922e24e6 (ceph): Merge pull request #2286 from dachary/wip-9153-jerasure-upgrade-firefly
erasure-code: preload the jerasure plugin (firefly)
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
05:08 PM Revision f71c8898 (ceph): Merge pull request #2282 from dachary/wip-9153-jerasure-upgrade
erasure-code: preload the jerasure plugin
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:21 PM Revision 98b24f39 (ceph): vstart.sh to enable support for simple RadosGW
Signed-off-by: Luis Pabón <lpabon@redhat.com> Luis Pabon
04:08 PM Bug #9153 (In Progress): erasure-code: jerasure_matrix_dotprod segmentation fault due to package ...
preloading jerasure is not enough : the plugin selects another plugin to be loaded depending on the CPU features (jer... Loïc Dachary
03:29 PM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
I still see this error in today's run http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-20_13:52:13-upgrade:dump... Yuri Weinstein
10:07 AM Bug #9153 (Resolved): erasure-code: jerasure_matrix_dotprod segmentation fault due to package upg...
Sage Weil
03:31 PM Revision 5bf472ae (ceph): os/FileJournal: Update the journal header when closing journal
When closing journal, it should check must_write_header and update
journal header if must_write_header alreay set.
It...
Ma Jianpeng
03:31 PM Revision 57778e2c (ceph): os/FileJournal: signal aio_cond even if seq is 0
This can happen if we write a journal but no events.
Reported-by: Somnath Roy <somnath.roy@sandisk.com>
Reported-by:...
Sage Weil
03:27 PM devops Bug #9185: incorrect Centos 6.5 fastcgi package
fcgi? how does that even enter into it? I thought our work was only with fastcgi?
Is this on teuthology, or cust...
Dan Mick
03:26 PM devops Bug #9185: incorrect Centos 6.5 fastcgi package
So this problem is with the fcgi package not mod_fastcgi? Sandon Van Ness
02:07 PM devops Bug #9185: incorrect Centos 6.5 fastcgi package
This should fix #9169 Yuri Weinstein
01:54 PM devops Bug #9185 (Rejected): incorrect Centos 6.5 fastcgi package
The fastcgi package that is being installed is, or either based off: fcgi-2.4.0-10.el6.x86_64. Not 100% sure that it ... Yehuda Sadeh
02:33 PM Feature #9031: List RADOS namespaces and list all objects in all namespaces
David Zafman
02:31 PM Bug #9186 (Duplicate): erasure-code: conditionally preload isa plugin
The isa plugin is only built on some platforms. When the OSD preloads plugins, it should not try to load plugins that... Loïc Dachary
02:05 PM rgw Bug #9169: 100-continue broken for centos/rhel
This seems to be due to idle timeout is not working, should be fixed by #9185 Yuri Weinstein
01:27 PM devops Feature #9136 (In Progress): ceph-deploy: use pre-existing ceph.conf
Alfredo Deza
10:54 AM Bug #9182: osd deadlock after ms_handle_reset
..and when i detached gdb the osd saw it was marked down, and came back to life after that. :/ Sage Weil
10:52 AM Bug #9182: osd deadlock after ms_handle_reset
... Sage Weil
10:51 AM Bug #9182 (Can't reproduce): osd deadlock after ms_handle_reset
ubuntu@teuthology:/a/teuthology-2014-08-19_02:30:02-rados-firefly-distro-basic-multi/435572... Sage Weil
10:47 AM CephFS Bug #9173: Crash in Server::_session_logged
Better log. John Spray
06:30 AM CephFS Bug #9173 (Resolved): Crash in Server::_session_logged

Hit by mds_client_recovery task...
John Spray
10:43 AM Bug #9181 (Resolved): Osd: segv in OpTracker::unregister_inflight_op
... Sage Weil
10:38 AM Bug #9179 (Resolved): unfound objects, recovery timeout
402/7722 unfound (
all osds up
ubuntu@teuthology:/a/teuthology-2014-08-19_02:30:02-rados-firefly-distro-basic-m...
Sage Weil
10:33 AM CephFS Bug #9178: samba: ENOTEMPTY on "rm -rf"
http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-10_23:14:02-samba-next-testing-basic-plana/415869/
Greg Farnum
10:30 AM CephFS Bug #9178 (Resolved): samba: ENOTEMPTY on "rm -rf"
... Greg Farnum
10:14 AM CephFS Bug #9177 (Resolved): ceph-fuse: failing MPI mdtest runs
... Greg Farnum
09:40 AM Bug #9176 (Resolved): mon: leaked MMonGetVersion
ubuntu@teuthology:/a/teuthology-2014-08-19_02:30:02-rados-firefly-distro-basic-multi/435589 Sage Weil
09:38 AM Bug #9175 (Duplicate): osd: stuck recovery
ubuntu@teuthology:/a/teuthology-2014-08-19_02:30:02-rados-firefly-distro-basic-multi/435529
pgs stuck recovery, ne...
Sage Weil
09:33 AM Feature #7238: erasure code : implement LRC plugin
Reserved three machines and run the following job on them:... Loïc Dachary
09:32 AM rgw Subtask #9068 (In Progress): rgw: add rgw setup to vstart
Pull request: https://github.com/ceph/ceph/pull/2292 Luis Pabon
09:31 AM rgw Documentation #9003: rgw: document development setup for rgw
Abhishek L wrote:
> Luis Pabon wrote:
> > I have edited vstart.sh so that it can setup rgw automatically. I have a...
Luis Pabon
09:30 AM rgw Documentation #9003: rgw: document development setup for rgw
patch has been submitted: https://github.com/ceph/ceph/pull/2292 Luis Pabon
05:21 AM rgw Documentation #9003: rgw: document development setup for rgw
Luis Pabon wrote:
> I have edited vstart.sh so that it can setup rgw automatically. I have also documented most of ...
Abhishek Lekshmanan
09:19 AM Bug #9128: Newly-restarted OSD may suicide itself after hitting suicide time out value because it...
sounds like we need to use the TPHandle and tp.reset_tp_handle() inside the search_For_missing loop Sage Weil
07:53 AM Documentation #9174: wrong picture on http://ceph.com/docs/master/cephfs/
... Dieter Kasper
07:46 AM Documentation #9174 (Closed): wrong picture on http://ceph.com/docs/master/cephfs/
The picture on page http://ceph.com/docs/master/cephfs/
is not correct.
ceph.ko is not on top of libcephfs / librad...
Dieter Kasper
04:24 AM Revision 790de974 (ceph): doc/start/quick-ceph-deploy: missing {ceph-node} from mon create-initial
Signed-off-by: Dan Mick <dan.mick@inktank.com> Dan Mick
03:56 AM Revision b3624500 (ceph): Merge pull request #2283 from somnathr/wip-sd-9145
CollectionIndex: Collection name is added to the access_lock name
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
03:11 AM devops Feature #8868 (Resolved): Update Fedora to 0.80.5 packages with ceph-common
The updated packages with spec file synced up with the upstream spec file were pushed to epel 7, fedora 22, fedora 21... Boris Ranto
01:50 AM Revision 615d2d90 (ceph): CollectionIndex: Collection name is added to the access_lock name
The CollectionIndex constructor is changed to accept the coll_t
so that the collection name can be used to form acces...
Somnath Roy
12:37 AM Revision 164f1a19 (ceph): erasure-code: preload the jerasure plugin
Load the jerasure plugin when ceph-osd starts to avoid the following
scenario:
* ceph-osd-v1 is running but did not ...
Loic Dachary
12:31 AM Revision 9b802701 (ceph): erasure-code: preload the jerasure plugin
Load the jerasure plugin when ceph-osd starts to avoid the following
scenario:
* ceph-osd-v1 is running but did not ...
Loic Dachary

08/19/2014

10:45 PM Revision bb77e3af (ceph): Merge pull request #2043 from guangyy/wip-pg-splitting
Support 'expected_num_objects' parameter when creating pool for pg folder splitting
Reviewed-by: Samuel Just <sam.ju...
Samuel Just
09:33 PM Revision fc412734 (ceph): mon: fix signed/unsigned warnings
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
09:31 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
A related thought is that the Intel 520s are plugged into the sata 6Gbit ports on the motherboard, so if there are an... Mark Kirkwood
06:52 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
It might be worth trying an Intel 530 if that is dramatically easier to source - as it is similar to the 520 in the m... Mark Kirkwood
06:26 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
That should have said unpatched wip-9073. Mark Kirkwood
06:25 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Doing a little more digging for the cause of 2/ (invalid argument erro). Using unpatched ipw-0973 and changing the jo... Mark Kirkwood
09:07 PM rgw Bug #9125 (Resolved): rgw: swift tests fail with civetweb
Sage Weil
08:15 PM Revision 1441ffe8 (ceph): rgw: subuser creation fixes
Fixes: #8587
There were a couple of issues, one when trying to identify whether swift
user exists, we weren't using t...
Yehuda Sadeh
08:06 PM Revision 23a9b763 (ceph): Merge pull request #2287 from ceph/wip-reweight-tunables
mon: make reweight-by-* sanity limits configurable
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
06:47 PM Revision 6271ca91 (ceph): Merge pull request #99 from ceph/wip-hadoop
reorg hadoop tests
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
06:47 PM Revision d9cf2991 (ceph): Merge pull request #2279 from ceph/wip-hadoop
fix and reorg hadoop workunits
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
06:32 PM Revision 82409ee6 (ceph): mon: make reweight-by-* sanity limits configurable
Also drop the somewhat redundant osd_sum.kb check; the main thing we care
about here is
Signed-off-by: Sage Weil <sa...
Sage Weil
05:44 PM Feature #7238: erasure code : implement LRC plugin
There is no need to test upgrade on a plugin that does not exist in LRC. Loïc Dachary
02:34 PM Feature #7238: erasure code : implement LRC plugin
canceled the previous job because it did not have enough OSD to complete (the LRC rule requires a minimum of 8 for ea... Loïc Dachary
12:22 PM Feature #7238: erasure code : implement LRC plugin
Cancel the "teuthology run that did not contain any LRC workload":http://pulpito.ceph.com/loic-2014-08-19_20:27:09-up... Loïc Dachary
11:27 AM Feature #7238: erasure code : implement LRC plugin
Fixed a few problems and running "a firefly upgrade suite":http://pulpito.ceph.com/loic-2014-08-19_20:27:09-upgrade:f... Loïc Dachary
05:40 PM Revision c36b72c1 (ceph): Merge pull request #2199 from ceph/wip-reweight
mon: allow reweighting of osds by pg (isntead of bytes used)
Reviewed-by: Guang Yang <yguang@yahoo-inc.com>
Sage Weil
03:16 PM Revision 33048410 (ceph): mon/OSDMonitor: respect CRUSH weights for reweight-by-pg
Do not assume that all OSDs are weighted equally for reweight-by-pg.
Note that reweight-by-utilization already rewei...
Sage Weil
03:16 PM Revision 1ecf44eb (ceph): mon/OSDMonitor: reweight-by-pg for pool(s)
Allow the reweight-by-pg to look at a specific set of pools. If the list
is ommitted, use PGs from all pools. This ...
Sage Weil
03:16 PM Revision 977f8527 (ceph): qa/workunits/cephtool/test.sh: test reweight-by-pg
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
03:16 PM Revision 8b971e94 (ceph): mon/OSDMonitor: adjust weights up, when possible
Note when OSDs are underloaded, as well. If that is the case, adjust the
OSD reweight value if, if possible. (It wo...
Sage Weil
03:16 PM Revision 01cb4054 (ceph): mon/OSDMonitor: reweight-by-pg
This is just like reweight-by-utilization, but looks purely at the PG to
OSD mapping, not at the number of bytes used...
Sage Weil
03:08 PM Bug #9156: SWIFT tests failed in upgrade:dumpling:rgw-dumpling-distro-basic-vps suite
Further analyzes and chants with Loic and Yehuda revealed that in apache access log we indeed have 30 sec not 1200 se... Yuri Weinstein
03:02 PM Bug #9156: SWIFT tests failed in upgrade:dumpling:rgw-dumpling-distro-basic-vps suite
Suspected backport apache 2.4 issue, test branch wip-rgw-dumpling for ceph-qa-suite
Running now ...
Yuri Weinstein
02:15 PM Fix #8914 (Need More Info): osd crashed at assert ReplicatedBackend::build_push_op
I'm not able to reproduce the problem on *ceph version 0.84-343-g92b227e (92b227e1c0b1533c359e74c81de58140b483ee8e)* ... Loïc Dachary
01:15 PM rgw Bug #9155: Swift Subuser - 403 Forbidden - during upload/post
I pushed a different fix to wip-8587, please take a look and see if you think it makes sense. Yehuda Sadeh
01:10 PM Feature #8155: Disallow changing cache_mode in nonsensical ways
c3f403293c7f8d946f66a871aa015a558120ce78 Samuel Just
01:10 PM Feature #8155 (Resolved): Disallow changing cache_mode in nonsensical ways
Samuel Just
01:09 PM devops Feature #9050: Calamari builds for ceph.com
Asking Ian and Neil, they confirm that what this means is "repos". The hard choice is going to be figuring out what ... Dan Mick
12:15 PM Bug #9170 (Resolved): erasure-code: preload erasure code plugins
Whitelist the plugins to be preloaded. Loïc Dachary
11:19 AM devops Feature #3019 (Closed): juju: modernize ceph charm, mon & osd bootstrap
Neil Levine
11:11 AM rgw Bug #9169 (Resolved): 100-continue broken for centos/rhel
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-18_16:07:27-upgrade:dumpling-firefly-x-firefly-dis... Yuri Weinstein
11:10 AM devops Feature #8868 (In Progress): Update Fedora to 0.80.5 packages with ceph-common
Ian Colle
11:04 AM Revision e6658e2a (ceph): mailmap: Anand Bhat affiliation
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
11:03 AM Revision 4535cc98 (ceph): mailmap: Federico Simoncelli affiliation
Reviewed-by: Federico Simoncelli <fsimonce@redhat.com>
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
09:20 AM rgw Feature #8911: RGW doesn't return 'x-timestamp' in header which is used by 'View Details' of Open...
I'll take a look. Seems like this is new functionality in RGW, not a bug, right? Luis Pabon
09:13 AM CephFS Bug #9152: mds: beacon needs to not take mds_lock
Hmm, the beacon send code doesn't need to hold the lock on its own, but it's triggered by the SafeTimer, which is jus... Greg Farnum
09:07 AM rgw Documentation #9003: rgw: document development setup for rgw
I have edited vstart.sh so that it can setup rgw automatically. I have also documented most of the steps needed by n... Luis Pabon
09:02 AM rgw Documentation #9003 (In Progress): rgw: document development setup for rgw
Luis Pabon
09:05 AM CephFS Bug #9151: mds should log/error/warn when segments are NOT getting trimmed
What kind of logging do we want? I assume you mean journal segments, and this is a bog standard operation...
If it's...
Greg Farnum
09:04 AM rgw Feature #8945: rgw: support swift /info api
After spending some time on this call, I am going to have to break it down to smaller tasks. I am currently investig... Luis Pabon
09:02 AM Bug #9143: Incorrect key sequence in encoding object name to key for GenericObjectMap
How did you run across this? Is it feasible to fix it by typing the escaped strings and writing a custom comparator? Greg Farnum
07:47 AM Bug #9079: osd: bad learned_addr during send_boot
"pending pull request":https://github.com/ceph/ceph/pull/2275 Loïc Dachary
07:41 AM Feature #9167 (Resolved): erasure-code: check plugin version when loading it
When loading the erasure code plugin, check the Ceph version against which it was built and fail if it does not match... Loïc Dachary
07:22 AM devops Bug #9166 (Closed): activate dmcrypt volumes via init script
Hi,
I don't know if this is more a bug or a feature request.
I think it would helpful if the activation of ceph ...
Manuel Lausch
07:16 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
"firefly backport":https://github.com/ceph/ceph/pull/2286 Loïc Dachary
07:10 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
The teuthology upgrade tests fails consistently with the same problem. Backporting to firefly seem to be the only way... Loïc Dachary
05:21 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
"Running upgrade:dumpling-firefly-x with the proposed fix":http://pulpito.ceph.com/loic-2014-08-19_14:23:09-upgrade:d... Loïc Dachary
07:10 AM Revision ad6a2be4 (ceph): Implement the collection hint transaction, add a new transation type as...
Signed-off-by: Guang Yang (yguang@yahoo-inc.com) Guang Yang
07:10 AM Revision dbf624e1 (ceph): Add tests for the collection hint OP: 1) Store Test 2) Idempotent Test.
Signed-off-by: Guang Yang (yguang@yahoo-inc.com) Guang Yang
07:08 AM Revision da37273d (ceph): Add a new field 'expected_num_objects' to pg_pool_t which denotes the e...
Signed-off-by: Guang Yang (yguang@yahoo-inc.com) Guang Yang
07:08 AM Revision 35f323d9 (ceph): Add a new monitor command to let user specify the expected number of ob...
Signed-off-by: Guang Yang (yguang@yahoo-inc.com) Guang Yang
07:08 AM Revision 7d266d13 (ceph): Add a new transaction OP (collection hint) to ObjectStore.
Signed-off-by: Guang Yang (yguang@yahoo-inc.com) Guang Yang
06:49 AM CephFS Fix #4286: SLES 11 - cfuse: disable 'big_writes'and 'atomic_o_trunc
Ian Colle
04:17 AM rbd Bug #9076: Can't completely remove a version 1 image on RHEL 7
Ok it's better with ceph.com packages. You can close this :)
Thanks!
Sébastien Han
04:16 AM rbd Bug #9075: Can't create a version 2 images on RHEL 7
Ok it's better with ceph.com packages. You can close this :)
Thanks!
Sébastien Han
04:10 AM Revision 92b227e1 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil

08/18/2014

11:21 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
FWIW - checked this myself on my home machine (which was *not* seeing this last issue recall, only the hang) by reboo... Mark Kirkwood
07:48 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
My linux version is 3.2 && 3.5. I'll test on 3.13.0-32-generic to find whether kernel cause this bug. jianpeng ma
07:00 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Excellent. Purely out of interest, any idea (now) why we only saw this bug on one particular system? Mark Kirkwood
04:04 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Hmm, thanks very much! I'll send the patch.
Thanks again, Mark!
jianpeng ma
03:44 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Retested with only debug-journal-header-3.diff on wip-9073. I did 200 test runs, good journal every time. Mark Kirkwood
02:39 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I think you should retest only using debug-journal-header-3.diff on wip-9073. And test more times to avoid the bug r... jianpeng ma
02:36 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I had your last debugging diff on there as well (I can retest without that if needed). Mark Kirkwood
02:34 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Are you only apply debug-journal-header-3.diff on wip-9073 to test ?
jianpeng ma
02:32 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Heh - sorry, means 'really fixed it well'! Mark Kirkwood
02:30 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
What's mean of nail it? sorry, i don't know. jianpeng ma
02:21 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Lol, you certainly have - been a pleasure debugging this with you!
I actually applied the patch attached in this n...
Mark Kirkwood
02:01 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
growl, make that 3.13.0-32-generic, typed 'uname -a' in wrong (x)window before! Mark Kirkwood
02:01 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I have a thought. It's strange.
Using aio, the kernel use user-space to write. But if before write to journal, the u...
jianpeng ma
01:58 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
...oh and kernel is 3.13.0-34-generic (sorry)! Mark Kirkwood
01:52 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Yeah, disabling dio seems to get a consistently good header (10 consecutive runs) Mark Kirkwood
01:22 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
From the latest ceph-osd.o.log. Before io_submit, the content is ok.
I found another issue.
2014-08-18 20:10:09.7...
jianpeng ma
01:10 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Seems I spoke too soon - a few more runs showed up:
$ hexdump -n8 journalblk-prestart--20864.txt
0000000 7000 033...
Mark Kirkwood
12:38 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
With *only* your latest pacth applied to wip-9073 I'm seeing a good journal header:
$ hexdump -n8 journalblk-prest...
Mark Kirkwood
12:12 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Using my latest patch, is journal-header corrupt?
From my debug info, before io_submit and after aio completed, the ...
jianpeng ma
09:44 PM rgw Bug #9155: Swift Subuser - 403 Forbidden - during upload/post
Here's the pull request:
[[https://github.com/ceph/ceph/pull/2281]]
Dhiraj Kamble
08:20 AM rgw Bug #9155: Swift Subuser - 403 Forbidden - during upload/post
That's duplicate of #8587, a pull request for your fix would be great. Yehuda Sadeh
07:49 AM rgw Bug #9155 (Resolved): Swift Subuser - 403 Forbidden - during upload/post
Swift Upload fails with HTTP error 403 for an subuser that was created with the required permissions. This happens ge... Dhiraj Kamble
09:29 PM Revision ab886c4a (ceph): doc: Removed quick guide and wireshark from top-level IA.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
09:28 PM Revision acee2e58 (ceph): doc: Move wireshark documentation to dev.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:57 PM Revision ce6e9a91 (ceph): doc/release-notes: v0.84
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:26 PM Bug #9062: Mon segfault in waitlist_or_zap_client
the fix was merged in commit:321d4defd4a0f5a53a41276e6dc048479cb3084a Greg Farnum
05:51 PM Revision ae787cfa (ceph): Work around an apparent binding bug (GCC 4.8).
A reference to h->seq passed to std::pair ostensibly could not bind
because the header structure is packed. At first...
Matt Benjamin
05:14 PM Bug #9145: recursive lock of CollectionIndex::access_lock (52)
The fix Sam suggested is to name the CollectionIndex lock based on the collection names. This will make lockdep happy... Somnath Roy
01:58 PM Bug #9145: recursive lock of CollectionIndex::access_lock (52)
Sage,
Yes, I am able to reproduce this following the steps you suggested. But, this time I am hitting the issue in _...
Somnath Roy
05:10 PM Revision 73e75fdb (ceph): use the right syntax for RHEL/CentOS distros to check for ceph status
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 4b15d0118ec000fe1e947131a62ef9990b2...
Alfredo Deza
05:10 PM Revision c1a2463c (ceph): Merge pull request #100 from ceph/wip-9154
use the right syntax for RHEL/CentOS distros to check for ceph status
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
05:05 PM Revision 209226f1 (ceph): Merge pull request #98 from ceph/rgw_civetweb
bug 8927: include frontend:civetweb for rgw tests Sage Weil
05:04 PM Revision a59bc865 (ceph): Merge pull request #2280 from ceph/wip-fs-docs
doc: add notes on using "ceph fs new"
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:51 PM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
"minimal fix":https://github.com/ceph/ceph/pull/2282 Loïc Dachary
09:05 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Stopping the daemons may not be the brightest idea because of http://tracker.ceph.com/issues/8849 . Pre-loading the p... Loïc Dachary
08:09 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
"proposed fix":https://github.com/ceph/ceph/pull/2278 Loïc Dachary
07:27 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Here is a possible scenario:
* ceph-osd-0.80.5 is running but did not load jerasure
* ceph-osd-0.83 is installed ...
Loïc Dachary
07:09 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Here is the part of the teuthology log dealing with the upgrade, which is immediately followed by a core dump from os... Loïc Dachary
06:43 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Trying a manual upgrade... Loïc Dachary
06:25 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
The ceph-libs package is obsolete and the jerasure plugin now lives in the ceph package. The problem does not come fr... Loïc Dachary
06:18 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
It looks like the ceph-libs package is not upgraded, which explains the core dump : master cannot successfully load a... Loïc Dachary
05:31 AM Bug #9153 (Fix Under Review): erasure-code: jerasure_matrix_dotprod segmentation fault due to pac...
"proposed fix":https://github.com/ceph/ceph/pull/2276 Loïc Dachary
05:22 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
If the ceph-libs package is upgraded before the ceph package, it is entirely possible that the shared library is repl... Loïc Dachary
04:47 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
The upgrade sequence
* dumpling
* firefly -> installs and load the jerasure plugin
* master -> installs an updat...
Loïc Dachary
04:41 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
The stack trace is bizarre. ECUtil::decode calls ErasureCodeJerasure::encode_chunks which makes no sense becase a) de... Loïc Dachary
04:29 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Got three VPS with rhel 6.5 installed, running the job on them with no "nuke-on-error" Loïc Dachary
03:43 AM Bug #9153 (In Progress): erasure-code: jerasure_matrix_dotprod segmentation fault due to package ...
As soon as VPS are available, lock three and run the job again hoping to repeat it... Loïc Dachary
01:22 AM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Ack Loïc Dachary
04:47 PM Revision b016f846 (ceph): doc: add notes on using "ceph fs new"
Signed-off-by: John Spray <john.spray@redhat.com> john
04:40 PM Revision 4b15d011 (ceph): use the right syntax for RHEL/CentOS distros to check for ceph status
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
04:22 PM Revision 948178aa (ceph): ceph_mon: check for existing mon store before opening db
This will make sure that the monitor data directory is populated before
proceeding to opening it up. We have to do t...
Joao Eduardo Luis
04:02 PM Revision 8336f81c (ceph): 0.84
Jenkins
03:47 PM Revision bda23018 (ceph): qa/workunits/rbd/qemu-iotests: touch common.env
This seems to be necessary on trusty.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry...
Sage Weil
03:41 PM Revision a02c90b8 (ceph): suites/hadoop: update workunit names
These were renamed/reorged.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
03:39 PM Revision 1dc1fb8a (ceph): qa/workunits/hadoop: move all hadoop tests into a hadoop/ dir
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
03:38 PM Revision 3d3fcc98 (ceph): qa/workunits/hadoop-wordcount: fix/use -rmr command
-rm -r -f ... doesn't seem to work; use -rmr instead.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
03:37 PM Revision adaf5a6a (ceph): qa/workunits/hadoop-wordcount: use -x
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
02:42 PM Feature #9161 (New): Cache warmup and ejection
Initial access of an object in a high performance cache tier can have high latency as the object is fetched from the ... Neil Levine
02:20 PM rgw Bug #9160 (Closed): rgw failures with 'NoneType' object has no attribute 'get_contents_as_string'

Several jobs in this suite failed with this error:
http://pulpito.ceph.com/john-2014-08-18_16:28:28-rgw-wip-object...
John Spray
01:56 PM rgw Bug #9125: rgw: swift tests fail with civetweb
looks like the fix is merged to master, tested it on master branch and it worked fine.
will mark it as "Resolved"...
Tamilarasi muthamizhan
10:45 AM Bug #9158 (Duplicate): osd crashed in upgrade:dumpling-x:stress-split-master-distro-basic-vps suite
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-17_11:40:01-upgrade:dumpling-x:stress-split-master... Yuri Weinstein
10:24 AM Bug #9072 (Resolved): error setting 'mon_pg_warn_min_objects' to '10K': (22) Invalid argument
Sage Weil
09:23 AM Bug #9072: error setting 'mon_pg_warn_min_objects' to '10K': (22) Invalid argument
I checked the firefly branch and Sage cherry-picked the required patches to it.
That ought to fix all issues with ...
Joao Eduardo Luis
09:08 AM devops Feature #9118: ceph-deploy: Add pre-generated keys to a Monitor
Keith Schincke wrote:
> Can the precreated/populated keyring be propagated with the ceph-deploy command when the clu...
Sage Weil
09:04 AM devops Feature #9118: ceph-deploy: Add pre-generated keys to a Monitor
Can the precreated/populated keyring be propagated with the ceph-deploy command when the cluster is created? Keith Schincke
08:23 AM Bug #9156 (Resolved): SWIFT tests failed in upgrade:dumpling:rgw-dumpling-distro-basic-vps suite
12 tests total failed in http://pulpito.front.sepia.ceph.com/teuthology-2014-08-17_12:05:01-upgrade:dumpling:rgw-dump... Yuri Weinstein
05:17 AM Bug #9112 (Resolved): (wip-objecter) librados notify calls freezing
No longer occurring after reinstating _recalc_linger_op_target and updating related bits of code John Spray
03:54 AM Revision c7948afa (ceph): qa/workunits/rbd/qemu-iotests: touch common.env
This seems to be necessary on trusty.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry...
Sage Weil
03:54 AM Revision 486deefd (ceph): qa/workunits/rbd/qemu-iotests: touch common.env
This seems to be necessary on trusty.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry...
Sage Weil
03:54 AM Revision 055be68c (ceph): qa/workunits/rbd/qemu-iotests: touch common.env
This seems to be necessary on trusty.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
03:49 AM Revision 313e60b3 (ceph): Merge pull request #2010 from ceph/wip-misplaced
osd: track misplaced objects separately from degraded objects
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
12:15 AM Revision a60267d8 (ceph): rgw: httpd instead of httpd.worker
httpd exists on rhel 6.5 too ...
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f7b32bcc318c4...
Sage Weil
12:15 AM Revision 211fd6d1 (ceph): rgw: need alll of mod unixd, version, authz
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6392758f1bf503c3bbb2b3ad2f2ee9e6f79d6841)
Sage Weil
12:15 AM Revision ea023597 (ceph): tasks/rgw: include mod_authz
As per http://www.webhostingtalk.com/showthread.php?t=1173594
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry pic...
Sage Weil
12:15 AM Revision d14fb136 (ceph): tasks/rgw: get mpm_event frmo mods-available, not mods-enabled
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2aae91929f720c54f406b7612631d82b27c98fe6)
Sage Weil
12:15 AM Revision 8b5f114b (ceph): Add extra conf for Apache 2.4
Inside a conditional to affect only 2.4, set User, Group, and the
module config to load mpm_event. This is normally ...
Dan Mick

08/17/2014

11:52 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Meanwhile, I have been doing a little digging of my own: if I disable dio or aio via
[osd]
journal [d,a]io = fals...
Mark Kirkwood
11:40 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Here's the log with that patch applied. Mark Kirkwood
07:27 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Hi Mark,
Could you test again? I add more debug message this time.
Thanks!
jianpeng ma
08:53 PM rbd Bug #8919 (Resolved): qemu-iotests fails to find common.env
Sage Weil
05:15 PM Bug #9153: erasure-code: jerasure_matrix_dotprod segmentation fault due to package upgrade race
Loic, can you take a look? Sage Weil
04:38 PM Bug #9153 (Resolved): erasure-code: jerasure_matrix_dotprod segmentation fault due to package upg...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-17_11:30:03-upgrade:dumpling-firefly-x-master-dist... Yuri Weinstein
03:35 PM Revision ff893d55 (ceph): mailmap: Andreas Peters name normalization
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
03:35 PM Revision 4cdec2f1 (ceph): mailmap: Owen Synge affiliation
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
03:35 PM Revision 4d5d0d96 (ceph): mailmap: Pavan Rallabhandi affiliation
Reviewed-by: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
01:00 PM CephFS Bug #9152 (Resolved): mds: beacon needs to not take mds_lock
any random task that holds the mds lock for a long time prevents beacons, which will trigger a failover Sage Weil
12:48 PM CephFS Bug #9151 (Resolved): mds should log/error/warn when segments are NOT getting trimmed
Sage Weil
11:20 AM Revision 725aa4e7 (ceph): mailmap: Huamin Chen affiliation
and name normalization
Reviewed-by: Huamin Chen <hchen@redhat.com>
Signed-off-by: Loic Dachary <loic-201408@dachary....
Loic Dachary
11:20 AM Revision 78de7e13 (ceph): mailmap: Zhiqiang Wang affiliation
and name normalization
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
11:20 AM Revision 309c98bd (ceph): mailmap: Xiaoxi Chen affiliation
Reviewed-by: Xiaoxi Chen <xiaoxi.chen@intel.com>
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
08:43 AM Revision 55227aae (ceph): mailmap: Shu, Xinxin affiliation
and name normalization
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
08:43 AM Revision a06affbb (ceph): mailmap: Thorsten Behrens affiliation
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
08:43 AM Revision bac88db0 (ceph): mailmap: Luis Pabón affiliation
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
08:43 AM Revision cc3b5ada (ceph): mailmap: Abhishek Lekshmanan affiliation
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
Loic Dachary
05:07 AM Revision 6d7a229c (ceph): qa/workunits/rest/test.py: fix rd_kb -> rd_bytes
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
05:07 AM Revision 3279f3e7 (ceph): qa/workunits/rest/test.py: do snap test on our data2/3 pool
This way it works when a 'data' pool doesn't already exist.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:07 AM Revision 5045c5cb (ceph): qa/workunits/rest/test.py: use rbd instead of data pool for size tests
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
05:04 AM Revision 284647f3 (ceph): Merge pull request #2272 from ceph/wip-8621
Wip 8621
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil

08/16/2014

11:44 PM Revision 6392758f (ceph): rgw: need alll of mod unixd, version, authz
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:44 PM Revision f7b32bcc (ceph): rgw: httpd instead of httpd.worker
httpd exists on rhel 6.5 too ...
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
10:24 PM Revision cc4e6258 (ceph): unittest_strtol: fix compilation warning
Was fixed in master by a4923f5bc373d530d1ffdf6c58a4d88139daedd2
Signed-off-by: Sage Weil <sage@redhat.com>
Joao Eduardo Luis
10:01 PM rgw Bug #8621 (Pending Backport): civetweb frontend fails authentication if URL has special chars
Sage Weil
09:55 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Sage's comment suggested I check something - reverting 4eb18dd487da4cb621dcbecfc475fc0871b356ac from wip-9073 and run... Mark Kirkwood
08:59 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I've reverted commit:4eb18dd487da4cb621dcbecfc475fc0871b356ac on next so we can release v0.84. once we sort this out... Sage Weil
12:47 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
jianpeng ma wrote:
> I read the #6003. I think they are not the same.
> You can see those two files (patch.diff ...
Sage Weil
09:53 PM Feature #9030 (Resolved): mon: quickly identify 'problem'  osds
Sage Weil
09:51 PM Revision 0e07f7f0 (ceph): osd: fix theoretical use-after-free of OSDMap
In practice, the map will remain pinned for a while, but this
will make coverity happy.
*** CID 1231685: Use after ...
Sage Weil
09:26 PM Bug #9150 (Can't reproduce): osd/ECBackend.cc: 529: FAILED assert(pop.data.length() == sinfo.alig...
... Sage Weil
08:57 PM rgw Bug #9137 (Resolved): AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
Sage Weil
04:56 PM rgw Bug #9137: AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
works on el6 and el7. fc20 fails the ceph-qa-chef because of tiobench. Sage Weil
02:16 PM rgw Bug #9137: AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
verfied to work on precise and trusty.
still need to test on el6, el7, and fedora.
Sage Weil
08:56 PM Revision 27b7ecee (ceph): tasks/rgw: include mod_authz
As per http://www.webhostingtalk.com/showthread.php?t=1173594
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:52 PM rgw Bug #9148 (Resolved): rgw: multiregion tests failing, s3tests.functional.test_s3.test_region_copy...
... Sage Weil
08:41 PM Revision 44a0e376 (ceph): Merge pull request #2259 from ceph/wip-9039
Wip 9039
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
08:37 PM Revision 2aae9192 (ceph): tasks/rgw: get mpm_event frmo mods-available, not mods-enabled
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:19 PM Revision dbc3f650 (ceph): osd: track last_undegraded pg stat
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:19 PM Revision 5168907f (ceph): osd: track last_fullsized in pg_stat_t
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:19 PM Revision 190dc2f3 (ceph): mon: track pgs that get stuck degraded
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:19 PM Revision 932e4787 (ceph): mon: track stuck undersized
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:19 PM Revision 904a5f1c (ceph): vstart.sh: make filestore fd cache size smaller
I hit the fd limit on a vstart cluster with the default 128; reduce this
to 16.
Signed-off-by: Sage Weil <sage@redha...
Sage Weil
08:19 PM Revision 1907066f (ceph): osd/osd_types: add last_undegraded, last_undersized to pg_stat_t
Keep track of the last time the PG was known to not be degraded or
undersized.
Signed-off-by: Sage Weil <sage@redhat...
Sage Weil
08:18 PM Revision 6c0a2134 (ceph): osd/PG: account for misplaces separately than degraded
A degraded object does not have enough replicas or shards, while a
misplaced object is not stored in the correct plac...
Sage Weil
08:18 PM Revision b037e47a (ceph): osd: add PG_STATE_UNDERSIZED
This is a distinct concept from degraded.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:18 PM Revision 6d6767d3 (ceph): osd/PG: track PG_STATE_UNDERSIZED separately from DEGRADED
DEGRADED means there are objects without complete reduncancy; also check
for needs_recovery().
UNDERSIZED means acti...
Sage Weil
08:18 PM Revision a3149994 (ceph): osd: num_objects_misplaced
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
08:18 PM Revision d734d7fc (ceph): mon: warn about misplaced objects, just like degraded
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:18 PM Revision 988d7419 (ceph): librados: approximate legacy 'degraded' value
The librados API returns a degraded count and no misplaced count. Sum them
to approximate the old behavior.
Signed-...
Sage Weil
08:15 PM Revision 34fe7a82 (ceph): Merge pull request #2217 from ceph/wip-problem-osds
mon: 'ceph osd blocked-by' for histogram of peers OSDs are waiting for
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
08:06 PM Revision 14614e01 (ceph): qa/workunits/rest/test.py: fix 'df' test to use total_used_bytes
This changed back in ee2dbdb0f5e54fe6f9c5999c032063b084424c4c
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
07:56 PM Revision 93c5b25b (ceph): Revert "os/FileJournal: Update the journal header when closing journal"
This reverts commit 4eb18dd487da4cb621dcbecfc475fc0871b356ac.
This may be responsible for #9073. Until that is reso...
Sage Weil
07:43 PM Revision 29e93f7c (ceph): os/FileStore: rename start_sync() -> do_force_sync()
Just to avoid confusion with start_sync(Context*c), which is totally
different.
Signed-off-by: Sage Weil <sage@redha...
Sage Weil
07:42 PM Revision dd11042f (ceph): os/FileStore: fix mount/remount force_sync race
Consider:
- mount
- sync_entry is doing some work
- umount
- set force_sync = true
- set done = true
- syn...
Sage Weil
04:18 PM Revision ee9e1ead (ceph): Merge pull request #2271 from ceph/wip-9053
paxos: fix problem with disjoint quorum members
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil
03:42 PM CephFS Bug #8574 (Resolved): teuthology: NFS mounts on trusty are failing
chef adds a dummy export and restarts nfs-kernel-server now Sage Weil
02:41 PM CephFS Bug #8574: teuthology: NFS mounts on trusty are failing
root@mira055:~# service nfs-kernel-server restart
* Stopping NFS kernel daemon ...
Sage Weil
02:08 PM Linux kernel client Bug #9147 (Closed): krbd: run_xfstests.sh fails
... Sage Weil
02:07 PM rbd Bug #9146 (Can't reproduce): EPERM from image_read.sh
... Sage Weil
01:54 PM rgw Bug #9039: Using COPY on radosgw to copy object from one bucket to another that's in another pool...
The restriping tool never made it to dumpling. It actually isn't even in firefly. Yehuda Sadeh
01:39 PM rgw Bug #9039 (Pending Backport): Using COPY on radosgw to copy object from one bucket to another tha...
the restriping fix patches also need to go to dumpling... Sage Weil
01:46 PM Bug #8997: ceph_test_rados_watch_notify hangs
ubuntu@teuthology:/a/sage-2014-08-15_21:44:35-rados-master-testing-basic-multi/427533 (probably) Sage Weil
01:43 PM Bug #9145 (Resolved): recursive lock of CollectionIndex::access_lock (52)
... Sage Weil
01:17 PM Feature #7238: erasure code : implement LRC plugin
"running teuthology test run":http://pulpito.ceph.com/loic-2014-08-16_22:17:50-upgrade:firefly-x:stress-split-wip-723... Loïc Dachary
12:41 PM Bug #9144 (Fix Under Review): filestore: commit triggered during journal replay
https://github.com/ceph/ceph/pull/2274 Sage Weil
09:26 AM Bug #9144 (Resolved): filestore: commit triggered during journal replay
... Sage Weil
09:38 AM Feature #9033 (Resolved): erasure-code: simplified LRC
"part of a larger pull request":https://github.com/dachary/ceph/commit/43b8f66797184b1138560184708573aa6930e8c4 Loïc Dachary
09:15 AM Bug #9053 (Pending Backport): mon/Paxos.cc: 628: FAILED assert(begin->last_committed == last_comm...
Sage Weil
09:06 AM Revision 03959147 (ceph): mailmap: Loic Dachary affiliation
Signed-off-by: Loic Dachary <loic-201408@dachary.org> Loic Dachary
07:47 AM Bug #9143 (Rejected): Incorrect key sequence in encoding object name to key for GenericObjectMap
For example, two oid has same hash and their name is:
A: "rb.data.123"
B: "rb-123"
In ghobject_t compare level, ...
Haomai Wang
06:02 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"all green !":http://pulpito.ceph.com/loic-2014-08-16_10:42:43-upgrade:firefly-x:stress-split-wip-9025-chunk-remappin... Loïc Dachary
05:37 AM Revision 9de5bd1d (ceph): Add extra conf for Apache 2.4
Inside a conditional to affect only 2.4, set User, Group, and the
module config to load mpm_event. This is normally ...
Dan Mick
04:47 AM Revision 7f9fe22a (ceph): Fix EINVAL err when use "ceph tell osd.* bench"
Signed-off-by: huangjun <hjwsm1989@gmail.com>
(cherry picked from commit 7dc93a9651f602d9c46311524fc6b54c2f1ac595)
huang jun
04:42 AM Revision 14607fc0 (ceph): qa/workunits/cephtool/test.sh: fix get erasure_code_profile test
Manual backport of 4d6899c7560e990650959b442980a7249f0ba4c1
Signed-off-by: Sage Weil <sage@redhat.com>
Ma Jianpeng
04:37 AM Revision c83c90ca (ceph): rgw: update civetweb submodule
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> Yehuda Sadeh
04:36 AM Revision 068b906e (ceph): mon: OSDMonitor: add 'osd pool get-quota' command
Enables us to obtain current quotas for a given pool.
Fixes: #8523
Signed-off-by: Joao Eduardo Luis <joao.luis@inkt...
Joao Eduardo Luis
04:30 AM Revision 2ed0fa03 (ceph): mon: name instead of id in "has tiers" message
Instead of "Pool foo has tiers 1 2" print
"Pool foo has tiers bar baz".
Signed-off-by: John Spray <jspray@redhat.com...
John Spray
04:01 AM Revision 7e509b1b (ceph): common/config.cc: allow integer values to be parsed as SI units
We are allowing this for all and any integer values; that is, OPT_INT,
OPT_LONGLONG, OPT_U32 and OPT_U64.
It's on th...
Joao Eduardo Luis
04:00 AM Revision 4e34d83e (ceph): test/strtol.cc: Test 'strict_strtosi()'
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 40587d4792fd55db72d33870aae8b6a80...
Joao Eduardo Luis
04:00 AM Revision 3b06b8f4 (ceph): common/strtol.cc: strict_strtosi() converts str with SI units to uint64_t
Accepts values with a suffix (B, K, M, G, T, P, E) and returns the
appropriate byte value.
E.g., 10B = 10, while 10K...
Joao Eduardo Luis
12:15 AM Revision 08772fd8 (ceph): ceph-disk: warn about falling back to sgdisk (once)
This way the user knows something funny might be up if dmcrypt is in use.
Signed-off-by: Sage Weil <sage@redhat.com>...
Sage Weil
12:15 AM Revision 92ab6ecf (ceph): ceph-disk: linter cleanup
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit d74ed9d53fab95f27a9ad8e9f5dab719299...
Alfredo Deza
12:15 AM Revision 8a8a9c35 (ceph): Revert "Fix for bug #6700"
This reverts commit 673394702b725ff3f26d13b54d909208daa56d89.
This appears to break things when the journal and data...
Sage Weil
12:15 AM Revision 1804a834 (ceph): ceph-disk: fix log syntax error
File "/usr/sbin/ceph-disk", line 303, in command_check_call
LOG.info('Running command: %s' % ' '.join(arguments))...
Sage Weil
12:15 AM Revision 6775f46e (ceph): ceph-disk: use partition type UUIDs, and blkid
Use blkid to give us the GPT partition type. This lets us distinguish
between dmcrypt and non-dmcrypt partitions. F...
Sage Weil
12:15 AM Revision b09e6592 (ceph): ceph-disk: move fs mount probe into a helper
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f80ed26d2403ba12e80da6459fc45c22584f72de)
Sage Weil
12:15 AM Revision 8c04e475 (ceph): ceph-disk: display information about dmcrypted data and journal volumes
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c7a1ceba441fa99a82e19ed2cd3c6782a5d77636)
Sage Weil
12:15 AM Revision a45e815c (ceph): ceph-disk: add get_partition_base() helper
Return the base devices/disk for a partition device.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from ...
Sage Weil
12:15 AM Revision 50166efd (ceph): ceph-disk: only fall back to sgdisk for 'list' if blkid seems old
If the blkid doesn't show us any ID_PART_ENTRY_* fields but we know it is
a GPT partition, *then* fallback. Otherwis...
Sage Weil
12:15 AM Revision 80896dcf (ceph): support dmcrypt partitions when activating
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit ef8a1281512c4ee70a3764b28891da691a1...
Alfredo Deza
12:15 AM Revision 40649902 (ceph): ceph-disk: fix list for encrypted or corrupt volume
Continue gracefully if an fs type is not detected, either because it is
encrypted or because it is corrupted.
Signed...
Stuart Longland
12:15 AM Revision c9847ef0 (ceph): better error reporting on incompatible device requirements
Signed-off-by: Alfredo Deza <alfredo@deza.pe>
(cherry picked from commit 1ac3a503a15ddf7f7c1a33310a468fac10a1b7b6)
Alfredo Deza
12:15 AM Revision 50b700fa (ceph): ceph-disk: fix verify_no_in_use check
We only need to verify that partitions aren't in use when we want to
consume the whole device (osd data), not when we...
Sage Weil
12:05 AM Revision 1b0da81d (ceph): init-ceph: don't use bashism
-z STRING
the length of STRING is zero
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from ...
Sage Weil

08/15/2014

11:42 PM Revision a14a700a (ceph): Merge pull request #2270 from ceph/wip-init-ceph
init-ceph: don't use bashism
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza
11:41 PM Revision 0d6d1aa7 (ceph): init-ceph: don't use bashism
-z STRING
the length of STRING is zero
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
11:40 PM Revision dc611e86 (ceph): Merge pull request #2247 from ceph/wip-ceph-disk
ceph-disk: fix various dmcrypt bugs
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza
10:33 PM Revision 9005f33d (ceph): osd: fix feature requirement for mons
These features should be set on the client_messenger, not
cluster_messenger.
Backport: firefly
Signed-off-by: Sage W...
Sage Weil
10:19 PM Revision 082db05c (ceph): Merge pull request #2269 from ceph/wip-osd-mon-feature
osd: fix mon feature requirement
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
10:14 PM Revision 48a8434f (ceph): Merge pull request #97 from dachary/firefly
rgw.yaml: keep the client.0 key even if empty (firefly) Yuri Weinstein
10:09 PM Revision 12ded3c9 (ceph): Merge pull request #96 from dachary/wip-8988-default-idle-timeout
rgw.yaml: keep the client.0 key even if empty Yuri Weinstein
10:01 PM Revision 1d0c66ae (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
10:00 PM Revision 7df67a54 (ceph): Fix -Wno-format and -Werror=format-security options clash
This causes build failure in latest fedora builds, ceph_test_librbd_fsx adds -Wno-format cflag but the default AM_CFL... Boris Ranto
09:50 PM Revision 703c5f4f (ceph): Merge pull request #89 from dachary/wip-9027-create-unique-pool
initialize ctx.manager in ceph.py
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:29 PM Revision ae0b9f17 (ceph): osd: fix feature requirement for mons
These features should be set on the client_messenger, not
cluster_messenger.
Backport: firefly
Signed-off-by: Sage W...
Sage Weil
09:11 PM Revision d9e96b17 (ceph): Merge pull request #2268 from ceph/wip-9119
Wip 9119
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:04 PM Revision 50404130 (ceph): ReplicatedPG::cancel_copy: clear cop->obc
Otherwise, an objecter callback might still be hanging
onto this reference until after the flush.
Fixes: #8894
Intro...
Samuel Just
09:04 PM Revision 0db3e511 (ceph): ReplicatedPG::maybe_handle_cache: do not forward RWORDERED reads
Even with READFORWARD, we can't forward RWORDERED reads.
Fixes: #9119
Backport: firefly
Signed-off-by: Samuel Just <...
Samuel Just
09:04 PM Revision 9e46c5ad (ceph): unittest_osdmap: test EC rule and pool features
TODO: tiering feature bits.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2f0e2951d773b6acce...
Sage Weil
09:04 PM Revision dbbe0c62 (ceph): unittest_osdmap: create an ec pool in test osdmap
This is part of 7294e8c4df6df9d0898f82bb6e0839ed98149310.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:56 PM Revision c2aa74a2 (ceph): osd: only require crush features for rules that are actually used
Often there will be a CRUSH rule present for erasure coding that uses the
new CRUSH steps or indep mode. If these ru...
Sage Weil
08:56 PM Revision 12430fed (ceph): crush: add is_v[23]_rule(ruleid) methods
Add methods to check if a *specific* rule uses v2 or v3 features. Refactor
the existing checks to use these.
Signed...
Sage Weil
08:55 PM Revision eb589428 (ceph): Merge pull request #2264 from ceph/wip-crush-features
do not require crush features for rules that aren't being used
Reviewed-by: Loic Dachary <loic@dachary.org>
Sage Weil
08:54 PM Revision 2f0e2951 (ceph): unittest_osdmap: test EC rule and pool features
TODO: tiering feature bits.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:41 PM Revision e4d238bb (ceph): Merge pull request #2266 from kevincox/removewirehsark
Remove Old Wireshark Dissectors
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
08:37 PM Revision cab47936 (ceph): Merge pull request #2070 from somnathr/wip-sd-filestore-optimization
Wip sd filestore optimization
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
08:06 PM Revision 984f3ce4 (ceph): rgw.yaml: keep the client.0 key even if empty
Because rgw.py iterates over it to run the rgw server. If it is removed
the rgw servers are not run and all fails.
S...
Loïc Dachary
08:03 PM Revision bcb9c1f8 (ceph): rgw.yaml: keep the client.0 key even if empty
Because rgw.py iterates over it to run the rgw server. If it is removed
the rgw servers are not run and all fails.
S...
Loïc Dachary
07:50 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I read the #6003. I think they are not the same.
You can see those two files (patch.diff Magnifier (571 Bytes) ji...
jianpeng ma
06:19 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
This is starting to sound a lot like #6003! Sage Weil
01:56 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
hexdump -n8 journalblk-prestart.txt
0000000 3000 021d 0000 0000
Mark Kirkwood
12:09 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you paste the journal head after this command. Only first 8byte. jianpeng ma
07:27 PM Revision 0b276103 (ceph): Remove Old Wireshark Dissectors
Remove the two old Wireshark plugins. They do not build and are
superseded by the dissector which is inside Wireshar...
Kevin Cox
05:43 PM Revision 2427ea8f (ceph): PGLog: fix clear() to avoid the IndexLog::zero() asserts
Introduced in:
c5b8d8105d965da852c79add607b69d5ae79a4d4
ac11ca40b4f4525cbe9b1778b1c5d9472ecb9efa
Signed-off-by: S...
Samuel Just
04:41 PM rgw Bug #9137: AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
The MPM selection is supposed to be made in the default config. Dan Mick
01:23 PM rgw Bug #9137: AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
Looking into this; my theory is two problems: 1) package structure changed in 2.4 and we might need to explicitly in... Dan Mick
11:46 AM rgw Bug #9137 (Resolved): AH00534: apache2: Configuration error: No MPM loaded. (rpm distros)
... Sage Weil
03:55 PM Revision 1d954867 (ceph): crush: add is_v[23]_rule(ruleid) methods
Add methods to check if a *specific* rule uses v2 or v3 features. Refactor
the existing checks to use these.
Signed...
Sage Weil
03:55 PM Revision 16dadb86 (ceph): osd: only require crush features for rules that are actually used
Often there will be a CRUSH rule present for erasure coding that uses the
new CRUSH steps or indep mode. If these ru...
Sage Weil
03:49 PM Bug #9130 (Resolved): (wip-objecter) FAILED assert(cur_con) in MonClient
fix in wip-objecter Sage Weil
06:42 AM Bug #9130 (Resolved): (wip-objecter) FAILED assert(cur_con) in MonClient

http://pulpito.front.sepia.ceph.com/john-2014-08-15_03:34:51-rbd-wip-mds-contexts-testing-basic-multi/425519/
<p...
John Spray
02:08 PM Bug #9119 (Pending Backport): READFORWARD ordering bug
Sage Weil
02:03 PM RADOS Bug #8963 (Resolved): erasure coding crush rulset breaks rbd kernel clients on non-ec pools on Ub...
backported to firefly Sage Weil
01:56 PM Revision 821b2a43 (ceph): replace locally instantiated CephManager
Use the ctx.manager instance created by ceph.py instead
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
01:56 PM Revision f53ea258 (ceph): move functions from ceph to ceph_manager
mount_osd_data and make_admin_daemon_dir are only used by
ceph_manager.py although they are defined in ceph.py
Signe...
Loïc Dachary
01:56 PM Revision 9782465c (ceph): initialize ctx.manager in ceph.py
instead of rados.py because ceph.py is only run once where rados.py
could be run multiple time, leading to race condi...
Loïc Dachary
01:42 PM Revision e908e42e (ceph): Merge pull request #93 from ceph/wip-9126-firefly
rgw: s/idle_timeout/default_idle_timeout/ Sage Weil
01:34 PM Bug #9142 (Can't reproduce): [ RUN ] LibRadosTwoPoolsPP.PromoteSnapScrub hang
ubuntu@teuthology:/a/samuelj-2014-08-14_18:41:07-rados-wip-sam-testing-testing-basic-multi/425498 Samuel Just
01:33 PM Bug #9140: [ FAILED ] LibRadosTwoPoolsPP.PromoteOn2ndRead (9913 ms)
ubuntu@teuthology:/a/samuelj-2014-08-14_18:41:07-rados-wip-sam-testing-testing-basic-multi/425458 Samuel Just
01:30 PM Bug #9140 (Duplicate): [ FAILED ] LibRadosTwoPoolsPP.PromoteOn2ndRead (9913 ms)
2014-08-15T05:48:20.619 INFO:tasks.workunit.client.0.plana16.stdout:[ OK ] LibRadosTwoPoolsPP.HitSetWrite (2908... Samuel Just
01:32 PM Bug #9141 (Can't reproduce): [ RUN ] LibRadosAio.IsCompletePP hang
ubuntu@teuthology:/a/samuelj-2014-08-14_18:41:07-rados-wip-sam-testing-testing-basic-multi/425497 Samuel Just
01:30 PM Revision 592a524d (ceph): rgw: s/idle_timeout/default_idle_timeout/
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from da00662191e9ec173be50b551253f4460f369f9da)
Loïc Dachary
01:01 PM Bug #9139 (Rejected): ceph_test_rados reports incorrectly missing object
ORDERSNAPS was fixing something important:
1) cache-primary send DELETE on object we are flushing
2) base-primary q...
Samuel Just
11:32 AM Revision 3326a9fe (ceph): Merge pull request #90 from dachary/wip-8988-default-idle-timeout
rgw: s/idle_timeout/default_idle_timeout/
Reviewed-by: John Spray <john.spray@redhat.com>
John Spray
11:28 AM devops Feature #9134 (Duplicate): ceph-deploy: add pre-generated client keys to MON
9118 Neil Levine
11:22 AM devops Feature #9134 (Duplicate): ceph-deploy: add pre-generated client keys to MON
User story: As an admin, I have already generated Ceph client keys and would like to add them to the cluster during t... Neil Levine
11:27 AM devops Feature #9136 (Resolved): ceph-deploy: use pre-existing ceph.conf
User story: As an admin, I have already generated a ceph,conf file and would like to use it for a new cluster install... Neil Levine
11:26 AM Bug #9135 (Can't reproduce): ENOENT on collection_add
... Sage Weil
11:08 AM CephFS Feature #8869 (Resolved): MDS: support standby-replay on old-format journals
This merged a couple of weeks ago in https://github.com/ceph/ceph/commit/440c820cce2c262570ab78e352bed8a630d41be5 John Spray
10:49 AM devops Feature #9133 (Rejected): create ceph user/group; run daemons as ceph (non-root)
this will involve lots of updates to packaging. Sage Weil
10:43 AM Revision cb4c5649 (ceph): Merge pull request #2213 from dachary/wip-9025-chunk-remapping
erasure-code: chunk remapping
Reviewed-by: Samuel Just <sam.just@inktank.com>
Loïc Dachary
10:34 AM Revision da006621 (ceph): rgw: s/idle_timeout/default_idle_timeout/
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
05:33 AM Feature #7238: erasure code : implement LRC plugin
Teuthology job description:... Loïc Dachary
04:45 AM CephFS Bug #9105: ~ObjectCacher behaves poorly on EBLACKLISTED
Punting on a general purpose fix for ObjectCacher for the time being, and just fixing this in librbd teardown. John Spray
04:44 AM CephFS Bug #9105 (Fix Under Review): ~ObjectCacher behaves poorly on EBLACKLISTED
https://github.com/ceph/ceph/pull/2263 John Spray
04:33 AM Revision b22d693f (ceph): lttng: Add distro packaging
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
03:53 AM Bug #9128 (Resolved): Newly-restarted OSD may suicide itself after hitting suicide time out value...
Stop one OSD daemon for a long time, like many hours even to 1 day, without marking it as out. During this time, ther... Zhi Zhang
03:40 AM Feature #9025 (Resolved): erasure-code: chunk remapping
Loïc Dachary
03:38 AM Feature #9025: erasure-code: chunk remapping
Teuthology job passes. Loïc Dachary
01:25 AM Revision 6891f4ed (ceph): lttng: Fix "make distcheck"
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
12:55 AM Revision edbf82ad (ceph): rgw: add default_idle_timeout to allow override
Globally overriding the rgw idle_timeout is not possible because it it
needs to be done on a per client.0, client.1, ...
Loïc Dachary
12:41 AM Revision 4f9f023c (ceph): rest-api tests: enable debugging
Hoping to catch #9058
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil

08/14/2014

11:56 PM Revision c54f1e4d (ceph): mon/Paxos: share state and verify contiguity early in collect phase
We verify peons are contiguous and share new paxos states to catch peons
up at the end of the round. Do this each ti...
Sage Weil
11:55 PM Revision 3e5ce5f0 (ceph): mon/Paxos: verify all new peons are still contiguous at end of round
During the collect phase we verify that each peon has overlapping or
contiguous versions as us (and can therefore be ...
Sage Weil
11:25 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
The strace attached. So this is the mkfs...and wip-9073 with *just* the last patch applied. Mark Kirkwood
11:20 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you using strace to catch the ceh-osd command? Please using strace -f to cache all child process.
Thanks!
jianpeng ma
11:14 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Back to seeing the same error (invalid argument) with this latest patch :-( Mark Kirkwood
10:58 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Makr, i don't find the reason. But i think this bug may caused by patch. So i modify my patch and hope the bug don't ... jianpeng ma
10:58 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
...suggests a memory overwrite problem - we really need to get the binaries running under valgrind! Mark Kirkwood
08:11 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
With that last patch applied, journal header looks good every mkfs and osd is starting every time. Mark Kirkwood
07:47 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Yes. It's a great step. A strange bug.
The attachment is a patch which add read_header on some place.Can you try t...
jianpeng ma
07:41 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Very interesting: *somethimes* after the mkfs the header looks like:
0000000 b000 02b5 0000 0000 0001 0000 0000 00...
Mark Kirkwood
07:12 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Will do. Mark Kirkwood
06:57 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
But from the code, when start osd, read journal-header is the first thing for journal.
I don't know the command 's...
jianpeng ma
06:54 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Hmmm - just checked again and got:
$ hexdump journalblk-prestart.txt|head -1
0000000 3000 02a0 0000 0000 0001 000...
Mark Kirkwood
06:45 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Funny you should mention that, I had just check that myself:
So, just after the mkfs, journal header is:
$ hexd...
Mark Kirkwood
06:30 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Hi Mark,
I use different on my hand but i can't reproduce this.
From the deply.sh, for osd operation
1:ceph-osd ...
jianpeng ma
03:33 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you use "strace -f ceph-osd .." to trace all syscall?
We may from the info find some clue.
jianpeng ma
03:20 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
No, sorry,
$ sudo dd if=/dev/zero of=/dev/sdc1 bs=512
$ sudo ./deploy.sh
is the prescription. The result is os...
Mark Kirkwood
03:08 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Are you mean if you zero the journal-disk then the osd can start? Otherwise, it will met this bug. jianpeng ma
03:04 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Sure - I'm running the script attached initially - now using a minor variation thereof (attached again).
The only ot...
Mark Kirkwood
02:40 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
The first 8 byte of journal-header destroyed. But the debug info show the content of journal-header is right.
Now ...
jianpeng ma
02:06 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Also, I note that running
$ sudo ceph-osd -i 0 --mkjournal
results in a journal state that lets the osd start, ...
Mark Kirkwood
01:38 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Shame about no 520, but here are the files. Mark Kirkwood
11:07 PM Revision 164cfe85 (ceph): erasure-code: parse function for the mapping parameter
Each D letter is a data chunk. For instance:
_DDD_DDD
is going to parse into:
[ 1, 2, 3, 5, 6, 7 ]
the 0 a...
Loïc Dachary
11:07 PM Revision 5c2d2320 (ceph): erasure-code: remap chunks if not sequential
If the remap vector is not empty, use it to figure out the sequence of
data chunks.
http://tracker.ceph.com/issues/9...
Loïc Dachary
11:07 PM Revision 298da45c (ceph): erasure-code: ErasureCodeInterface::get_chunk_mapping()
Add support for erasure code plugins that do not sequentially map the
chunks encoded to the corresponding index. This...
Loïc Dachary
11:06 PM Revision ffac52b3 (ceph): rgw: disable civetweb url decoding
Fixes: #8621
We want to have the raw request uri, as we do the decoding ourselves.
Signed-off-by: Yehuda Sadeh <yeh...
Yehuda Sadeh
11:06 PM Revision 1293e6ba (ceph): rgw: log civetweb messages
Handle the civetweb log_message callback, divert messages into our debug
log.
Signed-off-by: Yehuda Sadeh <yehuda@re...
Yehuda Sadeh
11:06 PM Revision 0e74b7a1 (ceph): rgw: don't allow negative / invalid content length
Certain frontends (e.g., civetweb) don't filter such requests.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh
11:06 PM Revision 240764fc (ceph): rgw: update civetweb submodule
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> Yehuda Sadeh
11:02 PM Revision 8fb47299 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
10:28 PM Revision 95ac43f3 (ceph): FileStore: remove the fdcache_lock
With the changes to the shared_cache, we no longer need the fdcache_lock
to prevent us from inserting a second fd for...
Greg Farnum
10:28 PM Revision b04d84db (ceph): shared_cache: pass key (K) by const ref in interface methods
Signed-off-by: Somnath Roy <somnath.roy@sandisk.com> Somnath Roy
10:28 PM Revision 78d70daf (ceph): FileStore: Index caching is introduced for performance improvement
IndexManager now has a Index caching. Index will only be created if not
found in the cache. Earlier, each op is creat...
Somnath Roy
10:28 PM Revision cfff9f6a (ceph): FileStore: FDCache lookup is rearranged
In lfn_open() there is no point of building the Index if the
cache lookup is successful and caller is not asking for ...
Somnath Roy
10:28 PM Revision 3e7848d5 (ceph): FileStore: No need to hold Index lock during omap calls
The Index lock is held during all the omap calls which is
not necessary.
Signed-off-by: Somnath Roy <somnath.roy@san...
Somnath Roy
10:28 PM Revision b24db81e (ceph): FileStore: Introduced a RLock instead of WLock
While calling index->collection_version, there is no need to
hold WLock at the index level. RLock should be sufficien...
Somnath Roy
10:26 PM Revision a9f76d43 (ceph): FDCache: implement a basic sharding of the FDCache
This is just a basic sharding. A more sophisticated implementation would
rely on something other than luck for keepin...
Greg Farnum
09:11 PM Revision 4c2828ed (ceph): shared_cache: expose prior existence when inserting an element
The LRU now handles you attempting to insert multiple values for the
same key, by telling you that you've done so and...
Greg Farnum
08:50 PM Revision 435c6d6c (ceph): Merge pull request #2235 from kevincox/wireshark
doc: Add documentation about Wireshark dissector.
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
08:45 PM Revision a1e79dbb (ceph): rgw_admin: add --min-rewrite-stripe-size for object rewrite
A new param to check whether the object has requires restriping,
checking whether a specific object stripe is bigger ...
Yehuda Sadeh
08:42 PM Revision 46d8c97e (ceph): doc: Add documentation about Wireshark dissector.
Signed-Off-By: Kevin Cox <kevincox@kevincox.ca> Kevin Cox
08:35 PM Revision 6a555434 (ceph): rgw: fix compilation
RGWRadosPutObj couldn't refer to the ceph context.
Reviewed-by: Sage Weil <sage@redhat.com>
Signed-off-by: Yehuda Sa...
Yehuda Sadeh
08:28 PM Revision f6771f20 (ceph): shared_cache: use a single lookup for lookup() too
We didn't convert this one to use iterators before.
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum
08:18 PM Revision cec40dae (ceph): qa/workunits/cephtool: verify setmaxosd doesn't let you clobber osds
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:58 PM Revision a1c3afb6 (ceph): OSDMonitor: Do not allow OSD removal using setmaxosd
Description: Currently setmaxosd command allows removal of OSDs by providing
a number less than current max OSD numbe...
Anand Bhat
06:38 PM Revision 16a43609 (ceph): rgw: pass set_mtime to copy_obj_data()
Sometimes we need to set the mtime when copying object data (e.g., when
we rewrite the obj).
Signed-off-by: Yehuda S...
Yehuda Sadeh
06:28 PM Revision 800eff24 (ceph): rgw: copy_obj_data() uses atomic processor
Fixes: #9089
copy_obj_data was not using the current object write infrastructure,
which means that the end objects w...
Yehuda Sadeh
06:27 PM Revision a8cabfa6 (ceph): Merge pull request #2257 from ceph/wip-8784
rgw: call throttle_data() even if renew_state() failed
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
06:25 PM Revision 5d3a7e59 (ceph): rgw: copy object data if target bucket is in a different pool
Fixes: #9039
Backport: firefly
The new manifest does not provide a way to put the head and the tail in
separate pool...
Yehuda Sadeh
06:06 PM rgw Bug #9125 (Resolved): rgw: swift tests fail with civetweb
logs are copied to ubuntu@mira042.front.sepia.ceph.com:/home/ubuntu/civetweb_swift... Tamilarasi muthamizhan
05:57 PM rgw Bug #8971 (Duplicate): rgw: s3 test failures with civetweb
Tamilarasi muthamizhan
05:56 PM rgw Bug #8971: rgw: s3 test failures with civetweb
s3tests now pass on wip-8621 branch. Tamilarasi muthamizhan
05:55 PM rgw Bug #8621: civetweb frontend fails authentication if URL has special chars
s3tests passed with recent changes to wip-8621.
Tamilarasi muthamizhan
05:39 PM Bug #9058 (Need More Info): rest-api: long-running process may fail 'tell osd...' due to stale os...
ok, my theory doesn't seem right.. Objecter is checking for a new map if it gets ENXIO or similar. enabled logging i... Sage Weil
05:36 PM devops Bug #8330 (Resolved): repodata on rpm repos do not list latest ceph-deploy (1.5.2)
Thanks for verifying. Sandon Van Ness
05:33 PM devops Bug #8976 (Fix Under Review): httpd on RHEL7 (RHEL repo) incompatible with mod_fastcgi (ceph repo)
We have a new version available out at:
http://gitbuilder.ceph.com/apache2-rpm-rhel7-x86_64-basic/ref/master/
A...
Sandon Van Ness
05:13 PM Bug #8895: ceph osd pool stats (displayed incorrect values)
Can probably close this as dupe of #5884? John Spray
04:14 PM CephFS Bug #9101: multimds: unlinked file is not pruned from replica mds caches
Sage Weil
03:36 PM Revision 8393fdea (ceph): Merge pull request #2251 from ceph/wip-9102
ceph-disk: linter cleanup
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
03:28 PM Revision 2fc76d6e (ceph): Merge pull request #86 from dachary/wip-9027-create-unique-pool
rados.py: avoid CephManager creation race Zack Cerza
03:24 PM Revision 4783b378 (ceph): Merge pull request #87 from dachary/wip-8988-default-idle-timeout
rgw: add default_idle_timeout to allow override Yuri Weinstein
03:20 PM CephFS Bug #9123 (Can't reproduce): kceph: had 130k+ inodes with write caps
in #9121 the client had more than 130k inodes open for write, resulting in a huge file recovery queue. there definit... Sage Weil
02:37 PM CephFS Bug #9121 (In Progress): mds: inode stuck recovering after client restart
recovery is working.. there are just a lot of inodes queued:
2014-08-14 14:40:06.695087 7fd45f757700 10 mds.0.cach...
Sage Weil
02:10 PM CephFS Bug #9121 (Resolved): mds: inode stuck recovering after client restart
... Sage Weil
02:02 PM Revision aec684bf (ceph): add calamari to the api/index section
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
01:51 PM CephFS Bug #9105: ~ObjectCacher behaves poorly on EBLACKLISTED
John Spray wrote:
> This is happening when the librbd-using client is blacklisted, ObjectCacher fails to flush when ...
Sage Weil
10:16 AM CephFS Bug #9105: ~ObjectCacher behaves poorly on EBLACKLISTED
This is happening when the librbd-using client is blacklisted, ObjectCacher fails to flush when requested, and ImageC... John Spray
09:44 AM CephFS Bug #9105: ~ObjectCacher behaves poorly on EBLACKLISTED
Started failing in 061c8e93f76dc4fd6290d6d15723d76e73267444 where rbd_cache and rbd_cache_writethrough_until_flush we... John Spray
01:50 PM Revision 321d4def (ceph): Merge pull request #2255 from ceph/wip-9062
msg/PipeConnection: make methods behave on 'anon' connection
Reviewed-by: John Spray <john.spray@redhat.com>
Sage Weil
01:44 PM Revision f10a1384 (ceph): Merge remote-tracking branch 'origin/next'
John Spray
01:29 PM Revision d0331a96 (ceph): Merge pull request #2254 from ceph/wip-8725
mds: fix MDSMap encoding to be backward-compatible
Reviewed-by: Loic Dachary <loic@dachary.org>
Reviewed-by: John Sp...
John Spray
01:17 PM rgw Bug #8988 (Resolved): AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
Sage Weil
12:33 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"the suite runs ok":http://pulpito.ceph.com/loic-2014-08-14_14:25:55-upgrade:firefly-x:stress-split-wip-9025-chunk-re... Loïc Dachary
05:55 AM rgw Bug #8988 (Fix Under Review): AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"need review":https://github.com/ceph/ceph-qa-suite/pull/87 Loïc Dachary
05:36 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
The reason why "the suite fails":http://pulpito.ceph.com/loic-2014-08-14_09:47:05-upgrade:firefly-x:stress-split-wip-... Loïc Dachary
12:53 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
It failed for the same reason. "Rescheduled once more, hoping the problem has been fixed":http://pulpito.ceph.com/loi... Loïc Dachary
01:13 PM Bug #8865 (Resolved): cep osd setmaxosd doesn't check if osds exist
Sage Weil
01:09 PM Revision ac70490d (ceph): doc: update kernel recommendations (avoid 3.15!)
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:53 PM Revision 54a7298c (ceph): rgw: add default_idle_timeout to allow override
Globally overriding the rgw idle_timeout is not possible because it it
needs to be done on a per client.0, client.1, ...
Loïc Dachary
12:37 PM Feature #9025: erasure-code: chunk remapping
Now that the teuthology + MDS bugs are fixed, the following job will be scheduled to exercise remapping:... Loïc Dachary
11:10 AM Bug #9119 (Resolved): READFORWARD ordering bug
READFORWARD is forwarding RWORDERED reads. Samuel Just
11:06 AM devops Feature #9118: ceph-deploy: Add pre-generated keys to a Monitor
Any keys (client.admin or otherwise) in the keyring file passed to "ceph-mon --mkfs --keyring <foo>" will get seeded ... Sage Weil
10:56 AM devops Feature #9118 (Resolved): ceph-deploy: Add pre-generated keys to a Monitor
ceph-authtool can be used to generate a key and keyring before a Ceph cluster is running, if a user has access to the... Neil Levine
10:54 AM Feature #9083 (Closed): Standalone script to generate Ceph keys
Feature already exists in ceph-authtool Neil Levine
09:34 AM Bug #9113: osd: snap trimming eats memory, linearly
a few notes:... Sage Weil
06:40 AM Bug #9113 (Resolved): osd: snap trimming eats memory, linearly
- rados pool snapshot taken weekly
- trimmed when >30 days old
- trimming makes some osds consume memory linearly
...
Sage Weil
09:06 AM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
ubuntu@teuthology:/a/sage-2014-08-13_15:28:18-rados-next-testing-basic-multi/422862 Sage Weil
09:05 AM Bug #9114: osd: segv in build_push_op
note: i manually killed ceph_test_rados to make teuthology clean up Sage Weil
07:09 AM Bug #9114 (Duplicate): osd: segv in build_push_op
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-08-13_15:28:18-rados-next-testing-basic-multi/422759... Sage Weil
08:57 AM Revision 6237acb3 (ceph): rados.py: avoid CephManager creation race
gevent may hold the rados.py thread when it has an opportunity. The
if not hasattr(ctx, 'manager'):
must therefo...
Loïc Dachary
08:33 AM Bug #9102 (Resolved): ceph-disk has undefined variables
Sage Weil
07:51 AM Bug #9102 (Fix Under Review): ceph-disk has undefined variables
PR opened https://github.com/ceph/ceph/pull/2251 Alfredo Deza
07:58 AM rgw Documentation #9116 (Resolved): rgw: broken link
From Luis Pabon:... Yehuda Sadeh
07:21 AM devops Bug #9066 (Rejected): Need ceph-deploy to be able to run to JUST generate ceph.conf and keyring w...
The initial issue was misunderstood, ceph-deploy already is able to create a ceph.conf and a mon keyring. Other requi... Alfredo Deza
06:47 AM Bug #9062 (Resolved): Mon segfault in waitlist_or_zap_client
Sage Weil
06:40 AM Bug #9112 (In Progress): (wip-objecter) librados notify calls freezing
John Spray
06:39 AM Bug #9112: (wip-objecter) librados notify calls freezing
Client log with objecter and librados debug logging at 20 in teuthology:~/jcsp/9112 John Spray
06:28 AM Bug #9112 (Resolved): (wip-objecter) librados notify calls freezing

Hitting this in rbd tests, periodically the ceph_test_rados_fsx process gets stuck inside IoCtxImpl::notify
<pre...
John Spray
06:34 AM CephFS Bug #8725 (Resolved): mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
Sage Weil
06:16 AM devops Feature #9103: create a (generic) webservice to handle Sphinx documentation versions
1.- Adding something to the Sphinx build is non-trivial. Sphinx extensions (the right way to do this) are very comple... Alfredo Deza
04:47 AM Revision 69ef8545 (ceph): s/apache2/apache
Signed-off-by: tamil <tamil.muthamizhan@inktank.com> Tamilarasi muthamizhan
03:36 AM Revision 5374386e (ceph): doc: Added user management link to quick start.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:30 AM Revision 5e8eae72 (ceph): doc: Removed cephx intro. Moved details to user management, config, and...
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:29 AM Revision 3aa3e96b (ceph): doc: Removed cephx guide. Replaced by user management.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:28 AM Revision aac6aa2a (ceph): doc: Removed auth intro and auth docs and added user management to index.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:28 AM Revision 4c651b7d (ceph): doc: Restructured auth configuration reference.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:27 AM Revision 3faf37a3 (ceph): doc: Put architectural details of authentication in to architecture doc.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
03:27 AM Revision 118ae726 (ceph): doc: Created a new User Management doc to replace authentication.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
02:48 AM Bug #9111: PG stuck with 'active+remapped' forever with cluster wide change (add/remove OSDs)
Right after I filed this bug, I got some clue, I found the problem came from those removed OSDs (which has status DNE... Guang Yang
02:01 AM Bug #9111 (Won't Fix): PG stuck with 'active+remapped' forever with cluster wide change (add/remo...
After adding/removing OSDs, some PGs stuck with 'active+remapped' forever.
1. ceph -s
-bash-4.1$ ceph -s...
Guang Yang
01:35 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
Thanks Sage , the issue has been resolved, cluster is Healthy now. karan singh
12:52 AM Revision 0a49db8e (ceph): msg/PipeConnection: make methods behave on 'anon' connection
The monitor does a create_anon_connection() to create a pseudo Connection
object for forwarded messages. If we try t...
Sage Weil
12:21 AM Revision 662af7ac (ceph): osd: fix pg epoch floor tracking
If you call erase() on a multiset it will delete all instances of a value;
we only want to delete one of them. Fix t...
Sage Weil
12:21 AM Revision 289360c5 (ceph): osd: prevent pgs from getting too far ahead of the min pg epoch
Bound the range of PG epochs between the slowest and fastest pg
(epoch-wise) with 'osd map max advance'. This value ...
Sage Weil
12:21 AM Revision ad82f281 (ceph): osd: add sanity check/warning on a few key configs
Warn when certain config values are set to bad values.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@in...
Sage Weil
12:21 AM Revision fac5bfaf (ceph): osd: introduce simple sleep during scrub
This option is similar to osd_snap_trim_sleep: simply inject an optional
sleep in the thread that is doing scrub work...
Sage Weil
12:21 AM Revision 9f29788f (ceph): common/io_priority: wrap ioprio_set() and gettid()
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit a2b49110ef65efd526c3430ad03c988ca9dde768)
Sage Weil
12:21 AM Revision b75f85a2 (ceph): common/Thread: allow io priority to be set for a Thread
Ideally, set this before starting the thread. If you set it after, we
could potentially race with create() itself.
...
Sage Weil
12:21 AM Revision 11858d7e (ceph): common/WorkQueue: allow io priority to be set for wq
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit dd6badcb5eedfec6748b3e6ca4d46e3b266038f6)
Con...
Sage Weil
12:21 AM Revision 9ca451a6 (ceph): osd: allow io priority to be set for the disk_tp
The disk_tp covers scrubbing, pg deletion, and snap trimming
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry pic...
Sage Weil
12:21 AM Revision 229a346d (ceph): osd: track per-pg epochs, min
Add some simple tracking so that we can quickly determine what the min
pg osdmap epoch is.
Signed-off-by: Sage Weil ...
Sage Weil

08/13/2014

11:49 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Mark, i can't find the ssd in lab.
And i also can't find the code.But from my two patch, i don't modify code which c...
jianpeng ma
07:08 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
hexdump journalblk.txt
0000000 1000 03ce 0000 0000 0001 0000 0000 0000
0000010 bdb9 29ac 51d7 a343 3bbf 1114 622e...
jianpeng ma
06:51 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Here's the 4096 bytes of sdc1 Mark Kirkwood
06:41 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
For the code,there is a logic error.
int r = ::pread(fd, bp.c_str(), bp.length(), 0);
bl.push_back(bp);
try ...
jianpeng ma
06:21 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you read the 4096 of /dev/sdc1 and send to me?
The journal header is in first 4096 size.
jianpeng ma
06:12 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
The info for the Intel 520:
Re more journal debugging - sure, I already have the following set:
[osd]
debug os...
Mark Kirkwood
06:09 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
The script puts in symlinks (also note slightly different osd data path on the work machine):
$ ls -l /var/lib/cep...
Mark Kirkwood
06:04 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
From you message, I found:
14-08-14 10:58:01.735317 7f944f5e4800 20 journal _check_disk_write_cache: disk write cach...
jianpeng ma
05:36 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you send the 520 disk-info using hdparm to me?
I'll search the lab try to find this ssd.
Thanks!
jianpeng ma
05:13 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Can you print more debuginfo about journal?
From the messages:
journal read_header error decoding journal header
...
jianpeng ma
03:58 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Doing a secure erase of the 520's changes nothing. Still seeing problem 2/ 'invalid argument' opening the journal. Mark Kirkwood
01:55 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
An aside thought - to rule out weird ssd related stuff I had performed a secure erase on the Crucial m4's while inves... Mark Kirkwood
01:40 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I'm happy to report that wip-9073 definitely fixes problem 1/ (the hang). Mark Kirkwood
01:04 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
to (hopefully) clarify the errors:
- Home machine: osd mkfs hangs (which I've called 1/)
- work machine: osd mkfs...
Mark Kirkwood
12:56 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Interesting... I'm just building wip-9073 on my home machine now, will update you with what I find.
The issue *mig...
Mark Kirkwood
11:36 PM Revision 85129047 (ceph): lttng: Support --with-lttng=check
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
11:17 PM Revision 6411dae9 (ceph): Merge pull request #85 from dachary/wip-rgw-idle-timeout
overrides: rgw must not be nested in ceph Yuri Weinstein
11:01 PM Revision bb046ed0 (ceph): mon/Paxos: put source mon id in a temp variable
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
10:49 PM Revision f00afcc4 (ceph): overrides: rgw must not be nested in ceph
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
10:05 PM Revision d74d3f1c (ceph): mds/MDSMap: fix incompat version for encoding
Back in 8f7900a09c8e490c9cd3a6f92ed1f0eb1f47f2a9 we added the new fields
before the 'extended' section, which made th...
Sage Weil
10:03 PM Revision 369c6392 (ceph): mds/MDSMap: drop trailing else in favor of early return
This keeps the old-version special cases in one place and make it obvious
what the current/forward-looking path is.
...
Sage Weil
09:17 PM rgw Feature #8473: rgw: Shard bucket index objects to improve single bucket PUT throughput
Here is the first patch - https://github.com/ceph/ceph/pull/2187 Guang Yang
09:16 PM Bug #7521 (Won't Fix): Add more events (hold object context) to OpTracker to better analyze perfo...
With more understanding of the tracker, I found actually the issue being tracked by this but can be achieved by the c... Guang Yang
09:14 PM Bug #7710 (Resolved): Multiple rados bench instance will overwrite the metadata object
Guang Yang
09:11 PM Revision 0f6956b3 (ceph): Merge remote-tracking branch 'upstream/next'
Samuel Just
09:10 PM Revision e74ad5f8 (ceph): Merge pull request #2252 from ceph/wip-9087
test/system/systest_runnable.cc: debugging on start and end
Reviewed-by: Sage Weil <sage@redhat.com>
Samuel Just
09:10 PM Revision 473f4bd3 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
09:10 PM Documentation #6142: Ceph needs mor than 32k pids
John, not sure where this should go in the doc structure... Sage Weil
09:08 PM Revision b2c1fa86 (ceph): test/system/systest_runnable.cc: debugging on start and end
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
09:00 PM Revision d74ed9d5 (ceph): ceph-disk: linter cleanup
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
08:58 PM Revision c6292416 (ceph): rbd-replay: Convert prep-for-replay.py to rbd-replay-prep.cc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision febcce92 (ceph): rbd-replay: Add man page and package binaries for rbd-replay-prep
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision b5bc14ab (ceph): rbd-replay: Move tracepoints from internal.cc to librbd.cc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 89c38602 (ceph): rbd-replay: Remove extent tracepoints and inline extents
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 520b75b5 (ceph): rbd-replay: Fix bug in rbd-replay-prep, Thread::issued_io with wrong IO
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 3efa30d2 (ceph): lttng: Remove tracing from libcommon
This is a short-term fix for issues caused by tracepoints in libcommon.
Code crashes at runtime if the same tracepoin...
Adam Crume
08:58 PM Revision 77cf0348 (ceph): rbd-replay: Switch ImageNameMap to using rbd_loc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 6be0c2b3 (ceph): rbd-replay: Switch ImageNameMap from regex to plain string matching
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision a602bd42 (ceph): rbd-replay: Fix compiler warning in unit tests
Was getting:
test/test_rbd_replay.cc:44:3: warning: converting ‘false’ to pointer type for argument 1 of ‘char testin...
Adam Crume
08:58 PM Revision 86d0b576 (ceph): rbd-replay: Remove boost_regex from LIBADD and LDADD
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision de9327ca (ceph): lttng: Replace Boost dependencies with custom string code
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 95e3b0b7 (ceph): rbd-replay: Remove outdated message that says regexes are used
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision dca21ee8 (ceph): rbd-replay: Add man page, and add binaries to Debian packaging
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision ecd20a35 (ceph): lttng: Preload liblttng-ust-fork.so in TESTS_ENVIRONMENT
This adds LD_PRELOAD=liblttng-ust-fork.so to TESTS_ENVIRONMENT.
This prevents lttng from complaining when processes a...
Adam Crume
08:58 PM Revision b9151612 (ceph): rbd-replay: Add unit tests
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision d66ee7a3 (ceph): rbd-replay: Use non-zero fake data
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision ee4d2170 (ceph): rbd-replay: Support replaying partial traces
Tracing may start after the application is started, and image open calls
may missed. To support replaying these trace...
Adam Crume
08:58 PM Revision 11f55de8 (ceph): lttng: Fix ceph_ctf_stringp
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 0b68f01a (ceph): rbd-replay: Use standard image@snap instead of image/snap
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 429a0891 (ceph): rbd-replay: Add rbd_replay::rbd_loc
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision bbdd0dab (ceph): rbd-replay: Add command-line arguments to prep-for-replay.py
Includes:
--print-on-read
--print-on-write
--window
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
08:58 PM Revision 3a20acb3 (ceph): lttng: Default to not storing buffer content
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 02743766 (ceph): rbd-replay: Add assertions
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision c83856c6 (ceph): rbd-replay: Add --pool option
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision ff2b9f5f (ceph): rbd-replay: Add --map-image option for image name mapping
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision ae4e02a3 (ceph): lttng: Fix 'make check' by setting up fake HOME
If the HOME variable is not set, LTTng complains, and the extra output
fails to match what the tests expect, so they ...
Adam Crume
08:58 PM Revision c938829b (ceph): lttng: Disable LTTng by default, add --with-lttng configure option
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision a7da7e05 (ceph): rbd-replay: Fix memory leak in PendingIO
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 0a629a9f (ceph): doc: Add usage comment to librbd::RBD::AioCompletion
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision d4e07f03 (ceph): rbd-replay: Add --latency-multiplier option to usage statement
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision b58797e0 (ceph): rbd-replay: Hash completions across multiple mutexes to reduce contention
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 4f6e775c (ceph): rbd-replay: Clean up prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 9a318083 (ceph): rbd-replay: Fix examples/rbd-replay/replay
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 11186aff (ceph): rbd-replay: Add --read-only flag
Removes ignoreWrites flag in prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
08:58 PM Revision b11ab2ea (ceph): rbd-replay: Fix async writes in prep-for-replay.py
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 9ec70b97 (ceph): rbd-replay: Add rbd_replay subsystem
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision d8a86bfb (ceph): rbd-replay: Bugfix: ensure PendingIOs are added before they're removed
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 84edf375 (ceph): rbd-replay: Printing pending IOs when worker threads waiting to die
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision c4ba9972 (ceph): rbd-replay: Change Worker::m_pending_ios from vector to map
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 0129841f (ceph): rbd-replay: Print "Still waiting for" more judiciously
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision e06e979b (ceph): rbd-replay: Support writing Actions to ostreams
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision a06e347a (ceph): rbd-replay: Check that return values are non-negative
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 289c98df (ceph): lttng: Remove 'ver' from trace in code for CEPH_OSD_OP_NOTIFY
'ver' is obsolete and variable exists only for proper deserialization
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
08:58 PM Revision 4001b28d (ceph): lttng: Basic tracing of librbd
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision ef5c30af (ceph): tracing: instrument filestore
This adds objectstore tracepoints for the filestore. It'd be nice to add
these to the objectstore interface some how ...
Noah Watkins
08:58 PM Revision cffb8f3b (ceph): lttng: Add more tracing to librbd
Includes:
- aio_complete
- aio_discard
- aio_flush
- aio_get_return_value
- aio_is_complete
- aio_read
- aio_wait_for...
Adam Crume
08:58 PM Revision b73af7c0 (ceph): lttng: Add rbd-replay
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision fc712844 (ceph): rbd-replay: Switch logging from cout to dout
To enable logs, we also have to use global_init to parse our
command-line args, so we now have other standard Ceph go...
Adam Crume
08:58 PM Revision 6d987ce7 (ceph): Add assertf macro
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision fb45d605 (ceph): lttng: Trace ReplicatedPG::do_osd_ops
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 3e21f8e4 (ceph): lttng: Add tracing-common.h
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 1dcf8477 (ceph): examples: Add example for the librados C API
Simply copied and modified the C++ API example.
Signed-off-by: Adam Crume <adamcrume@gmail.com>
Adam Crume
08:58 PM Revision c4f29afa (ceph): lttng: Trace librados C API
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision be3c9133 (ceph): lttng: Prevent creation of incorrect .o files when generating from .tp
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 50020b06 (ceph): lttng: Prevent deletion of tracing-common.h when cleaning
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 8e238b51 (ceph): lttng: Split up libtracepoints
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 090f63e1 (ceph): lttng: trace mutex::unlock
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
08:58 PM Revision 8a28388c (ceph): lttng: add pg and osd tracepoints
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
08:58 PM Revision 64fbc28f (ceph): lttng: Check for lttng/tracepoint.h when configuring
Adam Crume
08:58 PM Revision 2f2d6e1d (ceph): tracing: automake-ify tracepoint generation
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> Noah Watkins
08:58 PM Revision d5b16e4a (ceph): lttng: Trace OpRequest
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 469c9377 (ceph): lttng: Add rmw_flags to tracepoint in PG::queue_op
Signed-off-by: Adam Crume <adamcrume@gmail.com> Adam Crume
08:58 PM Revision 7a9eb3fc (ceph): tracing: bootstrap lttng-ust with mutex events
See src/tracing/README.md
Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
Noah Watkins
08:32 PM Revision aaeebceb (ceph): mon: fix divide by zero when pg_num adjusted and no osds
Fixes: #9052
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Manual backport of 239401db7b515...
Sage Weil
08:32 PM Revision 38c3a3c0 (ceph): mon: fix divide by zero when pg_num adjusted and no osds
Fixes: #9052
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 239401...
Sage Weil
08:31 PM Revision c5687b13 (ceph): mon: fix potential divide by zero on can_mark_{down,out}
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:31 PM Revision 239401db (ceph): mon: fix divide by zero when pg_num adjusted and no osds
Fixes: #9052
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:15 PM Revision f81d2b0e (ceph): mon: fix divide by zero when pg_num adjusted and no osds
Fixes: #9101
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:15 PM Revision 8a647f7d (ceph): mon: fix potential divide by zero on can_mark_{down,out}
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:55 PM Revision 87bf00a2 (ceph): ceph_test_rados_api_tier: fix cache cleanup (ec too)
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d7fb7bf5f2059f411633751e376c2270e6040fba)
Sage Weil
07:55 PM Revision 3be0d731 (ceph): ceph_test_rados_api: fix cleanup of cache pool
We can't simply try to delete everything in there because some items may
be whiteouts. Instead, flush+evict everythi...
Sage Weil
07:55 PM Revision 7f511ef7 (ceph): librados/TestCase: inheret cleanup_default_namespace
No need to duplicate this code.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 1d199fb1173dc9...
Sage Weil
07:54 PM Revision b3fd9612 (ceph): Merge pull request #2236 from ceph/wip-9055
ceph_test_rados_api_tier: fix cache pool cleanup during test
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
07:40 PM Revision 14f73e03 (ceph): Merge pull request #2222 from ceph/wip-9029
mds: Make min/max UID configurable for who is allowed to create a snapsh...
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
07:07 PM Revision 5ed9f4ee (ceph): mds: Revert from mds_mksnap_ setting to mds_snap_ settings
Wido den Hollander
07:00 PM Revision 6f7798e3 (ceph): ceph-disk: warn about falling back to sgdisk (once)
This way the user knows something funny might be up if dmcrypt is in use.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
06:52 PM Revision c7a1ceba (ceph): ceph-disk: display information about dmcrypted data and journal volumes
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:52 PM Revision b75e8a34 (ceph): ceph-disk: add get_partition_base() helper
Return the base devices/disk for a partition device.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
06:52 PM Revision b1651afb (ceph): ceph-disk: only fall back to sgdisk for 'list' if blkid seems old
If the blkid doesn't show us any ID_PART_ENTRY_* fields but we know it is
a GPT partition, *then* fallback. Otherwis...
Sage Weil
06:20 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
Loic, we had disk failed and possibly due to that suite failed (guessing), I re-started it http://pulpito.front.sepia... Yuri Weinstein
04:11 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
Waiting for "Shipping apache config":https://github.com/ceph/ceph-qa-suite/blob/master/tasks/rgw.py#L82 with... Loïc Dachary
04:04 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"running a suite using the new VPS.yaml":http://pulpito.ceph.com/loic-2014-08-14_01:02:11-upgrade:firefly-x:stress-sp... Loïc Dachary
03:47 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"fix indentation of rgw override":https://github.com/ceph/ceph-qa-suite/pull/85 Loïc Dachary
03:35 PM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
To confirm there is a large delay requiring a large idle_timeout:... Loïc Dachary
03:33 PM rgw Bug #8988 (In Progress): AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
Sage Weil
05:48 PM Revision 561208af (ceph): Merge pull request #2249 from ceph/wip-9096
osd: fix require_same_peer_instance from fast_dispatch
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: ...
Samuel Just
05:35 PM Revision 4a5c93a9 (ceph): osd/ReplicatedPG: only do agent mode calculations for positive values
After a split we can get negative values here. Only do the arithmetic if
we have a valid (positive) value that won't...
Sage Weil
05:34 PM Revision 5be56ff8 (ceph): osd/ReplicatedPG: only do agent mode calculations for positive values
After a split we can get negative values here. Only do the arithmetic if
we have a valid (positive) value that won't...
Sage Weil
04:38 PM Revision 7b3714ce (ceph): osd: fix some line wrapping
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:30 PM Bug #9109 (New): ceph CLI: Help is missing -k keyring option
The ceph command line should provide a -k keyring argument. "ceph --help" does not appear to list the -k option for t... John Wilkins
04:28 PM Bug #9087 (Need More Info): ceph_test_rados_list_parallel hang
Sage Weil
02:21 PM Bug #9087: ceph_test_rados_list_parallel hang
added some debugging. Samuel Just
12:47 PM Bug #9087: ceph_test_rados_list_parallel hang
Looking Samuel Just
04:22 PM Bug #9053: mon/Paxos.cc: 628: FAILED assert(begin->last_committed == last_committed)
Paxos::handle_last() bug.
the peon:...
Sage Weil
04:17 PM Bug #9053: mon/Paxos.cc: 628: FAILED assert(begin->last_committed == last_committed)
Sage Weil
04:13 PM Revision 0b2a8472 (ceph): s/apache/apache2
Signed-off-by: tamil <tamil.muthamizhan@inktank.com> Tamilarasi muthamizhan
04:09 PM Revision 90379e0b (ceph): Added a facet to test both apache and civetweb as frontend for rgw
Signed-off-by: tamil <tamil.muthamizhan@inktank.com> Tamilarasi muthamizhan
04:05 PM Revision 42d66dee (ceph): Merge pull request #83 from ceph/wip_add_manual_override2
Added idle_timeout: to fix time-outs problems for some tests Sage Weil
03:59 PM Revision b5f7d843 (ceph): bug 8927: enable civetweb testing for rgw
Tamilarasi muthamizhan
03:35 PM CephFS Bug #8964 (Resolved): kcephfs: client does not resend requests on mds restart
Sage Weil
03:30 PM Revision df945a90 (ceph): osd: fix require_same_peer_instance from fast_dispatch
The mark-down of old peers needs to take the session_dispatch_lock in order
to safely clear the Session ref cycle. H...
Sage Weil
03:29 PM Revision 74025f91 (ceph): Fixed syntax
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
03:24 PM Revision 90f647f4 (ceph): Added idle_timeout: to fix time-outs problems for some tests
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
03:20 PM Revision 3d7e2b35 (ceph): osd: inline require_osd_up_peer
There is only one caller.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
03:13 PM CephFS Bug #8725 (Fix Under Review): mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic...
https://github.com/ceph/ceph/pull/2254 Sage Weil
03:06 PM Revision 374c3a8e (ceph): Merge pull request #81 from ceph/wip_add_manual_override
Added overrides for vps time outs Sage Weil
02:46 PM Cleanup #9106: ceph-authtool: Modifying user without --gen-key overwrites the key
Wasn't able to reproduce this after retrying. Maybe just a usage issue. John Wilkins
02:24 PM Cleanup #9106 (Resolved): ceph-authtool: Modifying user without --gen-key overwrites the key
If you are trying to modify a user's caps/permissions using ceph-authtool, and the user has an existing key, specifyi... John Wilkins
02:37 PM RADOS Feature #9108 (New): ceph auth get: Get multiple users
The "ceph auth get <user>" command with the -o option is an ideal way to create a keyring for an individual user. How... John Wilkins
02:37 PM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
Hmm, most likely a bug in repair. We should start by creating a teuthology task which reproduces the bug. Once we h... Samuel Just
02:27 PM RADOS Feature #9107 (New): ceph-authtool: Delete a user.
Currently, there is no corresponding "delete" feature that allows a user to delete a user from a keyring. We should h... John Wilkins
02:25 PM Feature #8389 (Resolved): osd: clean up old ec objects more aggressively
Samuel Just
02:25 PM Feature #8480 (Resolved): modify scrub to detect/repair obsolete rollback objects
Samuel Just
02:15 PM CephFS Bug #9105 (New): ~ObjectCacher behaves poorly on EBLACKLISTED

In ceph master 78dc4df
http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-11_23:00:01-rbd-master-testing-bas...
John Spray
01:59 PM devops Feature #9103: create a (generic) webservice to handle Sphinx documentation versions

The calamari docs already include a version (albeit a rather verbose one including the git hash). I guess with a l...
John Spray
01:06 PM devops Feature #9103 (Resolved): create a (generic) webservice to handle Sphinx documentation versions
None of our docs allow a user to:
* Have a visual queue of what version of the docs they are seeing.
* be warned ...
Alfredo Deza
01:44 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
... Sage Weil
01:19 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
... Sage Weil
01:39 PM CephFS Bug #9101: multimds: unlinked file is not pruned from replica mds caches
looks like the problem is that another mds has the inode in its cache and isn't trimming it (or being asked to trim i... Sage Weil
01:13 PM CephFS Bug #9101 (Fix Under Review): multimds: unlinked file is not pruned from replica mds caches
https://github.com/ceph/ceph/pull/2250 Sage Weil
11:36 AM CephFS Bug #9101: multimds: unlinked file is not pruned from replica mds caches
Here is the debug data when using a ceph-fuse client.
We did reproduce the problem
Stephane Boisvert
11:15 AM CephFS Bug #9101 (New): multimds: unlinked file is not pruned from replica mds caches
as a result, deleted files stay pinned for a long time and space does not get removed. Sage Weil
01:35 PM Bug #9055 (Resolved): LibRadosTwoPoolsPP.HitSetWrite (and others) fail on remove of whiteout
Sage Weil
01:30 PM Bug #9052 (Resolved): ceph-mon crashes with *** Caught signal (Floating point exception) **
Sage Weil
12:38 PM CephFS Feature #9029 (Resolved): min/max uid for snapshot creation
Sage Weil
11:59 AM Bug #9102 (Resolved): ceph-disk has undefined variables
We fail to track them because the build doesn't yell at us, in the meantime, those should be fixed.... Alfredo Deza
10:46 AM Bug #9096 (Resolved): OSD::require_same_peer_instance fails to acquire lock
Sage Weil
10:23 AM Bug #9096 (Fix Under Review): OSD::require_same_peer_instance fails to acquire lock
https://github.com/ceph/ceph/pull/2249 Samuel Just
03:38 AM Bug #9096: OSD::require_same_peer_instance fails to acquire lock
It is the cause of http://tracker.ceph.com/issues/9074 Loïc Dachary
03:37 AM Bug #9096 (Resolved): OSD::require_same_peer_instance fails to acquire lock
It can be reproduced by running a few times (less than 5) *qa/workunits/cephtool/test.sh -t mon_osd*. It will eventua... Loïc Dachary
10:41 AM Revision e86fdefe (ceph): rgw: move generic server usage after all options
Since generic_server_usage() exits after printing the options, any other
help options should be printed before that. ...
Abhishek Lekshmanan
10:33 AM Bug #9082 (Resolved): Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before ...
Sage Weil
09:11 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
i've pushed wip-9082-firefly... can you please try this and see if it avoids the crash? i was looking for a divide b... Sage Weil
08:34 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
Hello Sage
Thanks for your time checking this bug. As required i have found some PG’s and 3 OSDs which are making...
karan singh
08:24 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
Hello Sage
As i have found some PG / OSD that make agent_choose_mode() unhappy. I am attaching logs of 2 differen...
karan singh
09:22 AM Feature #9097 (New): request for tools/commands to see hits/misses on cache pools
request for tools/commands to see hits/misses on cache pools Sheldon Mustard
07:23 AM Bug #9085 (Resolved): erasure-code: ISA plugin does not load
The isa plugin "wip-firefly-isa":https://github.com/ceph/ceph/tree/wip-firefly-isa does not have the bug. It was intr... Loïc Dachary
04:22 AM Revision 0479db8c (ceph): Merge pull request #2233 from majianpeng/fix1
os/chain_xattr: Remove all old xattr entry when overwrite the xattr.
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:17 AM Revision 3da1c711 (ceph): Merge pull request #2230 from ceph/wip-fsx-flatten
test_librbd_fsx: also flatten as part of randomize_parent_overlap
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:15 AM Revision 39bcafa6 (ceph): Merge pull request #2231 from ceph/wip-8944-firefly
Wip 8944 firefly
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:14 AM Revision 3bd1b331 (ceph): Merge pull request #2234 from kevincox/net-docs
doc: Initial network docs.
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:06 AM Revision b06d1b0a (ceph): Merge pull request #2237 from ceph/wip-8560
mon: instrument paxos
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Sage Weil
04:05 AM Revision b09b8563 (ceph): mon/Paxos: add perfcounters for most paxos operations
I'm focusing primarily on the ones that result in IO here.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
04:01 AM Revision 128aaa50 (ceph): Merge pull request #2242 from majianpeng/fix4
utime: Because class Clock didn't exist,so remove the declaration in class utime_t
Reviewed-by: Sage Weil <sage@redh...
Sage Weil
03:39 AM devops Bug #9074 (Duplicate): gitbuilder: make check does not complete, sometimes
It happens because of http://tracker.ceph.com/issues/9096 Loïc Dachary
01:57 AM devops Bug #9074: gitbuilder: make check does not complete, sometimes
Wrong diagnostic, the error is not from here. It loops while waiting for osds to come back up "a few lines below":htt... Loïc Dachary
01:02 AM devops Bug #9074: gitbuilder: make check does not complete, sometimes
"test.sh":https://github.com/ceph/ceph/blob/ea731ae14216bb479eff1f86ed6bd4a7cb71fb56/qa/workunits/cephtool/test.sh fa... Loïc Dachary
03:17 AM rbd Bug #9078: Removing an RBD is very slow whenever there is write's in other RBD which also belongs...
RBD's are created with different order parameter
Ramakrishnan P
02:00 AM rbd Bug #9078: Removing an RBD is very slow whenever there is write's in other RBD which also belongs...
setup is not available, unable to check "ceph -w", below are information based on IO tool(fio)
before rbd remove: io...
Ramakrishnan P
12:27 AM Bug #9077: Cluster is up in MON node even if Ceph is uninstalled in OSD node
Mon logs and dmesg logs of mon node are attached Ramakrishnan P
12:25 AM Revision f80ed26d (ceph): ceph-disk: move fs mount probe into a helper
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:25 AM Revision 6c77f5f2 (ceph): ceph-disk: use partition type UUIDs, and blkid
Use blkid to give us the GPT partition type. This lets us distinguish
between dmcrypt and non-dmcrypt partitions. F...
Sage Weil
12:14 AM rbd Bug #9075: Can't create a version 2 images on RHEL 7
Ok will do :). Sébastien Han

08/12/2014

11:34 PM Revision ea90d9f4 (ceph): Revert "ReplicatedPG: do not pass cop into C_Copyfrom"
The ref was introduced in 589b639af7c8834a1e6293d58d77a9c440107bc3
and is actually necessary to keep the buffers arou...
Samuel Just
11:21 PM Revision 300b5e8e (ceph): ReplicatedPG: do not pass cop into C_Copyfrom
We do not know when the objecter will finally let go of this Context. Thus, we
cannot know whether it will happen be...
Samuel Just
10:51 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
I can't reproduce.
From your messages, i can't find any error info.
Or am i missing something?
jianpeng ma
10:28 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Yeah I'm using that commmand.
Sorry - messed up the commit hash : 4eb18dd487da4cb621dcbecfc475fc0871b356ac
Mark Kirkwood
10:23 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Are you using this command "ceph-osd --id 0 --mkjournal --mkfs --osd-data /data1/cephdata --osd-journal /dev/sdc1"?
...
jianpeng ma
10:10 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Building wip-9073. Hmmm still getting the invalid argument error and osd down. I'm guessing this means there are two ... Mark Kirkwood
09:01 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Very quick work! Will test... Mark Kirkwood
08:47 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Mark, I've pushed this as wip-9073.. can you please test?
Thanks, Jianpeng! Sorry I missed the pull request earlier!
Sage Weil
08:36 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Yes, I already found this bug. If journal use aio mode. The bug occur.
The https://github.com/ceph/ceph/pull/2185 c...
jianpeng ma
08:28 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
When you say that reverting fixes it, do you mean that it allows an OSD that was erroring out on start to then start,... Sage Weil
06:31 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
...or maybe the ::open() Mark Kirkwood
06:14 PM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
On a different machine instead of a hang I am reliably getting:
2014-08-13 12:50:28.253439 7ffc701bb8c0 -1 ** ERR...
Mark Kirkwood
01:40 AM Bug #9073: OSD with device/partition journals down after fresh deploy or upgrade to 0.83
A correctio - the 'stuck on a mutex' comment is completely wrong - sorry - I'd attached strace to the ceph-osd proces... Mark Kirkwood
10:50 PM Revision bb121e41 (ceph): Merge pull request #2246 from ceph/wip-9064
ReplicatedPG::maybe_handle_cache: do not skip promote for write_ordered
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
10:27 PM Revision 8595e9be (ceph): OSD: introduce require_self_aliveness(OpRequestRef&,epoch_t) function
Take the self-aliveness checks out of require_same_or_newer_map() and use
the new function for that and for require_u...
Greg Farnum
10:27 PM Revision 7da121d2 (ceph): OSD: add require_same_peer_inst(OpRequestRef&,OSDMap&) helper
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit e99acf9810976b1fc74b84ad289773af43be973f)
C...
Greg Farnum
10:25 PM Revision d0f2c489 (ceph): OSD: use OpRequestRef& for a few require_* functions
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit eb2f1ea2c33647934af878b504383829f5a198ce)
Greg Farnum
10:25 PM Revision 8d395f1c (ceph): OSD: introduce require_up_osd_peer() function for gating replica ops
This checks both that a Message originates from an OSD, and that the OSD
is up in the given map epoch.
We use it in h...
Greg Farnum
10:25 PM Revision 24aeca9c (ceph): ReplicatedPG::maybe_handle_cache: do not skip promote for write_ordered
We cannot redirect a RW ordered read.
Fixes: #9064
Introduced: 0ed3adc1e0a74bf9548d1d956aece11f019afee0
Signed-off-b...
Samuel Just
10:20 PM Revision 72c2474e (ceph): init-ceph: conditionally update after argparsing
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 44ed1885fb588aad7e4729373e06c593317...
Alfredo Deza
10:20 PM Revision 8d7fb538 (ceph): make ceph-disk use the new init flag for cluster
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit d0ccb1cbef86b8ccd8967af892e3582ee80...
Alfredo Deza
10:20 PM Revision 260a9c84 (ceph): allow passing a --cluster flag to the init script
It will fallback to looking at /etc/ceph/$cluster.conf
otherwise.
Signed-off-by: Alfredo Deza <alfredo.deza@inktank....
Alfredo Deza
10:20 PM Revision cb5ecdce (ceph): use cluster name when checking the host
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit bdf5f30010117c71a227855a47fe22ab55b...
Alfredo Deza
10:20 PM Revision 79452540 (ceph): use name in /var/lib/ceph paths
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 3fb25c28e8e3cc62bc541538cc11f5ed414...
Alfredo Deza
10:20 PM Revision 92755d10 (ceph): when starting an OSD, pass in the cluster name
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 9e08e3a7794ce007c80c8fdb56e96900eb5...
Alfredo Deza
10:18 PM Revision 272b26f5 (ceph): osd_types: s/stashed/rollback_info_completed and set on create
Originally, this flag indicated that the object had already been stashed and
that therefore recording subsequent chan...
Samuel Just
10:18 PM Revision 7b5bcb94 (ceph): osd: use appropriate json types instead of stream when dumping info
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 363496ad878c685008ccd87f26f121d62...
Joao Eduardo Luis
10:17 PM Revision daec3226 (ceph): osd: have 'tid' dumped as a json unsigned int where appropriate
Most of the places dumping it would dump it as a string. Others as an int.
Just a couple would dump it as an unsigned...
Joao Eduardo Luis
10:15 PM Revision 439bce34 (ceph): rgw: need to pass need_to_wait for throttle_data()
need_to_wait wasn't passed into processor->throttle_data(). This was
broken in fix for #8937.
CID 1229541: (PW.PA...
Yehuda Sadeh
10:14 PM Revision fc1a4b5f (ceph): rgw: Don't send error body when it's a HEAD request
The main 'operation' know not to send any body for HEAD requests.
However for errors, this was not the case, the for...
Sylvain Munaut
10:13 PM Revision e3bc1534 (ceph): osd: prevent old clients from using tiered pools
If the client is old and doesn't understand tiering, don't let them use a
tiered pool. Reply with EOPNOTSUPP.
Fixes...
Sage Weil
10:12 PM Revision f3113e9a (ceph): mon/OSDMonitor: warn when cache pools do not have hit_sets configured
Give users a clue when cache pools are enabled but the hit_set is not
configured. Note that technically this will wo...
Sage Weil
10:03 PM Revision aaf732f7 (ceph): osd/ReplicatedPG: improve agent_choose_mode args
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit caf554bf2f752b50670d118c99843b7c4a2967b4)
Sage Weil
10:03 PM Revision 76a519b2 (ceph): osd/ReplicatedPG: evict blindly if there is no hit_set
If there is no hit set for a PG, blindly evict objects. This avoids an
assert(hit_set) in agent_estimate_atime_temp(...
Sage Weil
10:03 PM Revision 8ff48d79 (ceph): osd/ReplicatedPG: check agent_mode if agent is enabled but hit_sets aren't
It is probably not a good idea to try to run the tiering agent without a
hit_set to inform its actions, but it is tec...
Sage Weil
10:02 PM Revision e7c5fb58 (ceph): common/LogClient: fix sending dup log items
We need to skip even the most recently sent item in order to get to the
ones we haven't sent yet.
Fixes: #9080
Backp...
Sage Weil
10:01 PM Revision f98a6194 (ceph): common/LogClient: fix sending dup log items
We need to skip even the most recently sent item in order to get to the
ones we haven't sent yet.
Fixes: #9080
Backp...
Sage Weil
10:01 PM Revision bbb1e6a3 (ceph): RadosClient: Fixing potential lock leaks.
In lookup_pool and pool_delete, a lock is taken
before invoking wait_for_osdmap, but is not
released for the failure ...
Pavan Rallabhandi
09:58 PM Revision cbc9218e (ceph): librbd: fix error path cleanup for opening an image
If the image doesn't exist and caching is enabled, the ObjectCacher
was not being shutdown, and the ImageCtx was leak...
Josh Durgin
09:57 PM Revision e767254c (ceph): librbd: fix error path cleanup for opening an image
If the image doesn't exist and caching is enabled, the ObjectCacher
was not being shutdown, and the ImageCtx was leak...
Josh Durgin
09:38 PM Revision dec028d1 (ceph): Merge remote-tracking branch 'gh/firefly-next' into firefly
Sage Weil
09:37 PM Revision 7bd67fa1 (ceph): Merge pull request #2245 from dachary/wip-9085-isa-link
erasure-code: isa plugin must link with ErasureCode.cc
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:26 PM Revision 3f18b02c (ceph): Added overrides for vps time outs
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
09:07 PM Revision 984f6148 (ceph): erasure-code: isa plugin must link with ErasureCode.cc
Otherwise it will not get the methods it needs. A test is added to check
the plugin loads as expected, from the comma...
Loïc Dachary
09:03 PM Feature #8560 (Pending Backport): mon: instrument paxos
Sage Weil
08:53 PM Revision 1088d6cd (ceph): ceph-disk: fix log syntax error
File "/usr/sbin/ceph-disk", line 303, in command_check_call
LOG.info('Running command: %s' % ' '.join(arguments))...
Sage Weil
07:41 PM Revision afa176e7 (ceph): Merge pull request #2239 from ceph/wip-8912
librbd: fix error path cleanup for opening an image
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
07:36 PM Revision 41e44611 (ceph): doc/changelog: v0.67.10 notes
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:22 PM Revision ce7cd60a (ceph): Merge branch 'wip-8860'
Reviewed-by: Sage Weil <sage@redhat.com> Sage Weil
07:22 PM Revision 44ed1885 (ceph): init-ceph: conditionally update after argparsing
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
06:30 PM Revision 350b5124 (ceph): doc/release-notes: v0.67.10
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:27 PM Bug #8886: Miss some folders in PG's folder
I see. Thank you for your reply~ Jingjing Zhao
01:43 PM Bug #8886 (Closed): Miss some folders in PG's folder
./default.4281.322\u\ushadow\u.Ndfi3nAmRHjph\uXyzjJQutltgGi1Dkd\u1__head_17F630A2__1b_ffffffffffffffff_7
appears t...
Samuel Just
06:19 PM Revision 7998c322 (ceph): rgw: call throttle_data() even if renew_state() failed
Otherwise we're going to leak the aio callback handle.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh
06:18 PM Bug #9067 (Resolved): (wip-objecter) Objecter assertion in SIGINT handler
... John Spray
04:43 PM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
Samuel Just
04:20 PM Bug #8894 (Resolved): osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
Samuel Just
12:19 PM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
Samuel Just
12:19 PM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
wip-9054 Samuel Just
11:25 AM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
I think it's the C_Copyfrom which we gave the objecter in _copy_some. It's got a CopyOpRef. Samuel Just
04:35 PM Revision e3e48830 (ceph): Merge remote-tracking branch 'gh/next'
Conflicts:
PendingReleaseNotes
Sage Weil
04:34 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
This sounds right to me! Sage Weil
03:58 PM Bug #9082 (Need More Info): Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub b...
Sage Weil
10:57 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
I have injected debus osd 20 to one OSD , and and then tried to initiate rados bench on EC pool which is tired with c... karan singh
09:31 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...
can you reproduce this with debug osd = 20 and attach teh log? thanks! Sage Weil
08:27 AM Bug #9082: Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before tier agent ...

I have sent one email to ceph mailing list today , which is related to problem with Ceph pool . ...
karan singh
07:59 AM Bug #9082 (Resolved): Ceph Firefly 0.80.5 : PG has invalid (post-split) stats; must scrub before ...
Hello
Ceph version : 0.80.5
Centos 6.5
Features in use : erasure coding and cache tiering
Few hours back m...
karan singh
03:48 PM Bug #9064 (Resolved): RadosModel assertion failure
Samuel Just
03:48 PM Bug #9064 (Pending Backport): RadosModel assertion failure
Sage Weil
03:26 PM Bug #9064: RadosModel assertion failure
Samuel Just
03:26 PM Bug #9064: RadosModel assertion failure
wip-9064 Samuel Just
03:25 PM Bug #9064: RadosModel assertion failure
Got it: 0ed3adc1e0a74bf9548d1d956aece11f019afee0
We're redirecting RW ordered reads due to the second read promote...
Samuel Just
02:00 PM Bug #9064: RadosModel assertion failure

I've now seen this in a case where the client wasn't in the process of handling a new OSD map (but the server was),...
John Spray
05:17 AM Bug #9064: RadosModel assertion failure
This just reproduced on master 78dc4df, so looks like it's not wip-objecter specific. John Spray
03:24 PM Messengers Bug #8880 (Resolved): msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq featu...
Sage Weil
03:18 PM Bug #8860 (Resolved): ceph-disk issues with custom cluster name
Sage Weil
12:21 PM Bug #8860 (Pending Backport): ceph-disk issues with custom cluster name
Sage Weil
03:16 PM Bug #8625 (Resolved): EC pool - OSD creates an empty file for op with 'create 0~0, writefull 0~xx...
Sage Weil
03:11 PM rgw Bug #8539 (Resolved): civetweb backend responds with a body when a HEAD request yields an error
Sage Weil
03:08 PM Revision 7e7afafe (ceph): Merge pull request #2243 from ceph/wip-log-client
common/LogClient: fix sending dup log items
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
03:02 PM Bug #8982 (Resolved): cache pool osds crashing when data is evicting to underlying storage pool
Sage Weil
03:02 PM Bug #8714 (Resolved): we do not block old clients from breaking cache pools
Sage Weil
03:01 PM Bug #8944 (Resolved): Ceph daemon bad asok used in connection with cluster
Sage Weil
02:59 PM Bug #9080 (Resolved): LogClient: sends dup messages, misses some
Sage Weil
01:15 PM Bug #9080 (Pending Backport): LogClient: sends dup messages, misses some
Sage Weil
07:02 AM Bug #9080 (Resolved): LogClient: sends dup messages, misses some
noticed where where 'ceph -s' woudln't show the most recent log message. tracing things it turns out that it was alw... Sage Weil
02:58 PM Bug #9022 (Resolved): Potential lock leaks in RadosClient
Sage Weil
02:57 PM Bug #7999 (Resolved): osd: pgs share info that hasn't been persisted
Sage Weil
02:57 PM rgw Bug #8169 (Resolved): rgw: swift user manifest does not compute etag
Sage Weil
02:56 PM rgw Bug #8269 (Resolved): rgw: corrupted multipart object
Sage Weil
02:56 PM Bug #8438 (Resolved): erasure code: object are not cleanup
Sage Weil
02:56 PM rgw Bug #8442 (Resolved): rgw: does not detect/adapt to erasure pool stripe size
Sage Weil
02:56 PM rgw Bug #8586 (Resolved): Missing Swift API Header causes RadosGW to segfault
Sage Weil
02:55 PM rbd Bug #8912 (Resolved): librbd segfaults when creating new image (rbd-ephemeral-clone-stable-icehouse)
Sage Weil
12:38 PM rbd Bug #8912 (Pending Backport): librbd segfaults when creating new image (rbd-ephemeral-clone-stabl...
Sage Weil
02:54 PM Bug #8670 (Resolved): Cache tiering parameters can not be displayed for a pool
Sage Weil
02:48 PM Bug #8696 (Resolved): mon: 'osd pool set' must take into account pool's nature when setting some ...
Sage Weil
02:48 PM Bug #8701 (Resolved): osd: scrub found obsolete rollback obj
Sage Weil
02:47 PM rgw Bug #8702 (Resolved): RadosGW incorrectly converting + to space in URLs
Sage Weil
02:46 PM Bug #8733 (Resolved): OSD crashed at void ECBackend::handle_sub_read
Sage Weil
02:39 PM Bug #8882 (Resolved): osd: osd tier remove ... leaves incomplete clones behind, confusing scrub
Sage Weil
02:39 PM Bug #8889 (Resolved): osd/ReplicatedPG.cc: 5162: FAILED assert(got)
Sage Weil
02:38 PM rbd Bug #8920 (Resolved): rbd/singleton/{all/formatted-output.yaml} fails on trusty due to whitespace
Sage Weil
02:38 PM rgw Bug #8928 (Resolved): rgw: bad object created if stripe size is not a multiple of chunk size
Sage Weil
02:38 PM Bug #8931 (Resolved): failed write reply order from ceph_test_rados
Sage Weil
02:37 PM rgw Bug #8937 (Resolved): rgw: broken large(-ish) objects
Sage Weil
02:37 PM Bug #8943 (Resolved): "ceph df" cannot show pool available space correctly
Sage Weil
02:37 PM Bug #8969 (Resolved): PerfCounters.SinglePerfCounters failure on i386
Sage Weil
02:37 PM rgw Bug #8972 (Resolved): rgw: bucket index log wrong object name in multipart completion
Sage Weil
02:34 PM Bug #9085 (Pending Backport): erasure-code: ISA plugin does not load
Sage Weil
09:46 AM Bug #9085 (Fix Under Review): erasure-code: ISA plugin does not load
"need review":https://github.com/ceph/ceph/pull/2245 Loïc Dachary
09:20 AM Bug #9085 (Resolved): erasure-code: ISA plugin does not load
Because the plugin was not compiled with ErasureCode.cc Loïc Dachary
02:07 PM devops Bug #8160 (Duplicate): multipath-tools does not co-exist with ceph
If/when we implement multipath support in ceph-deploy, this should be resolved. Ian Colle
02:05 PM Revision 057c6808 (ceph): common/LogClient: fix sending dup log items
We need to skip even the most recently sent item in order to get to the
ones we haven't sent yet.
Fixes: #9080
Backp...
Sage Weil
01:53 PM Revision 07e1ceb5 (ceph): erasure-code: ErasureCodeJerasure::create_ruleset must return a ruleset
CrushWrapper::add_simple_ruleset does not return a ruleset, it returns
a ruleid that must be converted into a ruleset...
Loïc Dachary
01:53 PM Revision 96047f1d (ceph): erasure-code: OSDMonitor::crush_ruleset_create_erasure needs ruleset
When OSDMonitor::crush_ruleset_create_erasure checks the ruleset for
existence, it must convert the ruleid into a rul...
Loïc Dachary
01:43 PM rgw Bug #9089 (Resolved): rgw: copy_obj_data() does not stripe target object
copy_obj_data() is as it is now a reminiscent of a very old architecture. It should be modified to create a striped o... Yehuda Sadeh
01:36 PM Bug #8591 (Resolved): ceph-disk incorrectly colocates journal when using dm-crypt
wip-ceph-disk Sage Weil
01:35 PM Bug #8922: ceph-deploy mon create fails to create additional monitoring nodes.
does 'hostname' on those machines return the same string, or does it include a domain name, or somethign different? Sage Weil
01:34 PM Bug #8985: "[WRN] map e9 wrongly marked me down" in upgrade:dumpling-x-firefly---basic-vps suite
change the vps.yaml timeout to 90 seconds instead of 40.. these should go away then Sage Weil
01:33 PM Bug #8986 (Duplicate): "[WRN] map e62 wrongly marked me down" in upgrade:dumpling-x-firefly---bas...
Samuel Just
01:33 PM Bug #9012 (Duplicate): "[WRN] map e277 wrongly marked me down" in upgrade:dumpling-x-firefly---ba...
Samuel Just
01:32 PM Bug #9011 (Duplicate): osd memory leaks on next
#9023 Sage Weil
01:27 PM devops Bug #9061 (Resolved): dumpling to firefly upgrade on RH6 restarts the daemons
Sage Weil
01:26 PM Bug #8974 (Need More Info): osd crashed with merge_log assert due to removal of isds
Samuel Just
01:25 PM Bug #8974: osd crashed with merge_log assert due to removal of isds
We can probably make some progress if you reproduce with
debug ms = 1
debug osd = 20
debug filestore = 20
on ...
Samuel Just
01:14 PM Bug #8505 (Resolved): OSD osd/OSD.cc: 6222: FAILED assert(p->second.empty())
Samuel Just
01:13 PM Bug #8691 (Resolved): osd: PG::_lock, OSD::pg_map_lock lock cycle
Samuel Just
01:10 PM Bug #8939 (Duplicate): stalled LibRadosTwoPoolsPP.TryFlushReadRace; client failed to reconnect?
#8891 Sage Weil
01:09 PM Bug #8940 (Duplicate): 3.22s1 shard 0(2) missing ad166f62/benchmark_data_plana57_30491_object1036...
Sage Weil
01:06 PM Bug #9069 (Resolved): rgw tests reported as failed in teuthology-2014-08-11_10:35:04-upgrade:dump...
Sage Weil
01:00 PM Revision 95fbce22 (ceph): Merge pull request #2232 from dachary/wip-9044-use-ruleset
erasure-code: ErasureCodeJerasure::create_ruleset must return a ruleset
Reviewed-by: Xiaoxi Chen <xiaoxi.chen@intel....
Loïc Dachary
12:43 PM rgw Bug #8784: rgw: completion leak
Note that all the failures are at the copy object across regions path. I did find a missing cleanup at the error hand... Yehuda Sadeh
10:53 AM Bug #9058: rest-api: long-running process may fail 'tell osd...' due to stale osdmap
ubuntu@teuthology:/a/teuthology-2014-08-10_02:30:01-rados-next-testing-basic-plana/412468 Sage Weil
10:16 AM Revision 71e827f0 (ceph): utime: Because class Clock didn't exist,so remove the declaration in cl...
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com> Ma Jianpeng
10:08 AM Bug #9087 (Can't reproduce): ceph_test_rados_list_parallel hang
... Sage Weil
09:09 AM rbd Bug #6631 (Need More Info): disabling writethrough until flush appears to disable RBD cache
Amit Vijairania wrote:
> More repetition of tests..
>
> // IOPS for Sequential 4KB Write _with_ "rbd cache writet...
Sage Weil
09:07 AM rbd Bug #9078 (Need More Info): Removing an RBD is very slow whenever there is write's in other RBD w...
it sounds like the cluster is just under heavy load. can you confirm how many ops ceph -w shows before and during th... Sage Weil
05:09 AM rbd Bug #9078 (Rejected): Removing an RBD is very slow whenever there is write's in other RBD which a...
Configuration:
3 node with mon and 3 node with OSD connected via Enclosure/jbod, total 15 OSD's
Steps followed:
...
Ramakrishnan P
09:07 AM Feature #9083 (Closed): Standalone script to generate Ceph keys
Goal: To allow 3rd party products which will be acting as Ceph clients to be able to install & configure all Ceph-cli... Neil Levine
09:04 AM Bug #9077 (Need More Info): Cluster is up in MON node even if Ceph is uninstalled in OSD node
can you turn up mon logging (if it isn't up already) and attach teh log from the leader? tehse should get marked dow... Sage Weil
04:49 AM Bug #9077 (Can't reproduce): Cluster is up in MON node even if Ceph is uninstalled in OSD node
Configuration:
1 mon and 1 osd node, number of OSD's 7
Steps followed:
1. Make Cluster up in single node and e...
Ramakrishnan P
09:00 AM rbd Bug #8845 (Resolved): Flattening Clones of clone, results in command failure
Josh Durgin
09:00 AM rbd Bug #9075 (Need More Info): Can't create a version 2 images on RHEL 7
can you retry with the ceph.com package? the 0.81 from fedora is all kinds of busted. Sage Weil
02:45 AM rbd Bug #9075 (Resolved): Can't create a version 2 images on RHEL 7
Hi,
I can't create version 2 images, version 1 works though.
# rbd create -s 10240 --image-format 2 lesebb
20...
Sébastien Han
08:56 AM Bug #8595 (In Progress): osd: client op blocks until backfill starts (dumpling)
with this patch, i see filestore tripping over ENOENT on clone:
ubuntu@teuthology:/a/teuthology-2014-08-11_19:00:0...
Sage Weil
07:35 AM rgw Bug #9002: Creating swift key with --gen-secret in separate step from subuser creation fails
have meet on Wheezy and Ubuntu with Ceph0.80.5 too.
it can be sucessful when use :
radosgw-admin user create --su...
only debian
07:31 AM CephFS Bug #9056: fuse kmod + ceph-fuse triggers "BUG: sleeping function called from invalid context"
... John Spray
06:51 AM CephFS Bug #9056 (Resolved): fuse kmod + ceph-fuse triggers "BUG: sleeping function called from invalid ...
Sage Weil
05:10 AM CephFS Bug #9056: fuse kmod + ceph-fuse triggers "BUG: sleeping function called from invalid context"
This is supposed to be fixed upstream in v3.16-rc6 by commit c55a01d360af, will close this when we've seen a clean fs... John Spray
07:20 AM Bug #9044: erasure-code: use ruleset instead of ruleid
"backport to firefly":https://github.com/ceph/ceph/pull/2244 Loïc Dachary
05:58 AM Bug #9044 (Pending Backport): erasure-code: use ruleset instead of ruleid
Loïc Dachary
05:57 AM Bug #9044 (Resolved): erasure-code: use ruleset instead of ruleid
Loïc Dachary
05:55 AM Bug #9044: erasure-code: use ruleset instead of ruleid
Works. The problems of this run are
* "unrelated MDS decode bug":http://pulpito.ceph.com/loic-2014-08-12_10:00:07-...
Loïc Dachary
12:58 AM Bug #9044: erasure-code: use ruleset instead of ruleid
"scheduled upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-12_10:00:07-upgrade:firefly-x:stress-... Loïc Dachary
06:56 AM CephFS Bug #8648: Standby MDS leaks memory over time
Any change you can run one of these in standby under massif for a while? that will tell us what is leaking! Sage Weil
06:55 AM CephFS Bug #8651 (Won't Fix): crashing mds in an active-active mds setup
this MDS got blacklisted. there is an open issues somewhere to make the shutdown more friendly, but the behavior is ... Sage Weil
06:52 AM Bug #9023: valgrind failures in OSD
The leaks in the init stuff seem likely also to be present on master John Spray
06:50 AM CephFS Bug #8725: mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
we probably have to do a reencoding trick like we do in MOSDMap? Sage Weil
06:48 AM CephFS Bug #8876 (Resolved): kcephfs: hang on read of length 0
Sage Weil
06:22 AM Bug #9079 (Resolved): osd: bad learned_addr during send_boot
... Sage Weil
06:10 AM Bug #8520: osd: segv in PushOp::print()
... Sage Weil
03:27 AM rbd Bug #8385: RBD / QEMU Crash: Invalid fastbin entry (free)
Any interest in a lookalike bug from Cuttlefish?
/lib/x86_64-linux-gnu/libc.so.6(+0x7e566)[0x7f7cd15ad566]
/usr/...
Andrey Korolyov
02:55 AM rbd Bug #9076 (Resolved): Can't completely remove a version 1 image on RHEL 7
I can create version 1 image, however the deletion is not complete.
# rbd create -s 10240 --image-format 1 leseb
...
Sébastien Han
01:16 AM Revision 78dc4df1 (ceph): doc: Replace [default] with [global].
Fixes: #8955
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
01:05 AM Revision e5324edb (ceph): doc: Added yum-priorities procedure to manual install.
Fixes: #8734
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
01:04 AM Revision a01252a4 (ceph): doc: Added priority = 2 to packages.
Fixes: #8734
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
12:55 AM Revision c24c4eed (ceph): ceph_manager: longer osd restart timeout
We bumped this on master and dumpling already
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
12:54 AM devops Bug #9074: gitbuilder: make check does not complete, sometimes
"re-run the build to check if it fails always or sometimes":http://gitbuilder.sepia.ceph.com/gitbuilder-ceph-tarball-... Loïc Dachary
12:52 AM devops Bug #9074 (Duplicate): gitbuilder: make check does not complete, sometimes
It looks like i386 build fails because a timeout interrupts it before it gets a chance to complete.
It could be t...
Loïc Dachary

08/11/2014

11:43 PM Revision 3dfa72d5 (ceph): librbd: fix error path cleanup for opening an image
If the image doesn't exist and caching is enabled, the ObjectCacher
was not being shutdown, and the ImageCtx was leak...
Josh Durgin
10:58 PM Revision 2edf01ff (ceph): Revert "Fix for bug #6700"
This reverts commit 673394702b725ff3f26d13b54d909208daa56d89.
This appears to break things when the journal and data...
Sage Weil
10:57 PM Revision d6e6ba19 (ceph): ceph-disk: fix verify_no_in_use check
We only need to verify that partitions aren't in use when we want to
consume the whole device (osd data), not when we...
Sage Weil
09:15 PM Bug #9073 (Resolved): OSD with device/partition journals down after fresh deploy or upgrade to 0.83
Using a src build (and the packages built from it) on Ubuntu 14.04 x86_64. Ceph version is 0.83-399-gf77449c.
In ...
Mark Kirkwood
08:53 PM rbd Bug #9071 (Duplicate): mkfs.ext4 stuck in D state on RBD with kernel client
This is a bug in 3.15; it is not present in 3.14. The fix will make it into the next stable 3.15 release soon.
Sage Weil
07:32 PM rbd Bug #9071: mkfs.ext4 stuck in D state on RBD with kernel client
Please, mark this issue as duplicate of http://tracker.ceph.com/issues/8818 Ivan Mironov
06:06 PM rbd Bug #9071: mkfs.ext4 stuck in D state on RBD with kernel client
Reproducible on all my ceph hosts (all with the same kernel), with any image format (1 or 2). But only with mkfs.ext4... Ivan Mironov
05:47 PM rbd Bug #9071 (Duplicate): mkfs.ext4 stuck in D state on RBD with kernel client
I tried to create ext4 on newly created and mapped RBD image, but mkfs.ext4 stuck:
# mkfs.ext4 /dev/rbd/docker.rbd...
Ivan Mironov
07:36 PM Revision 0a2b4c25 (ceph): rgw: Don't send error body when it's a HEAD request
The main 'operation' know not to send any body for HEAD requests.
However for errors, this was not the case, the for...
Sylvain Munaut
07:30 PM Revision 616cb927 (ceph): Merge pull request #2214 from ceph/wip-rados-df-json
rados: fix 'rados df --format=json' field names
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
06:53 PM Revision 5663f91b (ceph): rados: fix {read,write}_bytes value
Fill this in with kb * 1024 instead of the ops count.
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@red...
Sage Weil
06:53 PM Revision b0f5ba95 (ceph): rados: remove {read,write}_kb fields from rados df json output
Redundant. Do not backport.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
06:15 PM Documentation #8955 (Resolved): doc refers to [default] section, don't think it exists
's/[default]/[global]/' John Wilkins
06:10 PM Documentation #8955 (In Progress): doc refers to [default] section, don't think it exists
John Wilkins
06:05 PM devops Bug #8734 (Resolved): EPEL / Ceph.com package priority issues
I added priorty=2 to the get packages document example for ceph.repo. I also added an install yum-priorities series o... John Wilkins
05:56 PM devops Bug #8734 (In Progress): EPEL / Ceph.com package priority issues
John Wilkins
05:51 PM Bug #9072: error setting 'mon_pg_warn_min_objects' to '10K': (22) Invalid argument
ubuntu@teuthology:/a/sage-2014-08-10_18:40:12-rados-firefly-next-distro-basic-multi/414556 Sage Weil
05:50 PM Bug #9072 (Resolved): error setting 'mon_pg_warn_min_objects' to '10K': (22) Invalid argument
... Sage Weil
05:25 PM Revision c2e2fc56 (ceph): Merge pull request #2238 from ceph/wip-9057
msg/Pipe: do not wait for self in Pipe::stop_and_wait()
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
05:25 PM Bug #9069: rgw tests reported as failed in teuthology-2014-08-11_10:35:04-upgrade:dumpling:rgw-du...
oh.. it' snot running as root.. or with daemon-helper. Sage Weil
05:24 PM Bug #9069: rgw tests reported as failed in teuthology-2014-08-11_10:35:04-upgrade:dumpling:rgw-du...
7585 ? Sl 0:05 radosgw -n client.0 -k /etc/ceph/ceph.client.0.keyring --rgw-socket-path /home/ubuntu/ceph... Sage Weil
03:57 PM Bug #9069 (Resolved): rgw tests reported as failed in teuthology-2014-08-11_10:35:04-upgrade:dump...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-11_12:05:02-upgrade:dumpling-dumpling---basic-vps/... Yuri Weinstein
04:58 PM rbd Bug #8912 (Fix Under Review): librbd segfaults when creating new image (rbd-ephemeral-clone-stabl...
https://github.com/ceph/ceph/pull/2239 Josh Durgin
01:39 PM rbd Bug #8912: librbd segfaults when creating new image (rbd-ephemeral-clone-stable-icehouse)
Looks like it was a race condition in a previously little-used error path. Josh Durgin
01:04 PM rbd Bug #8912 (In Progress): librbd segfaults when creating new image (rbd-ephemeral-clone-stable-ice...
Excellent report, your reproducer causes the same crash for me. Josh Durgin
04:14 PM Bug #9044: erasure-code: use ruleset instead of ruleid
gitbuilder is running Loïc Dachary
03:42 PM Revision b5dafe1c (ceph): osd: allow io priority to be set for the disk_tp
The disk_tp covers scrubbing, pg deletion, and snap trimming
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry pic...
Sage Weil
03:32 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
wip-9054 Samuel Just
03:07 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
When we go to flush clone 22, all we know is that 22 is dirty, has snaps
[21], and 4 is clean. As part of fl...
Samuel Just
02:24 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
Ok, we start with the following configuration in the cache (all dirty):
30:[29,21,20,15,10,4]:[22(21), 15(15,10), ...
Samuel Just
12:45 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
Actually, looks like this might already be handled correctly, re-consulting the log. Samuel Just
12:00 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
Thinking Samuel Just
11:52 AM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
Hmm, I think the bug is like this:
Normally, if we get the following op sequence:
- write 1:[]
- delete 10:[3] (...
Samuel Just
03:16 PM Bug #9040: clients can SEGV during package upgrade
I see no segmentation errors in the latest run: /a/teuthology-2014-08-11_12:05:02-upgrade:dumpling-dumpling---basic-v... Yuri Weinstein
12:34 PM rgw Bug #8539: civetweb backend responds with a body when a HEAD request yields an error
Merged, commit:0a2b4c25541bbd15776d3d35986518e37166910f Yehuda Sadeh
12:34 PM rgw Bug #8539 (Pending Backport): civetweb backend responds with a body when a HEAD request yields an...
Yehuda Sadeh
12:30 PM Revision 9d446bd4 (ceph): 0.67.10
Jenkins
12:24 PM Bug #9064: RadosModel assertion failure

The bug is happening when a new OSD map is received in the middle of the series of transactions. The read transact...
John Spray
11:40 AM Bug #9064: RadosModel assertion failure
Got an even more specific backtrace ... John Spray
09:57 AM Bug #9064: RadosModel assertion failure
trying to reproduce locally with objecter logging turned up and ``ms inject socket failures`` enabled as it is in the... John Spray
09:29 AM Bug #9064: RadosModel assertion failure
I understand this a little better now: the operations in this WriteOp are 1,2 (writes), 4 (setxattr), 5 (read). So t... John Spray
07:23 AM Bug #9064: RadosModel assertion failure
http://qa-proxy.ceph.com/teuthology/john-2014-08-10_02:14:59-rados-wip-mds-contexts-testing-basic-plana/411119/teutho... John Spray
07:22 AM Bug #9064 (Resolved): RadosModel assertion failure

http://qa-proxy.ceph.com/teuthology/john-2014-08-10_02:14:59-rados-wip-mds-contexts-testing-basic-plana/411119/teut...
John Spray
10:41 AM Bug #9057 (Resolved): mark_down from fast dispatch can deadlock
Sage Weil
09:57 AM rgw Subtask #9068 (Closed): rgw: add rgw setup to vstart
As part of the development documentation we need to update vstart to create an RadosGW development environment. Luis Pabon
09:53 AM Bug #9067 (Resolved): (wip-objecter) Objecter assertion in SIGINT handler

@ wip-mds-contexts 2550fc51f30a8a1e581dd9a90511732a3b70ad2a
When I start a "ceph status" while no mon is running...
John Spray
09:01 AM devops Bug #9066 (Rejected): Need ceph-deploy to be able to run to JUST generate ceph.conf and keyring w...
Mirror of issue: https://bugzilla.redhat.com/show_bug.cgi?id=1127852 Alfredo Deza
08:37 AM Bug #9065 (Resolved): LibRados* tests failed in upgrade:dumpling-x-firefly---basic-vps
This should be fixed by https://github.com/ceph/ceph/pull/2236 (in review)
Logs are in http://qa-proxy.ceph.com/te...
Yuri Weinstein
08:33 AM devops Bug #9032 (Rejected): ceph-deploy over proxy
The `--gpg-url` is only valid if you are pointing to a custom repo.
What you need to do is create a custom repo se...
Alfredo Deza
08:28 AM Feature #8580: Decrease disk thread's IO priority and/or make it configurable
Hi,
The backport to dumpling is missing the commit which provides the new configurable: https://github.com/ceph/ce...
Dan van der Ster
05:04 AM Bug #9062: Mon segfault in waitlist_or_zap_client
Note that this was wip-mds-clients which doesn't have any messenger changes and doesn't have any mon changes other th... John Spray
05:01 AM Bug #9062 (Resolved): Mon segfault in waitlist_or_zap_client

http://pulpito.front.sepia.ceph.com/john-2014-08-10_02:14:59-rados-wip-mds-contexts-testing-basic-plana/411054/
...
John Spray
04:37 AM Bug #9023: valgrind failures in OSD

Haven't seen the "new Session" one since rebasing on master, so I'm optimistic that it was the same thing as the le...
John Spray
04:09 AM CephFS Bug #8878 (In Progress): mds lock cycle (wip-objecter)
I think all these are OK now in wip-mds-contexts: remaining failures on that branch are all outside MDS. John Spray
04:09 AM Bug #9009 (Resolved): (wip-objecter) ObjectCacher assert in fs client
This is all good now in wip-mds-contexts (http://pulpito.ceph.com/john-2014-08-09_14:56:53-fs-wip-mds-contexts-testin... John Spray
03:22 AM Revision 8180713c (ceph): msg/Pipe: do not wait for self in Pipe::stop_and_wait()
The fast dispatch code necessitated adding a wait for the fast dispatch
to complete when taking over sockets back in ...
Sage Weil

08/10/2014

11:43 PM devops Bug #9061 (Resolved): dumpling to firefly upgrade on RH6 restarts the daemons
Hi,
When I upgrade the RPMs on a RH6 server from 0.67.9 to 0.80.5, the daemons are (cond)restarted. I believe these ...
Dan van der Ster
09:09 PM Revision fd421b26 (ceph): mon/MonitorDBStore: add get_{keys,bytes}() accounting to Transaction
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:48 PM Revision d7fb7bf5 (ceph): ceph_test_rados_api_tier: fix cache cleanup (ec too)
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:48 PM Revision ebbe8aab (ceph): ceph_test_rados_api: fix cleanup of cache pool
We can't simply try to delete everything in there because some items may
be whiteouts. Instead, flush+evict everythi...
Sage Weil
07:20 PM Linux kernel client Bug #8806: libceph: must use new tid when watch is resent
meanwhile, the MWatchNotify message now has a return value encoded at the end (s32) when header.version >= 0. See wi... Sage Weil
07:19 PM Linux kernel client Bug #8806: libceph: must use new tid when watch is resent
the bug is with the kernel client: it needs to use a new tid when resending the watch. this was partially fixed on t... Sage Weil
06:41 PM Revision 1d199fb1 (ceph): librados/TestCase: inheret cleanup_default_namespace
No need to duplicate this code.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:04 PM Bug #9057 (Fix Under Review): mark_down from fast dispatch can deadlock
https://github.com/ceph/ceph/pull/2238 Sage Weil
10:45 AM Bug #9057: mark_down from fast dispatch can deadlock
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-08-09_14:13:44-rados-next-testing-basic-multi/410713
3 (!...
Sage Weil
08:41 AM Bug #9057 (Resolved): mark_down from fast dispatch can deadlock
... Sage Weil
04:13 PM Feature #8639 (In Progress): mon: dispatch messages while blocked waiting for IO
Sage Weil
03:45 PM Bug #8620: rest/test.py occasional failure (dumpling)
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-08-10_13:22:17-rados-dumpling-distro-basic-multi/413788 Sage Weil
03:18 PM Revision 04a484a4 (ceph): erasure-code: OSDMonitor::crush_ruleset_create_erasure needs ruleset
When OSDMonitor::crush_ruleset_create_erasure checks the ruleset for
existence, it must convert the ruleid into a rul...
Loïc Dachary
03:18 PM Revision 0029a358 (ceph): erasure-code: ErasureCodeJerasure::create_ruleset must return a ruleset
CrushWrapper::add_simple_ruleset does not return a ruleset, it returns
a ruleid that must be converted into a ruleset...
Loïc Dachary
03:18 PM Revision 3f5d86af (ceph): erasure-code: ErasureCodeIsa::create_ruleset must return a ruleset
http://tracker.ceph.com/issues/9044 Refs: #9044
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
03:18 PM Revision 5808d6a6 (ceph): osd: improve ruleno/ruleset consistency
rename variables to reflect when they hold a ruleset and when they hold
a rule id / ruleno in the hope to avoid bugs ...
Loïc Dachary
02:07 PM Feature #8560 (Fix Under Review): mon: instrument paxos
Sage Weil
12:51 PM rgw Bug #8988 (Fix Under Review): AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
Two consecutive run with the increased timeout do not show the bug ("one":http://pulpito.ceph.com/loic-2014-08-10_15:... Loïc Dachary
02:03 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
In a few tickets it is suggested that this may be an idle timeout problem. I "rescheduled a suite":http://pulpito.cep... Loïc Dachary
01:31 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
In the attached file, each part separated with *-----------------------------* is the output between the last success... Loïc Dachary
01:09 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
The errors for each failure are different and suggests the tests are failing for an independent reason such as the cl... Loïc Dachary
01:03 AM rgw Bug #8988: AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
* http://pulpito.ceph.com/loic-2014-08-08_12:17:30-upgrade:firefly-x:stress-split-wip-9025-chunk-remapping-testing-ba... Loïc Dachary
12:46 PM Bug #9055 (Fix Under Review): LibRadosTwoPoolsPP.HitSetWrite (and others) fail on remove of whiteout
https://github.com/ceph/ceph/pull/2236 Sage Weil
11:05 AM Feature #9059 (Resolved): osd: store opportunistic whole-object checksum
when we deep scrub, we have a whole-object checksums that cover data and omap. store a copy in object_info_t, along ... Sage Weil
10:52 AM Bug #8935: operations not idempotent when enabling cache
sage-2014-08-09_14:13:44-rados-next-testing-basic-multi/410527 and 410528 Sage Weil
10:51 AM Bug #9058 (Can't reproduce): rest-api: long-running process may fail 'tell osd...' due to stale o...
sage-2014-08-09_14:13:44-rados-next-testing-basic-multi/410524 Sage Weil
10:48 AM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-08-09_14:13:44-rados-next-testing-basic-multi/410806
alwa...
Sage Weil
07:51 AM Revision 104440ce (ceph): Merge pull request #2201 from dachary/wip-8496-erasure-code-base-class
erasure code base class
Reviewed-by: Andreas-Joachim Peters <Andreas.Joachim.Peters@cern.ch>
Loïc Dachary
02:16 AM CephFS Bug #8725: mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
"same error":http://pulpito.ceph.com/loic-2014-08-10_09:59:49-upgrade:firefly-x:stress-split-wip-9025-chunk-remapping... Loïc Dachary
12:53 AM CephFS Bug #8725: mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
Another "similar crash":http://pulpito.ceph.com/loic-2014-08-08_12:17:30-upgrade:firefly-x:stress-split-wip-9025-chun... Loïc Dachary
12:39 AM CephFS Bug #8725: mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
And the same trace at "upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-08_12:13:20-upgrade:firef... Loïc Dachary
12:33 AM CephFS Bug #8725: mds crashed in upgrade:dumpling-x:stress-split-master-testing-basic-plana
Looks like a similar problem at "upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-08_12:13:20-upg... Loïc Dachary
01:04 AM Feature #9025: erasure-code: chunk remapping
The upgrade suite from firefly had one error related to an independant "MDS problem":http://pulpito.ceph.com/loic-201... Loïc Dachary
12:49 AM Feature #8496 (Resolved): erasure-code: ErasureCode base class
Loïc Dachary
12:41 AM Feature #8496: erasure-code: ErasureCode base class
The "upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-08_12:13:20-upgrade:firefly-x:stress-split-... Loïc Dachary
12:16 AM Bug #8978: ceph ping not working as expected
I'm experiencing the same (on newly installed ceph-cluster via Ubuntu server 14.04.1):
ceph status
cluster b6...
Kees Boogert

08/09/2014

11:34 PM Revision e087862d (ceph): Add rbdcache max dirty object option
Librbd will calculate max dirty object according to rbd_cache_max_size, it
doesn't suitable for every case. If user s...
Haomai Wang
11:34 PM Revision 5f74fb29 (ceph): librbd/internal.cc: check earlier for null pointer
Fix potential null ponter deref, move check for 'order != NULL'
to the beginning of the function to prevent a) deref ...
Danny Al-Gaaf
11:34 PM Revision b2189465 (ceph): librbd: add an interface to invalidate cached data
This is useful for qemu to guarantee live migration with caching is
safe, by invalidating the cache on the destinatio...
Josh Durgin
11:33 PM Revision f7dc5855 (ceph): librbd: check return code and error out if invalidate_cache fails
This will only happen when shrinking or rolling back an image is done
while other I/O is in flight to the same ImageC...
Josh Durgin
10:55 PM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
Unfortunately converting RBD to image format 2 did not fix it. User returned after being away for a week and her syst... Dmitry Smirnov
10:50 PM Revision a86e497a (ceph): Add rbdcache max dirty object option
Librbd will calculate max dirty object according to rbd_cache_max_size, it
doesn't suitable for every case. If user s...
Haomai Wang
10:49 PM Revision d0238131 (ceph): librbd/internal.cc: check earlier for null pointer
Fix potential null ponter deref, move check for 'order != NULL'
to the beginning of the function to prevent a) deref ...
Danny Al-Gaaf
10:47 PM Revision dda98b9a (ceph): librbd: add an interface to invalidate cached data
This is useful for qemu to guarantee live migration with caching is
safe, by invalidating the cache on the destinatio...
Josh Durgin
10:47 PM Revision d9357951 (ceph): librbd: check return code and error out if invalidate_cache fails
This will only happen when shrinking or rolling back an image is done
while other I/O is in flight to the same ImageC...
Josh Durgin
10:42 PM Revision 61791420 (ceph): Avoid extra check for clean object
We needn't to check clean object via buffer state, skip the clean object.
Signed-off-by: Haomai Wang <haomaiwang@gma...
Haomai Wang
10:36 PM Revision d474443b (ceph): rbd.cc: yes, cover formatted output as well. sigh.
Fixes: #7577
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherr...
Dan Mick
10:36 PM Revision 051d1b4e (ceph): rbd.cc: tolerate lack of NUL-termination on block_name_prefix
Fixes: #7577
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
(cherry picked ...
Dan Mick
10:30 PM Revision 61e2219d (ceph): rbd: don't forget to call close_image() if remove_child() fails
close_image() among other things unregisters a watcher that's been
registered by open_image(). Even though it'll tim...
Ilya Dryomov
08:24 PM Revision c7fbf272 (ceph): os/FileStore: dump open fds before asserting
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4e8de1792731cf30f27...
Sage Weil
08:23 PM Revision 2c6c23ce (ceph): os/FileStore: dump open fds before asserting
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 4e8de1792731cf30f27...
Sage Weil
08:23 PM Revision 4e8de179 (ceph): os/FileStore: dump open fds before asserting
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
07:52 PM Revision 281159bf (ceph): doc: Initial network docs.
Signed-Off-By: Kevin Cox <kevincox@kevincox.ca> Kevin Cox
06:51 PM Revision 978f5749 (ceph): ceph_test_rados_api_tier: do fewer writes in HitSetWrite
We don't need to do quite so many writes. It can be slow when we are
thrashing and aren't doing anything in parallel...
Sage Weil
05:50 PM CephFS Bug #9056: fuse kmod + ceph-fuse triggers "BUG: sleeping function called from invalid context"

http://pulpito.front.sepia.ceph.com/john-2014-08-09_14:56:53-fs-wip-mds-contexts-testing-basic-plana/409236/
http:...
John Spray
05:48 PM CephFS Bug #9056 (Resolved): fuse kmod + ceph-fuse triggers "BUG: sleeping function called from invalid ...

kernel 5f740d7e1531099b888410e6bab13f68da9b1a4d
wip-mds-contexts (aka wip-objecter) 7be59771bff09e2b46b5467627cb...
John Spray
12:53 PM Bug #9055 (Resolved): LibRadosTwoPoolsPP.HitSetWrite (and others) fail on remove of whiteout
2014-08-09T09:03:14.670 INFO:tasks.workunit.client.0.plana70.stdout:test/librados/TestCase.cc:93: Failure
2014-08-09...
Sage Weil
12:26 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
2014-08-08 10:55:12.312751 7f1237847700 10 osd.0 pg_epoch: 462 pg[2.1( v 462'2839 (0'0,462'2839] local-les=422 n=53 e... Sage Weil
10:04 AM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
almost there. on osd.0, we finish trimming 14a here:
2014-08-08 10:55:12.311901 7f1237847700 10 osd.0 pg_epoch: 4...
Sage Weil
12:03 PM Revision 4f6d5133 (ceph): os/chain_xattr: Remove all old xattr entry when overwrite the xattr.
Ceph use multiple xattrs to store the value of a single xattr which size
is larger than CHAIN_XATTR_MAX_BLOCK_LEN.
Bu...
Ma Jianpeng
11:43 AM Bug #8894: osd/ReplicatedPG.cc: 9281: FAILED assert(object_contexts.empty())
ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-08-08_22:30:19-rados-wip-sage-testing-testing-basic-burnupi/... Sage Weil
06:00 AM Revision 4d2d4dd5 (ceph): Merge branch 'next'
Conflicts:
src/osd/OSD.cc
Sage Weil
05:59 AM Revision 84c96547 (ceph): OSD: add require_same_peer_inst(OpRequestRef&,OSDMap&) helper
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit e99acf9810976b1fc74b84ad289773af43be973f)
Greg Farnum
05:59 AM Revision 4a0c9417 (ceph): OSD: introduce require_self_aliveness(OpRequestRef&,epoch_t) function
Take the self-aliveness checks out of require_same_or_newer_map() and use
the new function for that and for require_u...
Greg Farnum
05:59 AM Revision 48c9b384 (ceph): OSD: use OpRequestRef& for a few require_* functions
Signed-off-by: Greg Farnum <greg@inktank.com>
(cherry picked from commit eb2f1ea2c33647934af878b504383829f5a198ce)
Greg Farnum
05:59 AM Revision f86bf1d8 (ceph): OSD: introduce require_up_osd_peer() function for gating replica ops
This checks both that a Message originates from an OSD, and that the OSD
is up in the given map epoch.
We use it in h...
Greg Farnum
03:08 AM Revision 65706b11 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
02:16 AM Revision 3663233a (ceph): rgw: return error if accessing object in non-existent bucket
Fixes: #7064
Instead of trying to access the object, which is impossible as we don't
even have a proper bucket info. ...
Yehuda Sadeh
02:05 AM Revision 22aa8141 (ceph): Merge remote-tracking branch 'gh/firefly-next' into firefly
Sage Weil
01:39 AM Bug #9044 (Fix Under Review): erasure-code: use ruleset instead of ruleid
"associated pull request":https://github.com/ceph/ceph/pull/2232 Loïc Dachary
01:17 AM Revision 790f4c23 (ceph): Merge pull request #2228 from ceph/wip-pg-epoch
osd: fix pg epoch floor tracking
Reviewed-by: Samuel Just <sam.just@inktank.com>
Sage Weil
01:12 AM Revision c5110468 (ceph): mon: add 'osd blocked-by' command
Print a histogram of how many peers are blocked by each OSD. This is
useful for quickly diagnosing which OSDs are br...
Sage Weil
01:12 AM Revision 6380f478 (ceph): qa/workunits/cephtool: add simple 'ceph osd perf' test
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision bee79ec4 (ceph): qa/workunits/cephtool: add trivial 'ceph osd blocked-by' test
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision d8627313 (ceph): osd/PG: bound number of blocked_by OSDs we report per PG
We don't want to include an unbounded-sized vector of OSDs in the pg_stat_t
struct. Cap it, and report a random subs...
Sage Weil
01:12 AM Revision ee790e36 (ceph): osd/osd_types: s/int/int32_t/ in pg_stat_t
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision 6e96c2d9 (ceph): osd/PG: set blocked_by during peering GetInfo
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision ceaca9f1 (ceph): osd/PG: set blocked_by during peering GetLog
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision 6fcdbd21 (ceph): osd/PG: set blocked_by in Incomplete state
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision 47a9af35 (ceph): osd/PG: set blocked_by in GetMissing
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
01:12 AM Revision d55e9730 (ceph): mon/PGMap: track histogram of pg blocked_by
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:53 AM Revision 448edb19 (ceph): Merge pull request #2209 from somnathr/wip-sd-check-pool-existence
RadosClient: Adding pool existence checks for pool create interfaces.
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil

08/08/2014

11:57 PM Revision e4b00718 (ceph): RadosClient: Enable pool existence check
Pool create interfaces would now refer to updated osdmap
for pool existence, such that an appropriate message is
thro...
Pavan Rallabhandi
11:00 PM Bug #9054 (Resolved): ceph_test_rados: FAILED assert(!old_value.deleted())
ubuntu@teuthology:/a/teuthology-2014-08-06_02:30:01-rados-next-testing-basic-plana/403383... Sage Weil
10:58 PM Bug #8997: ceph_test_rados_watch_notify hangs
ubuntu@teuthology:/a/teuthology-2014-08-06_02:30:01-rados-next-testing-basic-plana/402968 Sage Weil
10:55 AM Bug #8997: ceph_test_rados_watch_notify hangs
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-08-06_02:30:01-rados-next-testing-basic-plana/402968 Sage Weil
10:54 PM Bug #9053 (Resolved): mon/Paxos.cc: 628: FAILED assert(begin->last_committed == last_committed)
ubuntu@teuthology:/a/teuthology-2014-08-06_02:30:01-rados-next-testing-basic-plana/402965
description: rados/monthra...
Sage Weil
07:36 PM Bug #9052: ceph-mon crashes with *** Caught signal (Floating point exception) **
With no OSDs in the cluster, the calculations for @pgs_per_osd@ can divide by zero (integer, but that still causes th... Dan Mick
07:29 PM Bug #9052 (Resolved): ceph-mon crashes with *** Caught signal (Floating point exception) **
I've found that I can crash ceph-mon by attempting to change pool values (such as pg_num) before adding OSDs to the c... Jamin Collins
06:59 PM rgw Documentation #9051 (Closed): Document rgw_defer_to_bucket_acls option
It appears that the only documentation right now is the commit message of 1d7c2041. Benjamin Gilbert
06:16 PM Bug #7576: osd: large skew in pg epochs (dumpling)
..and when we do, include commit:a52a855f6c92b03dd84cd0cc1759084f070a98c2 !! Sage Weil
06:16 PM Bug #7576 (Pending Backport): osd: large skew in pg epochs (dumpling)
still want to backport this to firefly ... Sage Weil
06:04 PM rgw Bug #8621: civetweb frontend fails authentication if URL has special chars
tested wip-8621 by executing s3tests, there are still a few failures,
logs are copied to ubuntu@mira042.front.sepi...
Tamilarasi muthamizhan
04:42 PM Fix #4205: librados: Improve Watch-notify semantics
http://pad.ceph.com/p/watch-notify Sage Weil
03:55 PM devops Feature #9050 (Rejected): Calamari builds for ceph.com
Neil Levine
03:24 PM devops Feature #6310 (Closed): Get Dumpling into CentOS Ceph repo
Neil Levine
02:53 PM Revision c34f9352 (ceph): test_librbd_fsx: also flatten as part of randomize_parent_overlap
With randomize_parent_overlap fsx will randomly truncate base images
after they have been cloned from. This throws f...
Ilya Dryomov
10:31 AM Bug #9046 (Resolved): Limiting the pool object quota stops the IO, however IO does not restart if...
Issue Title: Limiting the pool object quota stops the IO, however IO does not restart if we rest the pool object quot... Hirak Mazumder
10:03 AM Revision 214630b1 (ceph): mds: Also check min/max uid on snap removal
Not only check if the uid is correct when creating a snapshot, but also
verify if the UID is correct when trying to r...
Wido den Hollander
09:37 AM Bug #9040: clients can SEGV during package upgrade
Ian Colle
09:03 AM Bug #9023: valgrind failures in OSD
Another `new Session` at OSD.cc:3704
http://qa-proxy.ceph.com/teuthology/john-2014-08-07_18:44:20-fs-wip-mds-context...
John Spray
06:43 AM Bug #9044 (Resolved): erasure-code: use ruleset instead of ruleid
When "ruleset is looked up by name":https://github.com/ceph/ceph/blob/firefly/src/mon/OSDMonitor.cc#L2928 when creati... Loïc Dachary
03:15 AM Feature #9025: erasure-code: chunk remapping
"requeued, for ubuntu 14.04 to get quicker results":http://pulpito.ceph.com/loic-2014-08-08_12:17:30-upgrade:firefly-... Loïc Dachary
03:13 AM Feature #8496: erasure-code: ErasureCode base class
"requeued, for ubuntu 14.04 to get quicker results":http://pulpito.ceph.com/loic-2014-08-08_12:13:20-upgrade:firefly-... Loïc Dachary
02:43 AM rgw Bug #9043 (Duplicate): rgw:Cannot add object to Ceph using Openstack Dashboard(Horizon) in firefly
Uploading a new object fails with message "Error: Unable to upload object".
While adding an object using Horizon w...
Ashish Chandra
01:24 AM Revision a1449506 (ceph): qa/workunits/cephtool/test_daemon.sh: allow local ceph command
(cherry picked from commit 97a8d5a9fdbd3a25cc922c242ee57da58c57d0bc) Dan Mick
01:24 AM Revision 1ac0e543 (ceph): qa/workunits/cephtool/test_daemon.sh: typo
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 22d20f39b7355966554319d5a1aa888967607569)
Dan Mick
01:24 AM Revision cb48cd47 (ceph): qa/workunits/cephtool/test_daemon.sh: verify ceph -c works with daemon
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit aa9ae1f270293778aa937e7f7e4bcaee3099b9b2)
Dan Mick
01:24 AM Revision 76b593f6 (ceph): ceph.in: Pass global args to ceph-conf for proper lookup
Fixes: #8944
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 6d89a99648630f81b85ad115fe7662...
Dan Mick
12:42 AM Revision a52a855f (ceph): osd: fix pg epoch floor tracking
If you call erase() on a multiset it will delete all instances of a value;
we only want to delete one of them. Fix t...
Sage Weil

08/07/2014

10:56 PM Revision de0c7202 (ceph): Merge pull request #2226 from athanatos/wip-8396
Wip 8396
Reviewed-by: Sage Weil <sage@redhat.com>
Samuel Just
10:54 PM Revision 2120f4bb (ceph): OSD: move waiting_for_pg into the session structures
Each message belongs to a session. Further, no ordering is implied
between messages which arrived on different sessi...
Samuel Just
10:42 PM Revision f5d6818d (ceph): Merge pull request #2227 from athanatos/wip-8625
osd_types: s/stashed/rollback_info_completed and set on create
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
10:30 PM Revision 71cb4d3c (ceph): doc: Removed Debian reference from Upstart.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
10:29 PM Revision 8357cae0 (ceph): doc: Incorporated user feedback to clarify upgrade doc.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
09:15 PM Revision b6bf33cd (ceph): doc: Added configuration discussion at end of gateway install with link...
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
06:48 PM Revision 6e7b86a2 (ceph): OSD: fix wake_pg_waiters revert error in _open_lock_pg
231fe1b685bfbd3db9c81709ca39a29d696b13ad reintroduced erroneously
this call to wake_pg_waiters. All _create_lock_pg ...
Samuel Just
06:48 PM Revision 20fd714a (ceph): OSD: wake_pg_waiters outside of the pgmap write_lock, pg_lock
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
06:48 PM Revision 11d6e5ad (ceph): OSD: rename session_waiting_for_map_lock to session_waiting_lock
This lock will also protect the waiting_for_pg structures in each
session.
Signed-off-by: Samuel Just <sam.just@inkt...
Samuel Just
06:48 PM Revision 6aba9421 (ceph): OSD: clear_session_waiting_on_map in ms_handle_reset
Signed-off-by: Samuel Just <sam.just@inktank.com> Samuel Just
06:48 PM Revision 11cefcab (ceph): OSD::shutdown: actually drop sessions waiting on map
There might be messages for which we still don't have the
map. Dispatching waiting won't actually help.
Signed-off-...
Samuel Just
06:47 PM Revision 42607679 (ceph): osd_types: s/stashed/rollback_info_completed and set on create
Originally, this flag indicated that the object had already been stashed and
that therefore recording subsequent chan...
Samuel Just
06:21 PM Revision f95e82b1 (ceph): Merge remote-tracking branch 'gh/wip-filestore-bigxattr'
Sage Weil
03:56 PM Feature #8276: ceph-filestore-dump import-rados -p <pool> <archive>
Implemented syntax:
ceph_objectstore_tool import-rados pool [import_file|-]
Import into the specified pool on r...
David Zafman
03:54 PM Bug #8396 (Resolved): osd: message delayed in Session misdirected after split
Samuel Just
03:39 PM Bug #8625 (Pending Backport): EC pool - OSD creates an empty file for op with 'create 0~0, writef...
Sage Weil
03:11 PM Revision d0ccb1cb (ceph): make ceph-disk use the new init flag for cluster
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
02:48 PM Revision 23b4915f (ceph): allow passing a --cluster flag to the init script
It will fallback to looking at /etc/ceph/$cluster.conf
otherwise.
Signed-off-by: Alfredo Deza <alfredo.deza@inktank....
Alfredo Deza
02:34 PM Bug #9040: clients can SEGV during package upgrade
https://github.com/ceph/ceph-qa-suite/pull/77 seemed fixing this.
Testing now.
Yuri Weinstein
01:56 PM Bug #9040 (Won't Fix): clients can SEGV during package upgrade
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-06_16:30:35-upgrade:dumpling-dumpling---basic-vps/... Yuri Weinstein
02:24 PM Revision 4e1e929f (ceph): Update module references
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision 0e1df3cc (ceph): Import teuthology tasks (master branch)
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision 65060738 (ceph): Merge branch 'teuth_master' into tasks_master
Zack Cerza
02:24 PM Revision ad65d6bb (ceph): Update module references
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision 6f854e12 (ceph): Use newer orchestra API
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision a963e791 (ceph): Use teuthology's DaemonGroup
It has timeout support.
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
02:24 PM Revision 2e9a567d (ceph): Use Remote.user
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision 6967e6c6 (ceph): Merge branch 'teuth_firefly' into tasks_firefly
Zack Cerza
02:24 PM Revision f6a72a3d (ceph): Add wait_until_fuse_mounted()
This was removed in a refactor in master, but it would take lots of work
to backport the entire refactor.
Signed-off...
Zack Cerza
02:24 PM Revision 416f41e2 (ceph): Move write_secret_file() into task_util/kclient.py
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
02:24 PM Revision 4318e6ee (ceph): Import teuthology tasks (firefly branch)
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
01:24 PM Revision 6bd2b0fc (ceph): mds: Make min/max UID configurable for who is allowed to create a snapshot
This way administrators can configure which UIDs are allowed to create
snapshots on CephFS.
In certain situations it...
Wido den Hollander
01:12 PM Revision a5ecf151 (ceph): powerdns: Update README with better markdown
Wido den Hollander
12:37 PM rgw Bug #9039: Using COPY on radosgw to copy object from one bucket to another that's in another pool...
Well, I think data copy is the right thing to do. If I put bucket in different pool is because they're configured dif... Sylvain Munaut
10:40 AM rgw Bug #9039: Using COPY on radosgw to copy object from one bucket to another that's in another pool...
The problem is that it is implicitly assumed with the new manifest that the tail is going to reside at the same pool ... Yehuda Sadeh
10:07 AM rgw Bug #9039: Using COPY on radosgw to copy object from one bucket to another that's in another pool...
Really ? I didn't see anything in the code that checked whether the destination bucket was in the same pool or not an... Sylvain Munaut
09:59 AM rgw Bug #9039: Using COPY on radosgw to copy object from one bucket to another that's in another pool...
That sounds like an issue with the new (firefly) manifest. Yehuda Sadeh
07:21 AM rgw Bug #9039 (Resolved): Using COPY on radosgw to copy object from one bucket to another that's in a...
Currently if you copy an object from a bucket to another one which is in another rados pool, things will just break. ... Sylvain Munaut
09:34 AM Bug #9035 (Closed): ceph cluster is using more space than actual data after replication
the used is simply summing the statfs(2) results on all the OSDs. you can see this by doing a df on the osd volumes,... Sage Weil
02:24 AM Bug #9035 (Closed): ceph cluster is using more space than actual data after replication
Ceph cluster is using more space than estimated space to store data after replication.
Total cluster capacity is 5...
Srinivasula Reddy Maram
08:43 AM Revision bf9726a2 (ceph): mon/OSDMonitor : Use user provided ruleset for replicated pool
When creating a replicated pool, currently ceph ignore the ruleset
name provided by user but use a global default rul...
Xiaoxi Chen
07:52 AM rgw Bug #9037 (Duplicate): civetweb: error HEAD responses return body
Ian Colle
07:40 AM rgw Bug #9037: civetweb: error HEAD responses return body
Ah, sorry, somehow managed to miss it when I looked through the issue list. Please close this then. Valtteri Vuorikoski
07:34 AM rgw Bug #9037: civetweb: error HEAD responses return body
See #8539 Sylvain Munaut
02:59 AM rgw Bug #9037 (Duplicate): civetweb: error HEAD responses return body
0.80.5 radosgw with civetweb frontend returns body data when sending an error response to a HEAD request. This breaks... Valtteri Vuorikoski
06:41 AM CephFS Feature #9029: min/max uid for snapshot creation
Wido den Hollander
06:00 AM Bug #4254: osd: failure to recover before timeout on rados bench and thrashing; negative stats
I am seeing this issue again on v0.80.4. I stopped 3 osd processes and marked them as out to trigger data migration (... Zhi Zhang
03:08 AM Feature #8496: erasure-code: ErasureCode base class
"requeued on vps because plana are very busy":http://pulpito.ceph.com/loic-2014-08-07_12:09:48-upgrade:firefly-x:stre... Loïc Dachary
03:06 AM Feature #9025: erasure-code: chunk remapping
"queued the suite on vps because plana are very busy":http://pulpito.ceph.com/loic-2014-08-07_12:06:56-upgrade:firefl... Loïc Dachary
12:54 AM Feature #9025: erasure-code: chunk remapping
"upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-07_09:56:17-upgrade:firefly-x:stress-split-wip-... Loïc Dachary
01:42 AM Revision 9bbd1149 (ceph): Merge pull request #2220 from somnathr/wip-lock-leak-fix
RadosClient: Fixing potential lock leaks.
Backport: firefly
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
01:30 AM Revision f1aad8bc (ceph): RadosClient: Fixing potential lock leaks.
In lookup_pool and pool_delete, a lock is taken
before invoking wait_for_osdmap, but is not
released for the failure ...
Pavan Rallabhandi
01:23 AM Feature #9034 (New): erasure-code: better LRC strategy
The current LRC recovery strategy does not take advantage of all possibilities and may fail to discover a scenario th... Loïc Dachary
01:17 AM Feature #9033 (Resolved): erasure-code: simplified LRC
Add implicit parity and simplified LRC as "described by Andreas":https://www.mail-archive.com/ceph-devel@vger.kernel.... Loïc Dachary
12:45 AM Revision 0a1cea58 (ceph): os/FileStore: force any new xattr into omap on E2BIG
If we have a huge xattr (or many little ones), the _fgetattrs() for the
inline_set will fail with E2BIG. The conditi...
Sage Weil
12:28 AM Revision 26750fcf (ceph): os/FileStore: force any new xattr into omap on E2BIG
If we have a huge xattr (or many little ones), the _fgetattrs() for the
inline_set will fail with E2BIG. The conditi...
Sage Weil
12:11 AM Revision d050cf6c (ceph): Merge pull request #2218 from ceph/wip-rados-xattr
rados: fix get/setxattr commands up
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
Yehuda Sadeh
12:06 AM Revision e3819b62 (ceph): qa/workunits/rados/test_rados_tool: add a few xattr tests
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:06 AM Revision cc3112ef (ceph): rados: use STD{IN,OUT}_FILENO for magic values
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil

08/06/2014

10:43 PM Revision 189e4a92 (ceph): Removed upgrade/dumpling from master branch as it should be only in dum...
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
10:31 PM Revision 645c28a1 (ceph): rados: optionally read setxattr value from stdin
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
10:31 PM Revision 59a715a3 (ceph): rados: don't add \n to getxattr
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
09:53 PM Revision e5b67f75 (ceph): Update doc
File 'sysvinit' is necessary to allow the start of mon at each reboot.
Signed-off-by: François Lafont <francois.lafon...
François Lafont
08:58 PM Revision 789284c4 (ceph): mon/PGMap: s/osdid/osd/ in 'osd perf' plaintext heading
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:39 PM Revision 1b8a9ec2 (ceph): mon/OSDMonitor: clean up 'osd perf' command placement
This doesn't need an OSDMap; move it out of that block.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:22 PM Revision 940e42df (ceph): osd/PG: populate pg_stats_t blocked_by
Keep a PG member for OSDs that current block us, and fill it into the
stats struct when we publish. Use a vector<int...
Sage Weil
08:16 PM Revision ea731ae1 (ceph): qa/workunits/cephtool/test.sh: fix 'ceph df ...' tests
Broken by ee2dbdb0f5e54fe6f9c5999c032063b084424c4c and friends.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
07:30 PM Revision bdf5f300 (ceph): use cluster name when checking the host
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
07:30 PM Revision 3fb25c28 (ceph): use name in /var/lib/ceph paths
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
07:29 PM Revision 9e08e3a7 (ceph): when starting an OSD, pass in the cluster name
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
06:40 PM Bug #9022 (Pending Backport): Potential lock leaks in RadosClient
Sage Weil
02:58 AM Bug #9022: Potential lock leaks in RadosClient
Pull request on the way. Pavan Rallabhandi
02:58 AM Bug #9022 (Resolved): Potential lock leaks in RadosClient
While going through RadosClient, identified couple of interfaces librados::RadosClient::lookup_pool() and librados::R... Pavan Rallabhandi
06:37 PM Revision 0bd7c341 (ceph): PendingReleaseNotes: make note about rbd cache default change
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:32 PM Revision 64d5c406 (ceph): rgw: calc md5 and compare if user provided appropriate header
Fixes: #8436
Backport: firefly
This was broken in ddc2e1a8e39a5c6b9b224c3eebd1c0e762ca5782. The fix
resurrects and o...
Yehuda Sadeh
06:32 PM Revision ebecd80d (ceph): rgw: calculate user manifest
Fixes: #8169
Backport: firefly
We didn't calculate the user manifest's object etag at all. The etag
needs to be the m...
Yehuda Sadeh
06:28 PM Revision 8d8b8b11 (ceph): Merge pull request #2123 from ceph/wip-rbd-flush
librbd: enable rbd cache by default; writethrough until flush
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
06:08 PM Revision 7d820123 (ceph): Merge pull request #2205 from ceph/wip-librbd-snap-meta
librbd: fix crash with a chain of flattened images
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
05:51 PM Revision c0ef8b87 (ceph): rocksdb: fix i386 build
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
05:44 PM Revision 7a9b986a (ceph): rados: add {read,write}_ops to df json
Backport: firefly, dumpling
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:41 PM Revision 0e9f3759 (ceph): Merge remote-tracking branch 'gh/wip-test-ceph-disk'
Sage Weil
05:33 PM Revision 6168c3c9 (ceph): Merge pull request #2215 from ceph/wip-kb
mon: clean up _kb fields in json and perf counter output
Reviewed-by: John Spray <john.spray@redhat.com>
Sage Weil
05:33 PM Revision ee2dbdb0 (ceph): mon/PGMonitor: remove {rd,wr}_kb from pool stat dumps
These fields are replaced with corresponding *_bytes fields.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:33 PM Revision 4c8e3224 (ceph): mon: remove *_kb perf counters
This is an incompatible change.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:26 PM Revision 1f3f2982 (ceph): rgw: fix crash in swift CORS preflight request
Fixes: #8586
This fixes error handling, in accordance with commit 6af5a537 that fixed
the same issue for the S3 case...
Yehuda Sadeh
05:26 PM Revision 6e492150 (ceph): cls_rgw: fix object name of objects removed on object creation
Fixes: #8972
Backport: firefly, dumpling
Reported-by: Patrycja Szabłowska <szablowska.patrycja@gmail.com>
Signed-off...
Yehuda Sadeh
05:21 PM Revision 53aab689 (ceph): Merge pull request #75 from ceph/wip-readforward
include readforward cache mode in cache mode cycling test
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
04:18 PM Revision 3179bb10 (ceph): mon/PGMonitor: add _bytes perf counters
Leave the _kb ones in place for now.
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
04:14 PM Revision adb2791a (ceph): mon/PGMonitor: add _bytes fields for all usage dumps
Leave the _kb ones in place for now.
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
04:06 PM Revision fce2ed68 (ceph): Remove most ceph-specific tasks. They are in ceph-qa-suite now.
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
03:50 PM Revision 7dbd4401 (ceph): ignore errors if 'ps aux' fails
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit f9aa9c1e647dc3376a33d0f7147501bffcd...
Alfredo Deza
03:42 PM Revision d5f26aaf (ceph): Merge pull request #306 from ceph/wip-9024
ignore errors if 'ps aux' fails Zack Cerza
03:39 PM Feature #9031: List RADOS namespaces and list all objects in all namespaces

A way to implement this is to enhance the pg_ls_repsonse_t to include the namespace (or change object_t to hobject_...
David Zafman
02:30 PM Feature #9031 (Resolved): List RADOS namespaces and list all objects in all namespaces
We can currently create namespaces, but cannot easily view those that have been created. A method of listing namespac... Brian Andrus
03:23 PM Revision b416446c (ceph): Merge pull request #2165 from dachary/wip-mailmap
mailmap updates Loïc Dachary
03:23 PM devops Bug #9032 (Rejected): ceph-deploy over proxy
I have my servers working behind a proxy. When I run the ceph-deploy install command I get an error:
[ceph01][INFO ...
TJ Walker
03:16 PM Revision 895318c0 (ceph): README.md: word wrap
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
03:15 PM Revision 500b95e5 (ceph): README: symlink from README.md
It looks better as markdown than rendered as text via the markdown tool,
so just symlink it.
Signed-off-by: Sage Wei...
Sage Weil
02:53 PM Revision f9aa9c1e (ceph): ignore errors if 'ps aux' fails
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
02:46 PM Revision 25b9761e (ceph): Merge pull request #2202 from xinxinsh/enable-rocksdb-log-level
Enable rocksdb log level
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
02:05 PM Feature #9030 (Fix Under Review): mon: quickly identify 'problem'  osds
Sage Weil
02:05 PM Feature #9030 (Resolved): mon: quickly identify 'problem'  osds
Sage Weil
01:53 PM Revision 0114b335 (ceph): erasure-code: rework ErasureCode*::parse methods
The ErasureCode::parse virtual function is derived in
ErasureCode{Jerasure,Isa}. It is reworked to share parsing argu...
Loïc Dachary
01:44 PM Revision 77690f66 (ceph): erasure-code: move to ErasureCode::decode_concat
It does not belong to ErasureCodeInterface which is supposed to not have
any implementation.
Signed-off-by: Loic Dac...
Loïc Dachary
01:22 PM Revision 54394fa1 (ceph): erasure-code: move to ErasureCode::to_{int,bool}
The parameter parser helpers to_int and to_bool are moved from
ErasureCode{Jerasure,Isa} to ErasureCode.
The prototy...
Loïc Dachary
01:21 PM Revision 4ff981f4 (ceph): erasure-code: move to ErasureCode::{encode,decode}{,chunk}
The bulk of ErasureCode{Jerasure,Isa}::encode and
ErasureCode{Jerasure,Isa}::decode is dealing with alignment and buf...
Loïc Dachary
01:21 PM Revision b4b7c51f (ceph): erasure-code: move to ErasureCode::minimum_to_decode*
The ErasureCode{Jerasure,Isa}::minimum_to_decode and
ErasureCode{Jerasure,Isa}::minimum_to_decode_with_cost methods a...
Loïc Dachary
01:21 PM Revision 1ebce988 (ceph): erasure-code: ErasureCode base class
Add the ErasureCode base class for ErasureCodeJerasure and
ErasureCodeIsa, for methods that can be shared among all p...
Loïc Dachary
12:55 PM Bug #8860 (Fix Under Review): ceph-disk issues with custom cluster name
PR opened https://github.com/ceph/ceph/pull/2216 Alfredo Deza
12:25 PM CephFS Feature #9029 (Resolved): min/max uid for snapshot creation
On shared systems like shared hosting it might be useful to prevent regular users from creating snapshots on CephFS.
...
Wido den Hollander
12:20 PM rgw Feature #6747: PowerDNS backend for RGW bucket directing
Wido den Hollander
11:06 AM rbd Bug #8845 (Pending Backport): Flattening Clones of clone, results in command failure
Sage Weil
09:41 AM Bug #9019 (Resolved): Makefile.am: error: required file './README' not found
fixed it up with a symlink.. other solutions seemed more annoying :( Sage Weil
08:39 AM Linux kernel client Bug #8818 (Resolved): IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
OK, thanks everybody.... Ilya Dryomov
08:09 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
I switched to the good kernel (3.16.0-ceph-00037-g0532581) yesterday and re-ran my scripts overnight. The scripts co... Greg Wilson
08:39 AM Linux kernel client Bug #8464 (Resolved): krbd: deadlock
OK, thanks everybody.... Ilya Dryomov
08:06 AM Feature #8496: erasure-code: ErasureCode base class
"scheduled upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-06_17:07:04-upgrade:firefly-x:stress-... Loïc Dachary
06:22 AM Feature #8496: erasure-code: ErasureCode base class
The test "only had one job":http://pulpito.ceph.com/loic-2014-08-05_13:45:56-upgrade:firefly-x:stress-split-wip-8496-... Loïc Dachary
07:12 AM Feature #9025 (Fix Under Review): erasure-code: chunk remapping
"need review":https://github.com/ceph/ceph/pull/2213 Loïc Dachary
06:28 AM Feature #9025 (Resolved): erasure-code: chunk remapping
Interpret the *mapping* parameter and remap the chunks accordingly. For instance mapping=_DD means the data chunks ar... Loïc Dachary
07:11 AM CephFS Feature #9026 (Resolved): client: vxattr support for rctime, rsize, etc.
Sage Weil
05:44 AM Bug #9023 (Can't reproduce): valgrind failures in OSD

osd.2 from OSD.cc:462 (SafeTimer::init, pthread_create)
http://pulpito.front.sepia.ceph.com/john-2014-08-01_11:0...
John Spray

08/05/2014

11:48 PM Revision 859944d3 (ceph): test/osd/osd-test-helpers: mkdir -p for ceph-disk
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:38 PM Revision 4b45e25b (ceph): Add handles for the ceph-mon starting with upstart
François Lafont
11:22 PM Feature #9021 (Resolved): librbd: shared flag, object map
we need to consider to make a tradeoff between multi-client support and single-client support for librbd. In practice... Haomai Wang
10:43 PM Bug #8797: "ceph status" do not exit with python_2.7.8
For a moment Python maintainer in Debian kindly fixed this issue for us by adding patch to revert problematic change ... Dmitry Smirnov
10:25 PM Revision 818f5b70 (ceph): Merge pull request #2208 from lpabon/osd_dev_doc
Developer quick start guide
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
10:11 PM Revision c2f58e66 (ceph): test/ceph-disk.sh: mkdir -p
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
09:52 PM Revision 511a0597 (ceph): Merge pull request #76 from ceph/wip_tune2_upgrade-firefly
Added ec-readwrite.yaml to the mix to enable erasure code coverage
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
09:48 PM Revision 542dbd2a (ceph): Added ec-readwrite.yaml to the mix to enable erasure code coverage
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
09:06 PM Revision 3f9f135f (ceph): Merge remote-tracking branch 'upstream/next' into wip-sam-testing
Conflicts:
src/osd/OSD.cc
Samuel Just
07:34 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
The 00036 "bad" kernel started showing the problem in the /var/log/kern.log file within minutes of starting my test s... Eric Eastman
12:49 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Eric, Greg,
The fix on top of 3.16 + testing is in wip-request-fn.
http://gitbuilder.ceph.com/kernel-deb-precis...
Ilya Dryomov
06:51 PM Revision a923e2c9 (ceph): Renamed README to README.md to render in markdown
Signed-off-by: Luis Pabón <lpabon@redhat.com> Luis Pabon
06:48 PM Revision 52cf693e (ceph): Developer quick start guide
Signed-off-by: Luis Pabón <lpabon@redhat.com> Luis Pabon
06:16 PM Bug #9019 (Resolved): Makefile.am: error: required file './README' not found
commit(a923e2c9eb16823fa484c) Renamed README to README.md to render in markdown. After that, i can't generate Makefil... jianpeng ma
06:15 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
> I'm guessing the request is hung on teh OSD side of things...
Thanks Sage. Sadly after radosgw daemon restarting, ...
Guang Yang
08:28 AM Bug #9008 (Need More Info): Objecter: pg listing can deadlock when throttling is in use
Sage Weil
08:28 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
please query the admin socket for the process like so:
ceph daemon /var/run/ceph/ceph-client.*.asok objecter_requ...
Sage Weil
02:44 AM Bug #9008 (Resolved): Objecter: pg listing can deadlock when throttling is in use
In our Ceph cluster (with radosgw), we found that occasionally the processing threads hands forever and eventually ha... Guang Yang
06:11 PM Revision c09036ac (ceph): enable info_log_level config option for rocksdb
Signed-off-by: xinxin shu <xinxin.shu@intel.com> xinxin shu
05:21 PM Revision e6935dd4 (ceph): Merge pull request #2206 from ceph/wip-8875
doc: be a bit more explicit about 'ceph-deploy new' in quickstart
Reviewed-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins
04:52 PM Revision 66de51d1 (ceph): be a bit more explicit about 'ceph-deploy new' in quickstart
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
04:15 PM Revision 3876da15 (ceph): Merge branch 'master' of github.com:ceph/ceph
Sage Weil
04:15 PM Revision bc1d2b83 (ceph): Merge remote-tracking branch 'gh/wip-8880'
Conflicts:
src/osd/OSD.cc
Sage Weil
03:50 PM Revision 19b9ee94 (ceph): Merge pull request #2204 from osynge/wip-dont-mkdir-by-mistake2
Do not make directories by mistake.
Reviewed-by: Alfredo Deza <adeza@redhat.com>
Alfredo Deza
03:50 PM Revision b8c1cabd (ceph): Merge pull request #1883 from ceph/wip-msgr
messenger refactoring for xio
Reviewed-by: Greg Farnum <greg@inktank.com>
Sage Weil
03:28 PM Revision b4c80e31 (ceph): Do not make directories by mistake.
Rational: I found I had created a series of OSD directories under "/dev/" when disks I thought existed did not exist.... Owen Synge
02:57 PM Revision e5b3c093 (ceph): Merge pull request #2200 from theanalyst/typo
doc: typo s/loose/lose
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
02:33 PM Revision f773b24e (ceph): powerdns: Define a application variable when not invoked from Shell
This allows to be run directly using mod_wsgi behind Apache. Wido den Hollander
02:24 PM Bug #9018 (Resolved): "LibRadosTwoPoolsPP*" failed in upgrade:dumpling-x-firefly---basic-vps
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-05_09:22:33-upgrade:dumpling-x-firefly---basic-vps... Yuri Weinstein
02:13 PM devops Feature #8868: Update Fedora to 0.80.5 packages with ceph-common
So, there's a PR open for some restructuring of the .spec file now that we need to get in soon to make this more sane... Dan Mick
01:21 PM Fix #6278 (Resolved): osd: throttle snap trimming
Sage Weil
01:20 PM devops Fix #9017 (Rejected): [paddles] implement validation across all controller methods
paddles has a lot of boilerplate in controllers that look like:... Alfredo Deza
01:15 PM Feature #9015 (Resolved): msgr refactoring to support xio work
Sage Weil
01:09 PM Feature #9015 (Resolved): msgr refactoring to support xio work
Sage Weil
01:14 PM Fix #8905 (In Progress): msgr: encode osd epoch in nonce to avoid misc OSD reconnect races
Sage Weil
01:10 PM Feature #7516 (Fix Under Review): mon: reweight-by-pg
Sage Weil
01:06 PM Feature #7238 (In Progress): erasure code : implement LRC plugin
Samuel Just
12:55 PM Bug #8083: erasure-code: fix static code analysis errors found in gf-complete
Loïc Dachary
12:28 PM Documentation #8875 (Resolved): `ceph-deploy new` needs to be called for every node, not just the...
PR https://github.com/ceph/ceph/pull/2206
and merged commit e6935dd into master
Alfredo Deza
09:37 AM Documentation #8875 (In Progress): `ceph-deploy new` needs to be called for every node, not just ...
I noted the problem in the docs and will fix that shortly.
You are right, you need to run `ceph-deploy new {NODES}...
Alfredo Deza
11:19 AM Bug #9011: osd memory leaks on next
gonna see if this happens on plana too Sage Weil
11:13 AM Bug #9011: osd memory leaks on next
these look like static std::strings. and some other weird leaks that don't make sense... Sage Weil
08:00 AM Bug #9011 (Duplicate): osd memory leaks on next
ubuntu@teuthology:/a/sage-2014-08-04_11:34:19-rgw-next-testing-basic-vps/397606
need to clean these up
Sage Weil
09:26 AM rgw Feature #9013 (Resolved): rgw: set civetweb as a default frontend
Should add civetweb to the default frontends. Yehuda Sadeh
09:13 AM Messengers Bug #8880 (Pending Backport): msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_s...
Sage Weil
09:11 AM Bug #9012 (Duplicate): "[WRN] map e277 wrongly marked me down" in upgrade:dumpling-x-firefly---ba...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-08-04_14:18:17-upgrade:dumpling-x-firefly---basic-vps... Yuri Weinstein
09:05 AM rgw Feature #8218 (In Progress): rgw: object versioning manifest changes
Ian Colle
09:05 AM rgw Feature #8217 (In Progress): rgw: object versioning object overwrite / delete changes
Ian Colle
09:05 AM rgw Feature #8216 (In Progress): rgw: object versioning objclass support
Ian Colle
09:05 AM rgw Feature #8473 (In Progress): rgw: Shard bucket index objects to improve single bucket PUT throughput
Ian Colle
08:54 AM rbd Bug #8845 (Fix Under Review): Flattening Clones of clone, results in command failure
https://github.com/ceph/ceph/pull/2205 Josh Durgin
08:52 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
btw, the steps to reproduce this issue are mentioned by Sahana above & it can be reproduced on a single node too.
...
Dhiraj Kamble
08:47 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
Hi Greg,
No i did not intend to add any comments.
The reason i thought we should assert is, so that we can serv...
Dhiraj Kamble
08:25 AM Bug #9007 (Duplicate): Ceph Firefly 0.80.4 : Unable to get some pool values
you're right. this is fixed in master, and backported to firefly-next.. will be in next firefly point release. Sage Weil
01:50 AM Bug #9007 (Duplicate): Ceph Firefly 0.80.4 : Unable to get some pool values
h1. Hello Developers
I am curious to know if there is something missing from the code for Ceph pool values.
As...
karan singh
07:56 AM rgw Bug #8676: md5sum check failed during readwrite.py
ubuntu@teuthology:/a/sage-2014-08-04_11:34:19-rgw-next-testing-basic-vps/397522 Sage Weil
05:05 AM Revision 1682e62e (ceph): doc: typo s/loose/lose
Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com> Abhishek Lekshmanan
04:46 AM Feature #8496: erasure-code: ErasureCode base class
"upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-05_13:45:56-upgrade:firefly-x:stress-split-wip-... Loïc Dachary
12:48 AM Feature #8496 (Fix Under Review): erasure-code: ErasureCode base class
"pull request":https://github.com/ceph/ceph/pull/2201 Loïc Dachary
04:22 AM Bug #9009 (In Progress): (wip-objecter) ObjectCacher assert in fs client
OK, no big deal, just that there are contexts in the Client, like the MDS, which need updating to take client_lock wh... John Spray
03:49 AM Bug #9009 (Resolved): (wip-objecter) ObjectCacher assert in fs client

From branch wip-mds-contexts, which is a derivative of wip-objecter.
http://qa-proxy.ceph.com/teuthology/john-20...
John Spray
03:21 AM rgw Feature #8911: RGW doesn't return 'x-timestamp' in header which is used by 'View Details' of Open...
It also doesnot returns "Content-type" header as well. Swift does return this header aswell. So I would love to see r... Ashish Chandra
12:45 AM rgw Documentation #9003: rgw: document development setup for rgw
Much needed. Great! Abhishek Lekshmanan
12:41 AM Revision 541006c8 (ceph): Merge pull request #1875 from dachary/wip-8437
erasure-code: benchmarking jerasure
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil

08/04/2014

11:33 PM Feature #8496 (In Progress): erasure-code: ErasureCode base class
Because it needs work to adapt the isa plugin, it deserves a separate patch. Otherwise it mixes two unrelated topics. Loïc Dachary
05:12 AM Feature #8496 (Rejected): erasure-code: ErasureCode base class
It is part of a "larger pull request":https://github.com/ceph/ceph/pull/1911 Loïc Dachary
11:21 PM Bug #8736: thrash and scrub combination lead to error
http://pulpito.ceph.com/loic-2014-08-04_15:06:02-upgrade:firefly-x:stress-split-wip-8475-testing-basic-plana/396887/
...
Loïc Dachary
11:16 PM Revision 466aba42 (ceph): osd/osd_types: add blocked_by to pg_stat_t
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:09 PM Revision b65cef67 (ceph): ceph_test_rados_api_tier: fix [EC] HitSet{Read,Write,Trim} tests
The hit_set_ fields can only be set on tier pools as of
f131dfbaedf6f451572e7aa3a83f653912122953.
Fixes: #8823
Signe...
Sage Weil
11:02 PM Feature #8475 (Resolved): erasure-code: oversized objects when using the Cauchy technique
Loïc Dachary
06:05 AM Feature #8475: erasure-code: oversized objects when using the Cauchy technique
"scheduled upgrade:firefly-x:stress-split":http://pulpito.ceph.com/loic-2014-08-04_15:06:02-upgrade:firefly-x:stress-... Loïc Dachary
02:07 AM Feature #8475: erasure-code: oversized objects when using the Cauchy technique
"Rebased and repushed":https://github.com/ceph/ceph/pull/1890 , running gitbuilder Loïc Dachary
10:24 PM Revision b9b022e5 (ceph): add annotation for rocksdb config option
Signed-off-by: xinxin shu <xinxin.shu@intel.com> xinxin shu
10:12 PM Revision dceab8dc (ceph): Merge pull request #2198 from ceph/wip-8998
fix OSD SEGV in heartbeat()
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
10:01 PM Revision 8dcfbd8e (ceph): osd: simplify dout_prefix macros
Use a get_osdmap_epoch() helper that is a bit lighter weight (by avoiding
copying around an OSDMapRef).
Signed-off-b...
Sage Weil
09:57 PM Revision 80829d78 (ceph): osd: reorder OSDService methods under proper dout_prefix macro
The dout_prefix for OSDService uses get_osdmap() to grab a shared_ptr for
the epoch printout. The OSD one does not, ...
Sage Weil
09:02 PM Revision 97f317d8 (ceph): rados/singleton: make cache mode cycling include readforward mode
forward -> off -> writeback -> readforward -> ...
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
08:56 PM Revision 0b445e0f (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
08:48 PM Revision 047c18db (ceph): doc/release-notes: make note about init-radosgw change
This changed back in 524aee6f95f9c397b7c8508934f3c0577f9df1dd but
was not mentioned in the release notes.
Signed-off...
Sage Weil
08:38 PM Revision b93abaf5 (ceph): Merge pull request #305 from ceph/wip-8640
implement `--version` and improve `setup.py` Zack Cerza
08:00 PM rgw Feature #3454: Support temp URLs for Swift API
This should be documented somewhere too, at least in the table at http://ceph.com/docs/master/radosgw/swift/ Blair Bethwaite
07:12 PM Revision fe6f4611 (ceph): create a --version flag for teuthology
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
07:11 PM Revision 7d0a072d (ceph): pull the version and description in setup.py
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
07:11 PM Revision a2bb186f (ceph): add an initial version to teuthology
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
06:47 PM Revision 354c4112 (ceph): doc: Added 'x' to monitor cap.
Signed-off-by: John Wilkins <john.wilkins@inktank.com> John Wilkins
05:50 PM Revision 0d1fe79b (ceph): Tweak usage statement
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
05:33 PM Revision 18b7a37c (ceph): Merge pull request #2166 from majianpeng/bug-fix
os/FileJournal: When dump journal, using correctly seq avoid misjudging joural corrupt.
Reviewed-by: Samuel Just <sa...
Samuel Just
05:32 PM Revision 6878e8cd (ceph): Merge pull request #2184 from majianpeng/fix2
ECBackend: Don't directyly use get_recovery_chunk_size() in RecoveryOp::WRITING state
Reviewed-by: Samuel Just <sam....
Samuel Just
05:31 PM Revision 6f05ff8f (ceph): Merge pull request #2194 from majianpeng/fix1
osd/ECBackend: clean up assert(r==0) in continue_recovery_op.
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
05:30 PM Revision 3897f09a (ceph): Merge pull request #2192 from ceph/wip-8891
msg/SimpleMessenger: drop msgr lock when joining a Pipe
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
05:04 PM Revision 7b2c8b33 (ceph): cls_rgw: fix object name of objects removed on object creation
Fixes: #8972
Backport: firefly, dumpling
Reported-by: Patrycja Szabłowska <szablowska.patrycja@gmail.com>
Signed-off...
Yehuda Sadeh
05:03 PM Revision 8519e9ab (ceph): rgw: need to pass need_to_wait for throttle_data()
need_to_wait wasn't passed into processor->throttle_data(). This was
broken in fix for #8937.
CID 1229541: (PW.PA...
Yehuda Sadeh
04:52 PM Revision 06206247 (ceph): rgw: call processor->handle_data() again if needed
Fixes: #8937
Following the fix to #8928 we end up accumulating pending data that
needs to be written. Beforehand it ...
Yehuda Sadeh
04:51 PM Revision 2f44d768 (ceph): Merge pull request #2191 from ceph/wip-rgw-need-to-wait
rgw: need to pass need_to_wait for throttle_data()
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
04:41 PM Revision 78b2c10a (ceph): Merge pull request #2195 from apeters1971/wip-ec-isa-fast-xor
EC-ISA: provide a 10% faster simple parity operation for (k, m=1)
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
04:28 PM Revision b22a8a9e (ceph): Merge pull request #2193 from ceph/wip-ceph-conf
ceph-conf: flush log on exit
Reviewed-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
04:10 PM Revision ef210806 (ceph): Merge pull request #304 from ceph/wip-fix-build-matrix
suite: fix build_matrix when non-yaml present in % dir Zack Cerza
04:07 PM Revision 21e27637 (ceph): suite: fix build_matrix when non-yaml present in % dir
First, if we get a non-yaml file, return an empty list.
Second, if we are in a % mode, exclude an empty raw result f...
Sage Weil
03:19 PM Revision dc3ae881 (ceph): add another call to use "ps aux"
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 4686115c4830939ea6c9d7a27fd89d7cec9...
Alfredo Deza
03:19 PM Revision c949ad1d (ceph): tell us if you are really not running any more
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit ec342d814676d08eb833f478ac7ffdb21cc...
Alfredo Deza
03:09 PM Bug #8998 (Pending Backport): osd: SEGV in OSD::heartbeat()
Sage Weil
03:00 PM Bug #8998 (Fix Under Review): osd: SEGV in OSD::heartbeat()
https://github.com/ceph/ceph/pull/2198 Sage Weil
09:14 AM Bug #8998: osd: SEGV in OSD::heartbeat()
ubuntu@teuthology:/a/teuthology-2014-08-03_02:30:01-rados-next-testing-basic-plana/394893 Sage Weil
02:18 PM rgw Feature #9004 (New): rgw: multi-site: multi-master
As a user, I want to be able to write to any available RGW and have that file available on other RGWs for read and wr... Neil Levine
02:06 PM Bug #8891 (Resolved): rados bench hang during thrashing
Sage Weil
09:17 AM Bug #8891 (Fix Under Review): rados bench hang during thrashing
Sage Weil
01:53 PM rgw Documentation #9003: rgw: document development setup for rgw
While we're at it, beefing up the rgw support in vstart.sh would be great. right now you can pass RGW=1 and it will ... Sage Weil
01:49 PM rgw Documentation #9003 (Closed): rgw: document development setup for rgw
Yehuda Sadeh
01:51 PM Revision 516101ae (ceph): EC-ISA: provide a 10% faster simple parity operation for (k, m=1). Add ...
Andreas Peters
11:20 AM rgw Bug #9002 (Duplicate): Creating swift key with --gen-secret in separate step from subuser creatio...
Customer reported on CentOS with Ceph v0.80.4
Steps to reproduce:
radosgw-admin user create --uid=testuser1 --dis...
Brian Andrus
11:00 AM rgw Bug #9001 (Won't Fix): Starting gateway with radosgw init script fails to create socket
Ceph Version: v0.80.4
Distro: CentOS
Customer reported, unable to reproduce.
/var/run/ceph directory owned by ...
Brian Andrus
10:02 AM Revision 985b7c2f (ceph): osd/ECBackend: clean up assert(r==0) in continue_recovery_op.
After the commit(d9106ce5e4437ab02), the assert(r==0) is no longer
necessary.
Ma Jianpeng
09:42 AM Revision 8363a94a (ceph): erasure-code: HTML display of benchmark results
The ceph_erasure_code_benchmark output is converted into a JSON series
suitable to display in HTML with the http://ww...
Loïc Dachary
09:42 AM Revision 3cc72345 (ceph): COPYING: Cloudwatt copyright is inline
Remove partial list of contributions since Cloudwatt copyright has been
placed in the copyright notices of the files ...
Loïc Dachary
09:42 AM Revision e11c3fcc (ceph): erasure-code: rework benchmark suite
Expand the default suite to enumerate all cases that are relevant to the
current code base so that it is easier to co...
Loïc Dachary
09:42 AM Revision 90592e9d (ceph): erasure-code: properly indent ErasureCodePluginSelectJerasure.cc
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
09:41 AM Revision be3e1e40 (ceph): erasure-code: control jerasure plugin variant selection
The jerasure-variant parameter is interpreted as the name of the plugin
variant to be loaded regardless of the availa...
Loïc Dachary
09:41 AM Revision 5fb4354f (ceph): erasure-code: reduce jerasure verbosity
Only output a message about adjusting the buffer size when it is
adjusted, not when the size does not need adjustment...
Loïc Dachary
09:16 AM Bug #7986: 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/2044 dirty, 0/0
ubuntu@teuthology:/a/teuthology-2014-08-03_02:30:01-rados-next-testing-basic-plana/395219 Sage Weil
08:54 AM Revision 3987ac2a (ceph): erasure-code: cauchy techniques allow w 8,16,32
Enforce the restriction at initialization time, the same way it is done
for Reed Solomon. Choosing a w value differen...
Loïc Dachary
08:54 AM Revision c7daaaf5 (ceph): erasure-code: implement alignment on chunk sizes
jerasure expects chunk sizes that are aligned on the largest possible
vector size that could be used by SSE instructi...
Loïc Dachary
07:07 AM Linux kernel client Bug #8979: GPF kernel panics - auth?
pushed wip-8979 which removes the fixed buffer size. but, we still need to make things not crash when the auth reply... Sage Weil
06:57 AM Linux kernel client Bug #8979: GPF kernel panics - auth?
yeah:
#define TEMP_TICKET_BUF_LEN 256
Sage Weil
06:48 AM Linux kernel client Bug #8979: GPF kernel panics - auth?
... Sage Weil
06:52 AM Revision cb54605b (ceph): mailmap: sort entries
to help avoid duplicates (found one)
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
06:50 AM Revision accf8c26 (ceph): mailmap: Tommi Virtanen is not with Red Hat
The entry was added by s/inktank/redhat/ and did not acknowledge that
Tommi Virtanen left Inktank before it was acqui...
Loïc Dachary
06:50 AM Revision 794f70a8 (ceph): mailmap: Kevin Cox affiliation
Reviewed-by: Kevin Cox <kevincox@kevincox.ca>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
06:50 AM Revision 6652494e (ceph): mailmap: Accela Zhao affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
06:50 AM Revision 6f238fa0 (ceph): mailmap: Adam Crume affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
06:50 AM Revision 6f24a049 (ceph): mailmap: Stephen Jahl affiliation
Reviewed-by: Stephen Jahl <stephenjahl@gmail.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
06:50 AM Revision b1ba72f8 (ceph): mailmap: George Ryall affiliation
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
06:50 AM Revision 2f21a110 (ceph): mailmap: Brian Rak affiliation
and name normalization
Reviewed-by: Brian Rak <dn@devicenull.org>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
06:50 AM Revision ef620cc3 (ceph): mailmap: Sebastien Ponce name normalization
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
06:50 AM Revision f870b344 (ceph): mailmap: João Eduardo Luís name normalization
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
06:48 AM Revision d8e64157 (ceph): mailmap: Ma Jianpeng affiliation
and name normalization
Reviewed-by: Ma Jianpeng <jianpeng.ma@intel.com>
Signed-off-by: Loic Dachary <loic@dachary.org>
Loïc Dachary
06:36 AM Documentation #8875: `ceph-deploy new` needs to be called for every node, not just the admin one
I was able to complete install.
The first step above granted sudo rights on each node.
The way I was able to get it...
Bobby Yakov
05:56 AM Documentation #8875: `ceph-deploy new` needs to be called for every node, not just the admin one
You still need a user that can call sudo without a password prompt on remote nodes.
And it looks like you only pas...
Alfredo Deza
05:47 AM devops Bug #8893 (Resolved): ceph-deploy install command on centos 6.5 reports exception
merged commit eb9ea33 into ceph:master Alfredo Deza
04:00 AM Revision 3230060f (ceph): ceph-conf: flush log on exit
This makes it deterministic whether we output
2014-08-03 20:59:45.482614 4036c80 -1 did not load config file, using ...
Sage Weil
03:09 AM Revision 076f33af (ceph): ECBackend: Don't directly use get_recovery_chunk_size() in RecoveryOp::...
We cannot guarantee that conf->osd_recovery_max_chunk don't change when
recoverying a erasure object.
If change betwe...
Ma Jianpeng
01:47 AM Bug #8601 (Resolved): erasure-code: default profile does not exist after upgrade
Loïc Dachary
01:26 AM Revision 98997f3b (ceph): msg/SimpleMessenger: drop msgr lock when joining a Pipe
Avoid this deadlock:
- a fault
- delay thread entry gets a fast dispatch message
- drops delay_lock
- calls into f...
Sage Weil

08/03/2014

09:48 PM rgw Bug #8864: radosgw help doesn't seem to display some debug options
I pushed a couple of commits to fix most of undocumented options in man pages & help for #8112. Can you let me know w... Abhishek Lekshmanan
09:35 PM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
Finally I've isolated the issue.
Something was wrong with a particular RBD image (format 1) that was created on Ceph...
Dmitry Smirnov
09:11 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
another similar hang:... Sage Weil
07:47 PM Revision 70250755 (ceph): Merge branch 'wip-8701-firefly' into firefly-next
Samuel Just
07:47 PM Revision cbfbe637 (ceph): os/LFNIndex: handle long object names with multiple links (i.e., rename)
When we rename an object (collection_move_rename) to a different name, and
the name is long, we run into problems bec...
Sage Weil
07:47 PM Revision 816ed7e9 (ceph): os/LFNIndex: FDCloser helper
Add a helper to close fd's when we leave scope. This is important when
injecting failures by throwing exceptions.
S...
Sage Weil
07:47 PM Revision 4fc72d8d (ceph): os/LFNIndex: remove alt xattr after unlink
After we unlink, if the nlink on the inode is still non-zero, remove the
alt xattr. We can *only* do this after the ...
Sage Weil
07:47 PM Revision b213866e (ceph): os/LFNIndex: only consider alt xattr if nlink > 1
If we are doing a lookup, the main xattr fails, we'll check if there is an
alt xattr. If it exists, but the nlink on...
Sage Weil
07:47 PM Revision 94fee997 (ceph): os/LFNIndex: use FDCloser for fsync_dir
This prevents an fd leak when maybe_inject_failure() throws an exception.
Signed-off-by: Sage Weil <sage@redhat.com>...
Sage Weil
07:47 PM Revision f8494c87 (ceph): ceph_test_objectstore: clean up on finish of MoveRename
Otherwise, we leave collections around, and the next test fails.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry ...
Sage Weil
07:47 PM Revision e18777ed (ceph): store_test: add long name collection_move_rename tests
Currently fails.
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 6aa48a485e03ca100f3d9e...
Samuel Just
07:47 PM Revision 5db6c12b (ceph): ceph_test_objectstore: fix warning
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit cf98805c09a38cce78ac08317899dc4152ae55a5)
Sage Weil
06:27 PM Bug #8891: rados bench hang during thrashing
i think this was the same repaer vs fast dispatch that i tracked down in wip-msgr. Sage Weil
06:23 PM Revision e36babc8 (ceph): os/MemStore: fix lock leak
CID 1228868 (#2-1 of 2): Missing unlock (LOCK)
12. missing_unlock: Returning without unlocking oc->lock.L.
Signed-of...
Sage Weil
03:40 PM Revision 4e03d5b5 (ceph): Revert "enforce rados put aligment"
This reverts commit 7a58da53ebfcaaf385c21403b654d1d2f1508e1a.
This was alread backported in dece65064d949b5afcc359cd...
Sage Weil
02:48 PM devops Bug #8330: repodata on rpm repos do not list latest ceph-deploy (1.5.2)
Agreed, this is fixed. Current repodata works perfectly with all packages showing correctly (on the same host btw, I'... Simon Ironside
08:40 AM rgw Bug #8784: rgw: completion leak
ubuntu@teuthology:/a/teuthology-2014-08-01_23:02:01-rgw-master-testing-basic-plana/394054 Sage Weil
08:39 AM Bug #8996 (Resolved): "Segmentation fault" in upgrade:dumpling-x-firefly---basic-vps suite
botched (double) backport, fixed by commit:4e03d5b512c8d2f7fa51dda95c6132e676529f9b Sage Weil
03:40 AM Revision e93818df (ceph): rgw: need to pass need_to_wait for throttle_data()
need_to_wait wasn't passed into processor->throttle_data(). This was
broken in fix for #8937.
CID 1229541: (PW.PA...
Yehuda Sadeh

08/02/2014

08:20 PM Revision 6f1a5443 (ceph): rgw: fix crash in swift CORS preflight request
Fixes: #8586
This fixes error handling, in accordance with commit 6af5a537 that fixed
the same issue for the S3 case...
Yehuda Sadeh
08:19 PM Revision dfcd9956 (ceph): rgw: fix decoding + characters in URL
Fixes: #8702
Backport: firefly
Only decode + characters to spaces if we're in a query argument. The +
query argument...
Brian Rak
08:18 PM Revision f68720a6 (ceph): rgw: align object chunk size with pool alignment
Fixes: #8442
Backport: firefly
Data pools might have strict write alignment requirements. Use pool
alignment info whe...
Yehuda Sadeh
08:18 PM Revision 4cd1b60c (ceph): rgw: object write should not exceed part size
Fixes: #8928
This can happen if the stripe size is not a multiple of the chunk size.
Backport: firefly
Signed-off-...
Yehuda Sadeh
08:18 PM Revision f9716214 (ceph): rgw: call processor->handle_data() again if needed
Fixes: #8937
Following the fix to #8928 we end up accumulating pending data that
needs to be written. Beforehand it ...
Yehuda Sadeh
08:09 PM Revision 67d73451 (ceph): cls_rgw: fix object name of objects removed on object creation
Fixes: #8972
Backport: firefly, dumpling
Reported-by: Patrycja Szabłowska <szablowska.patrycja@gmail.com>
Signed-off...
Yehuda Sadeh
06:28 PM Revision d90148da (ceph): Merge remote-tracking branch 'origin/wip-8438' into firefly-next
Backport of c5b8d8105d965da852c79add607b69d5ae79a4d4 Samuel Just
06:22 PM Revision 223a0a9e (ceph): Merge remote-tracking branch 'origin/wip-7999' into firefly-next
Backport of 830940bf242a73403ec1882a489e31f7694b7f7e Samuel Just
05:01 PM Bug #8998 (Resolved): osd: SEGV in OSD::heartbeat()
... Sage Weil
04:58 PM Bug #8997 (Can't reproduce): ceph_test_rados_watch_notify hangs
... Sage Weil
04:55 PM Bug #8996 (Resolved): "Segmentation fault" in upgrade:dumpling-x-firefly---basic-vps suite
There are lots of these errors in:
http://pulpito.front.sepia.ceph.com/teuthology-2014-08-02_08:50:33-upgrade:dumpli...
Yuri Weinstein
04:31 PM Messengers Bug #8880: msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq feature")
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-08-01_02:32:01-rados-master-testing-basic-plana/392461 Sage Weil
08:14 AM Bug #8396: osd: message delayed in Session misdirected after split
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-08-01_02:32:01-rados-master-testing-basic-plana/392256 Sage Weil
08:07 AM Bug #6003: journal Unable to read past sequence 406 ...
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-08-01_02:32:01-rados-master-testing-basic-plana/392342... Sage Weil
05:49 AM Revision 14b03bc7 (ceph): adding multi-version suite to test compatibility of clients running newer
version of ceph against servers running older version.
Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
Tamilarasi muthamizhan
04:19 AM Revision 3de7b7c5 (ceph): doc/release-notes: fix syntax error
Attempt 2...
ERROR: /srv/autobuild-ceph/gitbuilder.git/build/doc/release-notes.rst:22: Unknown target name: "leveldb...
Sage Weil
03:48 AM Revision 84bd3cd7 (ceph): unittest_crush_wrapper: fix build
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit f36cffc986c973014c89aa37ca73740b2fc194ca)
Sage Weil
02:31 AM Revision 12154c22 (ceph): Merge branch 'master' of github.com:ceph/ceph-qa-suite
Tamilarasi muthamizhan
02:27 AM Revision 3caf1e3f (ceph): Merge pull request #2188 from wonzhq/obj-mtime
osd: add local_mtime to struct object_info_t
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
02:26 AM Revision 4983aa32 (ceph): Merge pull request #2178 from dachary/wip-erasure-code-profile-default-...
erasure-code: create default profile if necessary (firefly) Sage Weil
02:24 AM Revision c95e91ef (ceph): os/KeyValueStore: clean up operator<< for KVSuperBlock
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
02:23 AM Revision e408f98d (ceph): Merge pull request #2174 from yuyuyu101/kvstore-superblock
Kvstore superblock
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
02:18 AM Revision 640c291c (ceph): Merge pull request #74 from ceph/wip_fix_fuse_upgrade-firefly
Yet anothe indentation fix Yuri Weinstein
02:15 AM Revision 0928d3e3 (ceph): Yet anothe indentation fix
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
01:28 AM Revision 9f7107b3 (ceph): Merge pull request #303 from ceph/wip-8930
Tasks are failing since using "data" pool no longer part of default install
Reviewed-by: Sage Weil <sage@redhat.com>
David Zafman
01:02 AM Revision d89a5a37 (ceph): mon: s/%%/%/
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d700076a42a5a5ebe769a8311fd3b52bf2e98cd2)
Sage Weil
01:01 AM Revision 3a05ff92 (ceph): Merge pull request #2169 from ceph/wip-double-pc
mon: s/%%/%/
Realized where these came from; it was an accident.
Sage Weil
12:50 AM Revision da0de4ee (ceph): atomic: fix read() on i386, clean up types
Among other things, fixes #8969
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from com...
Sage Weil
12:50 AM Revision 1e3b9277 (ceph): include/atomic: make 32-bit atomic64_t unsigned
This fixes
In file included from test/perf_counters.cc:19:0:
./common/perf_counters.h: In member function ‘std::pair...
Sage Weil
12:50 AM Revision 34ac3060 (ceph): Define AO_REQUIRE_CAS (fixes FTBFS on 'hppa')
to fix FTBFS due to undeclared atomic functions.
As reported
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=...
Dmitry Smirnov
12:50 AM Revision a0c12b70 (ceph): atomic_t: add atomic64_t
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
(cherry picked from commit bf3ba6001c7b4cf37edfe6551d3ef298ebcbf421)
Yehuda Sadeh
12:48 AM Revision 5a8c8c01 (ceph): test/cli-integration/rbd: fix trailing space
Newer versions of json.tool remove the trailing ' ' after the comma. Add
it back in with sed so that the .t works on...
Sage Weil
12:46 AM Revision 234b2d19 (ceph): tests: don't depend on 'data' pool in rbd test
Since we removed the default 'data' and 'metadata' pools,
tests which need a pool should create it themselves.
Signe...
John Spray
12:45 AM Revision 751258db (ceph): PGMonitor: fix bug in caculating pool avail space
Currently for pools with different rules, "ceph df" cannot report
right available space for them, respectively. For d...
Xiaoxi Chen
12:45 AM Revision 7b1546cc (ceph): mon: set min_size to data chunk count for erasure pools
Make the min_size value meaningful for erasure pools.
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked fro...
Sage Weil
12:44 AM Revision e8595879 (ceph): mon: include 'max avail' in df output
Include an estimate of the maximum writeable space for each pool. Note
that this value is a conservative estimate fo...
Sage Weil
12:44 AM Revision a5e0ebc3 (ceph): mon: right justify df values
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 2f63a309df4b7086725949bc0a532595cf927edf)
Sage Weil
12:44 AM Revision 3d578479 (ceph): mon: Fix % escaping (\% should be %%)
Clang's -Wpedantic points this out.
Signed-off-by: John Spray <john.spray@inktank.com>
(cherry picked from commit f0...
John Spray
12:08 AM Revision 6164c3ef (ceph): crush: add get_rule_weight_map
Calculate a weight map of OSDs for a given rule.
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from com...
Sage Weil
12:07 AM Revision 2f90bfdb (ceph): Fix the PG listing issue which could miss objects for EC pool (where th...
Backport: firefly
Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
(cherry picked from commit 228760ce3a7109f50fc0f8...
Guang Yang
12:06 AM Revision 61c90252 (ceph): osd/ReplicatedPG: requeue cache full waiters if no longer writeback
If the cache is full, we block some requests, and then we change the
cache_mode to something else (say, forward), the...
Sage Weil
12:06 AM Revision f701688a (ceph): osd/ReplicatedPG: fix cache full -> not full requeueing when !active
We only want to do this if is_active(). Otherwise, the normal
requeueing code will do its thing, taking care to get ...
Sage Weil
12:05 AM Revision a1ddce36 (ceph): qa/workunits/cephtool/test_daemon.sh: verify ceph -c works with daemon
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit aa9ae1f270293778aa937e7f7e4bcaee3099b9b2)
Dan Mick
12:05 AM Revision 50a1c08e (ceph): qa/workunits/cephtool/test_daemon.sh: typo
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 22d20f39b7355966554319d5a1aa888967607569)
Dan Mick
12:05 AM Revision 4cfb7dbc (ceph): qa/workunits/cephtool/test_daemon.sh: allow local ceph command
(cherry picked from commit 97a8d5a9fdbd3a25cc922c242ee57da58c57d0bc) Dan Mick
12:05 AM Revision 120b5d5a (ceph): ceph.in: Pass global args to ceph-conf for proper lookup
Fixes: #8944
Signed-off-by: Dan Mick <dan.mick@inktank.com>
(cherry picked from commit 6d89a99648630f81b85ad115fe7662...
Dan Mick
12:05 AM Revision e0c27cd7 (ceph): qa/workunits/cephtool/test.sh: test osd pool get erasure_code_profile
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit ce9f12d7a2202948532fed9da4d763ed03f6b8fa)
Con...
Sage Weil
12:03 AM Revision cf22759c (ceph): mon: OSDMonitor: add "osd pool get <pool> erasure_code_profile" command
Enable us to obtain the erasure-code-profile for a given erasure-pool.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel...
Ma Jianpeng
12:00 AM Revision b94d2f14 (ceph): osd/ReplicatedPG: observe INCOMPLETE_CLONES in is_present_clone()
We cannot assume that just because cache_mode is NONE that we will have
all clones present; check for the absense of ...
Sage Weil
12:00 AM Revision 22cc1b64 (ceph): osd/ReplicatedPG: observed INCOMPLETE_CLONES when doing clone subsets
During recovery, we can clone subsets if we know that all clones will be
present. We skip this on caching pools beca...
Sage Weil
12:00 AM Revision c42232e3 (ceph): osd/ReplicatedPG: do not complain about missing clones when INCOMPLETE_...
When scrubbing, do not complain about missing cloens when we are in a
caching mode *or* when the INCOMPLETE_CLONES fl...
Sage Weil

08/01/2014

11:59 PM Revision bde3a9f0 (ceph): osd/osd_types: add pg_pool_t FLAG_COMPLETE_CLONES
Set a flag on the pg_pool_t when we change cache_mode NONE. This
is because object promotion may promote heads witho...
Sage Weil
11:59 PM Revision f19aa8ef (ceph): qa/workunits: cephtool: adjust pool name where missing as it has changed
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 50e93c2138978f7f7c2fbafacc1611c87...
Joao Eduardo Luis
11:59 PM Revision 163440a7 (ceph): qa/workunits: cephtool: cleanup after pool creation
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 6cd345732b15e84de17d743e06bc4d855...
Joao Eduardo Luis
11:58 PM Revision c71c99ff (ceph): qa/workunits: cephtool: pool needs to be a tier to be used as such
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 704b0a33f2071eabeb8c5b000a6805ef6...
Joao Eduardo Luis
11:58 PM Revision ba0a9bf0 (ceph): qa/workunits: cephtool: test erroneous 'tier remove'
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 49db6767152092d503ccf8ead6f7cb069...
Joao Eduardo Luis
11:58 PM Revision ffec081a (ceph): qa/workunits: cephtool: test get/set on both tier and non-tier pools
Make sure gets and sets of tiering-specific variables succeed on tier
pools and fail on non-tier pools.
Signed-off-b...
Joao Eduardo Luis
11:58 PM Revision 55166c3a (ceph): qa/workunits: cephtool: split get/set on tier pools from get/set tests
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit df5944955d96c041e65964a13b802028e...
Joao Eduardo Luis
11:58 PM Revision 0425eb1d (ceph): qa/workunits: cephtool: test for 'osd pool {get,set}-quota'
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit b927c0de7d5c7a78bf3c133be52cbc1d7...
Joao Eduardo Luis
11:58 PM Revision 68197bc2 (ceph): mon: OSDMonitor: 'osd pool' - if we can set it, we must be able to get it
Add support to get the values for the following variables:
- target_max_objects
- target_max_bytes
- cache_target_...
Joao Eduardo Luis
11:57 PM Revision 01498ce8 (ceph): qa: support running under non privileged user
If the test is run against a cluster started with vstart.sh (which is
the case for make check), the --asok-does-not-n...
Loïc Dachary
11:57 PM Revision 99259dff (ceph): qa/workunits/cephtool/test.sh: sudo ceph daemon
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked from commit bcc09f93761d46202742ca85cce498a352edd494)
Sage Weil
11:57 PM Revision 4f911d45 (ceph): qa/workunits: cephtool: fix 'osd bench' test
Commit 7dc93a9651f602d9c46311524fc6b54c2f1ac595 fixed an incorrect
behavior with the OSD's 'osd bench' value hard-cap...
Joao Eduardo Luis
11:57 PM Revision bad2f513 (ceph): qa/workunits: cephtool: only run heap profiler test if tcmalloc enabled
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 4b0809a13eb363a15e52a6a57372a0a31...
Joao Eduardo Luis
11:57 PM Revision 26f9860d (ceph): qa/workunits: cephtool: set +e for the tcmalloc tests
Avoids failing the tests when tcmalloc is not present
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cher...
Joao Eduardo Luis
11:57 PM Revision a0720b60 (ceph): qa/workunits: cephtool: delete unnecessary function calls
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 67255435151627314cc2fc38732d4fb5e...
Joao Eduardo Luis
11:57 PM Revision 37322472 (ceph): qa/workunits: cephtool: disable bash debug when not running tests
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 946bd0dad4b027326b03c13405782b99e...
Joao Eduardo Luis
11:56 PM Revision 52d733c7 (ceph): qa/workunits: cephtool: allow running individual tests
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 5d26575ef2d31d745ec4aa69ca1501cd7...
Joao Eduardo Luis
11:56 PM Revision 96e58d6d (ceph): qa/workunits: cephtool: cleanup state after erasure-code-profile test
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit f4184086d0d647e064e34308f678ef889...
Joao Eduardo Luis
11:56 PM Revision 4cd5d179 (ceph): qa/workunits: cephtool: add/remove comments
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 780424df3a107c7da57fc28d64f9e7a4b...
Joao Eduardo Luis
11:56 PM Revision 5bcf690b (ceph): qa/workunits: cephtool: split into properly indented functions
The test was a big sequence of commands being run and it has been growing
organically for a while, even though it has...
Joao Eduardo Luis
11:55 PM Revision 2af75180 (ceph): qa/workunits: cephtool: move test line to where it's more appropriate
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 04658b7b2b5f98ae81ffb3f77303745e6...
Joao Eduardo Luis
11:55 PM Revision 5bf9e3c7 (ceph): qa/workunits: cephtool: split into functions
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit db6cc133ba4bb38b3c11eb835fd3983dc...
Joao Eduardo Luis
11:54 PM Revision c898c973 (ceph): mon: test that pools used in tiers cannot be removed
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 39a4b78177cb9896ff35ab05bcf8774bfc934f3a)
Loïc Dachary
11:54 PM Revision 5b726742 (ceph): qa/workunits/cephtool: test setting options using SI units
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 38405d3554dfb0caf2b0a2c010b95a61b...
Joao Eduardo Luis
11:48 PM Revision 5ed92f67 (ceph): Merge pull request #73 from ceph/wip_fix_fuse_upgrade-firefly
Fixed typos
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
11:46 PM Revision aacf67d1 (ceph): mon: OSDMonitor: be scary about inconsistent pool tier ids
We may not crash your cluster, but you'll know that this is not something
that should have happened. Big letters mak...
Joao Eduardo Luis
11:46 PM Revision ea1a3170 (ceph): Fixed typos
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
11:46 PM Revision af59cbf5 (ceph): osd: pg_pool_t: clear tunables on clear_tier()
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 64bdf6c92bc44adad5a49b9dc4f674789...
Joao Eduardo Luis
11:46 PM Revision 8f788e14 (ceph): mon: OSDMonitor: limit tier-specific pool set/get on non-tier pools
Fixes: 8696
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit f131dfbaedf6f451572e...
Joao Eduardo Luis
11:40 PM Revision 4b768a8f (ceph): mon/OSDMonitor: improve no-op cache_mode set check
If we have a pending pool value but the cache_mode hasn't changed, this is
still a no-op (and we don't need to block)...
Sage Weil
11:40 PM Revision c3f40329 (ceph): mon: OSDMonitor: disallow nonsensical cache-mode transitions
Fixes: 8155
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit fd970bbc95d89bf66c95...
Joao Eduardo Luis
11:40 PM Revision d9aacd96 (ceph): mon: OSDMonitor: return immediately if 'osd tier cache-mode' is a no-op
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit d01aa5bff30441eec1ffaa3e59a21187f...
Joao Eduardo Luis
11:39 PM Revision 981add9c (ceph): osd/ReplicatedPG: debug obc locks
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 356af4bf46d6387e2f1a59646548f9a77e49e5f8)
Sage Weil
11:39 PM Revision e32e2559 (ceph): osd/ReplicatedPG: greedily take write_lock for copyfrom finish, snapdir
In the cases where we are taking a write lock and are careful
enough that we know we should succeed (i.e, we assert(g...
Sage Weil
11:39 PM Revision 68651e91 (ceph): osd: allow greedy get_write() for ObjectContext locks
There are several lockers that need to take a write lock
because there is an operation that is already in progress an...
Sage Weil
11:06 PM Revision 54af8b2b (ceph): Tasks are failing since using "data" pool no longer part of default ins...
Create a pool specifically for each task
Fixes: 8930
Signed-off-by: David Zafman <david.zafman@inktank.com>
David Zafman
10:57 PM Revision 4aa2b873 (ceph): qa/workunits/rest/test.py: make osd create test idempotent
Avoid possibility that we create multiple OSDs do to retries by passing in
the optional uuid arg. (A stray osd id wi...
Sage Weil
10:57 PM Revision a2681c93 (ceph): qa/workunits/rest/test.py: make osd create test idempotent
Avoid possibility that we create multiple OSDs do to retries by passing in
the optional uuid arg. (A stray osd id wi...
Sage Weil
10:53 PM Revision 7a58da53 (ceph): enforce rados put aligment
Signed-off-by: Lluis Pamies-Juarez <lluis.pamies-juarez@hgst.com>
(cherry picked from commit 304b08a23a3db57010078046...
Lluis PJ
10:50 PM Revision 935e75a8 (ceph): mon: Monitor: suicide on start if mon has been removed from monmap
If the monitor has been marked as having been part of an existing quorum
and is no longer in the monmap, then it is s...
Joao Eduardo Luis
10:37 PM Revision 79d1aff1 (ceph): Merge branch 'wip-cache-second'
Reviewed-by: Samuel Just <sam.just@inktank.com> Sage Weil
10:37 PM Revision 0ed3adc1 (ceph): osd: promotion on 2nd read for cache tiering
Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com> Zhiqiang Wang
10:37 PM Revision 1417eded (ceph): ceph_test_rados_api_tier: test promote-on-second-read behavior
Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Zhiqiang Wang
10:33 PM Revision 7ae15549 (ceph): use llrintl when converting double to micro
This avoids rounding error (noticeable on i386).
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
(cherr...
Sage Weil
09:25 PM Bug #8776 (Won't Fix): osd: runaway memory on dumpling
this is a result of a very large omap object and us building a transaction to delete the keys. the problem is the bi... Sage Weil
09:57 AM Bug #8776: osd: runaway memory on dumpling
Argh, it's building up a leveldb operation to atomically remove all of the keys associated with the object. I *think... Samuel Just
08:31 PM Revision 52c2182f (ceph): Merge pull request #2183 from majianpeng/master
ECBackend: Using ROUND_UP_TO to refactor function get_recovery_chunk_size()
Reviewed-by: Samuel Just <sam.just@inkta...
Samuel Just
08:30 PM Revision dad092c6 (ceph): Merge pull request #2175 from majianpeng/fix1
ReplicatedPG: For async-read, set the real result after completing read.
Reviewed-by: Samuel Just <sam.just@inktank....
Samuel Just
08:27 PM Revision 3440d664 (ceph): Expand '~' in ssh key paths
... since paramiko doesn't bother to do this
Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
Zack Cerza
07:46 PM Revision ceb37c62 (ceph): Merge branch 'master' of github.com:ceph/ceph-qa-suite
Tamilarasi muthamizhan
07:25 PM Revision f4c953c1 (ceph): Merge pull request #301 from ceph/wip-nuke-vpm
nuke: allow nuking vpm hosts Zack Cerza
06:26 PM Bug #8930 (Resolved): osd: test unable to produce unfound objects
David Zafman
04:07 PM Bug #8930 (Fix Under Review): osd: test unable to produce unfound objects
David Zafman
09:41 AM Bug #8930: osd: test unable to produce unfound objects
David Zafman
05:00 PM Revision f752ff49 (ceph): Merge pull request #2180 from ceph/wip-ec-isa
osd: add support for intel ISA-L EC library Sage Weil
04:57 PM Revision f335c73b (ceph): Merge pull request #2172 from ceph/wip-8714
osd: prevent old clients from using tiered pools
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
04:04 PM Revision 30e61373 (ceph): Merge pull request #72 from ceph/wip_fix_fuse_upgrade-firefly
Fixed to run blogbench.sh on client:0 Sage Weil
03:56 PM devops Bug #8849 (Resolved): rpm restarts daemons on upgrade
already backported, commit:e75dd2e4b7adb65c2de84e633efcd6c19a6e457b and ^ Sage Weil
03:55 PM Bug #8728 (Resolved): rest/test.py osd create not idempotent
Sage Weil
03:54 PM Bug #8670: Cache tiering parameters can not be displayed for a pool
non trivial to backport.. need to get all the rados test refactoring, too! Sage Weil
03:51 PM CephFS Bug #8622 (Resolved): erasure-code: rados command does not enforce alignement constraints
commit:7a58da53ebfcaaf385c21403b654d1d2f1508e1a Sage Weil
03:48 PM Bug #6789 (Resolved): cannot remove the leader when there only are two monitors
Sage Weil
03:39 PM Bug #8944 (Pending Backport): Ceph daemon bad asok used in connection with cluster
Sage Weil
03:37 PM Bug #8714 (Pending Backport): we do not block old clients from breaking cache pools
Sage Weil
03:35 PM Feature #8674 (Pending Backport): osd: cache tier: avoid promotion on first read
commit:79d1aff1821bc9f21477636df4d0d4e57f2cd008 Sage Weil
03:32 PM rgw Bug #8937 (Pending Backport): rgw: broken large(-ish) objects
Sage Weil
03:17 PM Revision e08d1a49 (ceph): Fixed to run blogbench.sh on client:0
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
03:05 PM Documentation #8995 (Resolved): Preflight Checklist Clarifications
There are several small clarifications that can be made to the Ceph Preflight Checklist to help new users try out Cep... Christopher Hertel
02:44 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
No need to do that just yet. I now fully understand the problem and working on a proper fix that I'd like you to tes... Ilya Dryomov
02:37 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
I have done some testing and I am seeing the same thing as Eric. With the deadlock-bad kernel I hit the deadlock iss... Greg Wilson
02:16 PM Revision c86a9877 (ceph): valgrind: ignore ec plugin factory leaks
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 7c3fc404d6a84c6601400e364bd98ea89df13922)
Sage Weil
02:13 PM Revision 5fa79fa9 (ceph): use notcmalloc builder for valgrind leak checking
This reverts 453d72b294af6d3db1ae7a212ba9b43209d69f26. It turns out that
tcmalloc defeats the leak checking in valgr...
Sage Weil
02:11 PM Revision f5fc2436 (ceph): msg/SimpleMessenger: drop local_conneciton priv link on shutdwon
This breaks ref cycles between the local_connection and session, and let's
us drop the explicit set_priv() calls in O...
Sage Weil
02:08 PM Revision cd64a63d (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
02:08 PM Revision a00777f4 (ceph): Merge pull request #2190 from ceph/wip-osd-leaks
osd: do not leak Session* ref in _send_boot()
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
02:06 PM Bug #8625: EC pool - OSD creates an empty file for op with 'create 0~0, writefull 0~xxx, setxattr...
Making it not an rgw bug. Samuel Just
02:06 PM Bug #8625: EC pool - OSD creates an empty file for op with 'create 0~0, writefull 0~xxx, setxattr...
wip-8625, versioning should never be necessary after a create (it will be necessary before the create if the object a... Samuel Just
09:53 AM Bug #8625: EC pool - OSD creates an empty file for op with 'create 0~0, writefull 0~xxx, setxattr...
It's the create 0~0 followed by a writefull. Arguably, we still shouldn't version the object, I'll take a look. Samuel Just
01:02 PM Fix #8993 (Closed): osd_pool_default_pgp_num woes
When setting osd_pool_default_pgp_num and not osd_pool_default_pg_num you can create pools with more pgp than pg.
...
Alexandre Marangone
12:57 PM devops Bug #8893 (Fix Under Review): ceph-deploy install command on centos 6.5 reports exception
PR opened https://github.com/ceph/ceph-deploy/pull/226 Alfredo Deza
06:51 AM devops Bug #8893 (In Progress): ceph-deploy install command on centos 6.5 reports exception
Alfredo Deza
12:11 PM Revision fc2d18bb (ceph): Merge pull request #2182 from ceph/wip-round
use llrintl when converting double to micro
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
09:15 AM rbd Bug #8416 (Closed): Client Crash when try to map a volume (ubuntu)
OK, I'm going to assume this was indeed the missing features handling bug. I looked into it, it was introduced in 3.... Ilya Dryomov
08:23 AM Bug #8989 (Rejected): Failed running iogen.sh in upgrade:firefly-firefly-testing-basic-vps suite
It was a test mis-configuration. When we added a new client to run workload on, we had to be more specific about on ... Yuri Weinstein
08:09 AM Revision 13b9dc70 (ceph): osd: add local_mtime to struct object_info_t
This fixes a bug when the time of the OSDs and clients are not
synchronized (especially when client is ahead of OSD),...
Zhiqiang Wang
07:10 AM Bug #8717 (Resolved): teuthology: valgrind leak checks broken for osd (at least)
Sage Weil
05:57 AM Bug #8601: erasure-code: default profile does not exist after upgrade
... Loïc Dachary
04:22 AM Revision 03be707d (ceph): kcephfs/thrash: add standby mds
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision c2fc1a94 (ceph): Merge branch 'wip-rocksdb'
Sage Weil
04:11 AM Revision 213e0acf (ceph): debian, rpm: build rocksdb statically if we can
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision c574e653 (ceph): autogen.sh: debug with -x
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision 57fd60cd (ceph): rocksdb: -Wno-portability
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision eb1cd780 (ceph): rocksdb: update submodule makefile
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> Josh Durgin
04:11 AM Revision bcebf9c0 (ceph): do_autogen.sh: build with static librocksdb
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision 33293523 (ceph): rocksdb: require less shiny autoconf
precise has 2.59
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
04:11 AM Revision 5502dd3e (ceph): configure: verify c++11 is present for rocksdb
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision 7500b890 (ceph): Makefile: include rocksdb in dist tarball always
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision 8fcbb659 (ceph): configure: enable static rocksdb if check and c++11
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision da5edb81 (ceph): add rocksdb bz2 dep
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:11 AM Revision a497a3b4 (ceph): debian, rpm: build require bzip2 dev
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
04:06 AM Revision 2e549b4f (ceph): unittest_erasure_code_[plugin_]isa: conditionally compile
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
03:59 AM Revision 04037ef9 (ceph): osd: do not leak Session* ref in _send_boot()
The get_priv() call returns a ref; make sure we drop it if it exists.
This doesn't happen on every run because usuall...
Sage Weil
03:59 AM Revision 53256d33 (ceph): Merge pull request #71 from ceph/wip_fix_fuse_upgrade-firefly
Fixed miss-config for ceph-fuse Sage Weil
03:57 AM Revision f789bcd9 (ceph): Fixed ceph-fuse: [client.0] syntax
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
02:23 AM Feature #8992 (New): Uniqueness between two or more CRUSH ruleset choose statements
Assuming that ceph-node1 is in default root, when we define and assign following crush rule:... Szymon Zacher
01:44 AM Bug #8641: Cache tiering agent cannot flush or evict objects during the benchmark
In my opinion problem affect also cache_min_evict_age cache_min_flush_age and others. It's impossible to force ceph c... Szymon Zacher
01:34 AM Revision 26e9105f (ceph): Fixed miss-config for ceph-fuse
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
01:11 AM Revision 9b05a87e (ceph): Merge pull request #2157 from ceph/wip-8937
rgw: call processor->handle_data() again if needed
Reviewed-by: Sage Weil <sage@redhat.com>
Yehuda Sadeh
12:46 AM Revision d00f23be (ceph): erasure-code/isa: fix signed/unsigned comparison
warning: erasure-code/isa/ErasureCodeIsa.cc:313: comparison between signed and unsigned integer expressions
Signed-o...
Sage Weil
12:45 AM Revision 8a193ab0 (ceph): erasure-code/isa: don't use typename outside of template
error: erasure-code/isa/ErasureCodeIsa.h:141: using 'typename' outside of template
Signed-off-by: Sage Weil <sage@re...
Sage Weil
12:40 AM CephFS Bug #8962: kcephfs: client does not release revoked cap
... Zheng Yan
12:32 AM Revision 92d0bbe1 (ceph): ECBackend: Using ROUND_UP_TO to refactor function get_recovery_chunk_si...
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com> Ma Jianpeng
12:25 AM Revision e4ed2cb2 (ceph): ReplicatedPG: For async-read, set the real result after completing read.
When reading an object from replicated pool, ceph uses sync mode,
so it can set the results in execute_ctx correctly....
Ma Jianpeng

07/31/2014

11:19 PM Revision 6e52efab (ceph): doc/release-notes: tweak quoting
ERROR: /srv/autobuild-ceph/gitbuilder.git/build/doc/release-notes.rst:22: Unknown target name: "leveldb".
Signed-off...
Sage Weil
10:39 PM Revision 80911736 (ceph): use llrintl when converting double to micro
This avoids rounding error (noticeable on i386).
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
10:18 PM Revision d8b291f3 (ceph): configure: check for 'better' yasm that can build ISA-L
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
09:04 PM rgw Bug #8972 (Pending Backport): rgw: bucket index log wrong object name in multipart completion
Sage Weil
09:31 AM rgw Bug #8972 (Fix Under Review): rgw: bucket index log wrong object name in multipart completion
Sage Weil
08:54 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
Zheng Yan wrote:
> Sage Weil wrote:
> > Zheng Yan wrote:
> > > no clue what happened. please dump the mds cache wh...
Sage Weil
07:32 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
Sage Weil wrote:
> Zheng Yan wrote:
> > no clue what happened. please dump the mds cache when it happens next time
...
Zheng Yan
10:11 AM CephFS Bug #8962: kcephfs: client does not release revoked cap
Zheng Yan wrote:
> no clue what happened. please dump the mds cache when it happens next time
We have a dump, act...
Sage Weil
08:48 PM rgw Bug #8991 (Resolved): rgw: RGWRados::list_bi_log_entries() doesn't clear list
... Yehuda Sadeh
08:43 PM Revision 1520b471 (ceph): erasure-code/isa/Makefile: add missing \, and missing headers
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:43 PM Revision e1f32d9f (ceph): yasm-wrapper: fix -f ...
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:42 PM Revision 1670fd63 (ceph): yasm-wrapper: turn -I foo into -i foo
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
08:41 PM Revision 6d95797c (ceph): yasm-wrapper: echo original args
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:53 PM Revision d76af5a6 (ceph): Merge branch 'master' of github.com:ceph/ceph-qa-suite
Tamilarasi muthamizhan
06:40 PM Revision b7d0017d (ceph): EC: add plugin for Intel ISA-L library
Andreas Peters
06:28 PM Revision fe99b253 (ceph): Merge pull request #2179 from ceph/wip-8982
osd: fix cache pools without hit_sets configured
Reviewed-by: Samuel Just <sam.just@inktank.com>
Samuel Just
06:03 PM Revision ce4e559d (ceph): vstart.sh: limit open files
Most users only get 1024 fds.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
06:03 PM Revision caf554bf (ceph): osd/ReplicatedPG: improve agent_choose_mode args
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
06:03 PM Revision 383536a2 (ceph): mon/OSDMonitor: warn when cache pools do not have hit_sets configured
Give users a clue when cache pools are enabled but the hit_set is not
configured. Note that technically this will wo...
Sage Weil
06:03 PM Revision 5d1c76f6 (ceph): osd/ReplicatedPG: check agent_mode if agent is enabled but hit_sets aren't
It is probably not a good idea to try to run the tiering agent without a
hit_set to inform its actions, but it is tec...
Sage Weil
06:03 PM Revision ea4996d3 (ceph): osd/ReplicatedPG: evict blindly if there is no hit_set
If there is no hit set for a PG, blindly evict objects. This avoids an
assert(hit_set) in agent_estimate_atime_temp(...
Sage Weil
04:55 PM Revision f77449cb (ceph): Merge pull request #2168 from ceph/wip-8972
cls_rgw: fix object name of objects removed on object creation
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
03:52 PM Bug #8977: osd: didn't discard sub_op_reply from previous interval?
Added some debugging to dump the OpWQ queue information if there are stale ops, running in loop. Samuel Just
12:53 PM Bug #8977: osd: didn't discard sub_op_reply from previous interval?
2014-07-30 10:40:58.317063 7fc2164da700 0 log [WRN] : slow request 960.196157 seconds old, received at 2014-07-30 10... Samuel Just
02:35 PM Bug #8989 (Rejected): Failed running iogen.sh in upgrade:firefly-firefly-testing-basic-vps suite
There majority of failures related to this in this run: http://pulpito.front.sepia.ceph.com/teuthology-2014-07-30_12:... Yuri Weinstein
12:59 PM Revision f3d1f396 (ceph): Merge pull request #2173 from ceph/wip-8969
fix i386 atomics
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
12:52 PM Feature #131 (In Progress): bring wireshark plugin is up to date
Sage Weil
12:51 PM Documentation #7 (Resolved): Document Monitor Commands
ceph -h Sage Weil
12:47 PM Revision 2fff5f03 (ceph): common: s/stringstream/ostream/ in str_map
There is no need to specialize more than ostream : it only makes it
impossible to use cerr or cout as a parameter to ...
Loïc Dachary
12:47 PM Revision b6d8feab (ceph): erasure-code: create default profile if necessary
After an upgrade to firefly, the existing Ceph clusters do not have the
default erasure code profile. Although it may...
Loïc Dachary
12:16 PM Revision 70fede92 (ceph): Merge pull request #1990 from dachary/wip-erasure-code-profile-default
erasure-code: create default profile if necessary
Reviewed-by: Joao Eduardo Luis <jluis@redhat.com>
Loïc Dachary
11:29 AM rgw Bug #8988 (Resolved): AssertionError(s) in upgrade:firefly-x:stress-split-next---basic-plana
"Related issue":http://tracker.ceph.com/issues/9100
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-201...
Yuri Weinstein
11:25 AM Bug #8982 (Pending Backport): cache pool osds crashing when data is evicting to underlying storag...
Sage Weil
11:14 AM Bug #8982 (Fix Under Review): cache pool osds crashing when data is evicting to underlying storag...
Sage Weil
08:47 AM Bug #8982 (In Progress): cache pool osds crashing when data is evicting to underlying storage pool
Sage Weil
07:36 AM Bug #8982 (Resolved): cache pool osds crashing when data is evicting to underlying storage pool
We have a erasure coded pool 'ecdata' and a replicated(size=3) pool 'cache' acting as writeback cache upon it.
When...
Kenneth Waegeman
11:17 AM Bug #8969 (Pending Backport): PerfCounters.SinglePerfCounters failure on i386
Sage Weil
09:48 AM rgw Feature #8987 (New): rgw: data sync for multipart upload
Yehuda Sadeh
09:46 AM Bug #8986 (Duplicate): "[WRN] map e62 wrongly marked me down" in upgrade:dumpling-x-firefly---bas...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-07-30_13:00:44-upgrade:dumpling-x-firefly---basic-vps... Yuri Weinstein
09:43 AM Bug #8985: "[WRN] map e9 wrongly marked me down" in upgrade:dumpling-x-firefly---basic-vps suite
... Yuri Weinstein
09:42 AM Bug #8985 (Resolved): "[WRN] map e9 wrongly marked me down" in upgrade:dumpling-x-firefly---basic...
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-07-30_13:00:44-upgrade:dumpling-x-firefly---basic-vps... Yuri Weinstein
09:35 AM Bug #8970 (Won't Fix): Injectargs - inconsistent parsing of bool values
these will also work:
--my-boolean-option=0
--my-boolean-option=false
but you're right, the others won't, be...
Sage Weil
09:33 AM Feature #8973: Add support for collecting usage information by namespace
We decided not to do this when designing namespaces because we wanted namespaces to scale independnetly of the size o... Sage Weil
08:49 AM Bug #8947 (Duplicate): Writing rados objects with max objects set for cache pool crashed osd
Oh, i see it now. This is a dup of #8982. Sage Weil
08:29 AM RADOS Support #8600: MON crashes on new crushmap injection
In addition to the choose vs. chooseleaf issue that Joao is mentioning here, we have also seen problems when min_size... Henning Stener
08:13 AM Bug #8966: ceph.conf "osd pool default size = 2" not working
Then the documentation (http://ceph.com/docs/master/start/quick-ceph-deploy/) on point 2 should be updated.... Christoph Pedro
07:58 AM RADOS Bug #8984 (Won't Fix): creating erasure-code pool when not having a root item default
When creating a EC pool:
> ceph osd pool create poolio 128 128 erasure profile15
It returns
> Error ENOENT: root ...
Kenneth Waegeman
07:46 AM Bug #8983 (Resolved): rados bench -b option does not take orders of magnitude (k,M,..) but also d...
When running this:
> rados -p <pool> bench 1000 write -t 10 -b 4M
It runs with -b 4 instead of expected
> rados -...
Kenneth Waegeman
06:04 AM Bug #8601: erasure-code: default profile does not exist after upgrade
Apparently having an EC pool is still sufficient to prevent kernel clients from mounting, so I don't think we can bac... Greg Farnum
05:52 AM Bug #8601: erasure-code: default profile does not exist after upgrade
"firefly backport":https://github.com/ceph/ceph/pull/2178 Loïc Dachary
05:16 AM Bug #8601 (Pending Backport): erasure-code: default profile does not exist after upgrade
Loïc Dachary
04:35 AM Revision 9e6a3e79 (ceph): ceph_manager: increase osd revival timeout from 75 -> 150 seconds
Saw this fail on a firefly run just bc of valgrind slowness. Make the
timeout higher!
Signed-off-by: Sage Weil <sag...
Sage Weil
04:34 AM Revision f84458bd (ceph): ceph_manager: increase osd revival timeout from 75 -> 150 seconds
Saw this fail on a firefly run just bc of valgrind slowness. Make the
timeout higher!
Signed-off-by: Sage Weil <sag...
Sage Weil
03:57 AM Revision 5c6a0512 (ceph): Merge pull request #2162 from ceph/wip-8586
rgw: fix crash in swift CORS preflight request
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
02:53 AM Linux kernel client Bug #8979 (Resolved): GPF kernel panics - auth?
From James Eckersall, "GPF kernel panics" on ceph-users.
I've had a fun time with ceph this week.
We have a clust...
Ilya Dryomov
02:14 AM Revision 51c1f2a8 (ceph): FileStore: Add omap_backend to "<<" operator
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
02:13 AM Revision 7faed148 (ceph): Add superblock to KeyValueStore
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
01:51 AM Revision b879e746 (ceph): KeyValueStore: use generic KeyValueDB::create()
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
01:14 AM Revision 7c3fc404 (ceph): valgrind: ignore ec plugin factory leaks
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:50 AM Revision 9df9d280 (ceph): mon/OSDMonitor: fix i386 floating point rounding error
This would round .123 to .122999 on i386; now it does not.
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil

07/30/2014

11:57 PM Revision 9eda0bbe (ceph): Merge pull request #2171 from ceph/wip-cli-integration
fix rbd cli-integratoin tests
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
10:59 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
no clue what happened. please dump the mds cache when it happens next time Zheng Yan
07:01 AM CephFS Bug #8962: kcephfs: client does not release revoked cap
and the code that did it is in teuthology.git/teuthology/misc.py:... Sage Weil
07:00 AM CephFS Bug #8962: kcephfs: client does not release revoked cap
here is the final state of the directory:... Sage Weil
10:37 PM Revision 9e2c8282 (ceph): Merge pull request #300 from ceph/wip-mpi-vers
task/mpi: Explicit check for version mismatch John Spray
10:25 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
The deadlock-bad kernel showed the error after a few minutes of running multiple dd writes to rbd device. Here is one... Eric Eastman
11:33 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
All,
Can you try and confirm that deadlock-bad fails and deadlock-good works for you?
deadlock-bad:
http://g...
Ilya Dryomov
05:18 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Update: At this point I'm almost certain this is not an rbd/ceph problem. Trying to track down the exact culprit. Ilya Dryomov
04:59 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
I can reproduce this with 100% certainty now on Trusty, 3.15.6-031506-generic.
Running:
bonnie++ -n 512
agai...
Karl Austin
09:57 PM Bug #8752 (New): firefly: scrub/repair stat mismatch
This problem manifests only on caching pools.
I have two EC pools with the following settings:...
Dmitry Smirnov
09:56 PM Revision 02dd04ac (ceph): Merge pull request #2164 from ceph/wip-8944
ceph.in: Pass global args to ceph-conf for proper lookup
Backport: firefly
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
09:52 PM Revision 22d20f39 (ceph): qa/workunits/cephtool/test_daemon.sh: typo
Signed-off-by: Dan Mick <dan.mick@inktank.com> Dan Mick
09:52 PM Revision aa9ae1f2 (ceph): qa/workunits/cephtool/test_daemon.sh: verify ceph -c works with daemon
Signed-off-by: Dan Mick <dan.mick@inktank.com> Dan Mick
09:52 PM Revision 97a8d5a9 (ceph): qa/workunits/cephtool/test_daemon.sh: allow local ceph command
Dan Mick
09:52 PM Revision 96863128 (ceph): atomic: fix read() on i386, clean up types
Among other things, fixes #8969
Backport: firefly
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
09:51 PM Revision 6d89a996 (ceph): ceph.in: Pass global args to ceph-conf for proper lookup
Fixes: #8944
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Dan Mick
09:44 PM Bug #8229 (Closed): 0.80~rc1: OSD crash (domino effect)
Closing: nothing left to track here; did not have this problem with 0.80.4. Dmitry Smirnov
09:42 PM Bug #8978 (Can't reproduce): ceph ping not working as expected
Reading the doc: http://ceph.com/docs/master/rados/troubleshooting/troubleshooting-mon/
I came across command: cep...
Eric Eastman
09:26 PM Bug #8977 (Can't reproduce): osd: didn't discard sub_op_reply from previous interval?
/a/teuthology-2014-07-29_02:30:02-rados-firefly-distro-basic-plana/384397
an op gets stuck in limbo because we are...
Sage Weil
08:57 PM Revision 0190df53 (ceph): osd: prevent old clients from using tiered pools
If the client is old and doesn't understand tiering, don't let them use a
tiered pool. Reply with EOPNOTSUPP.
Fixes...
Sage Weil
08:54 PM rgw Bug #8586 (Pending Backport): Missing Swift API Header causes RadosGW to segfault
Sage Weil
08:41 PM Revision 605064dc (ceph): test/cli-integration/rbd: fix trailing space
Newer versions of json.tool remove the trailing ' ' after the comma. Add
it back in with sed so that the .t works on...
Sage Weil
08:31 PM Revision 094cd5b3 (ceph): Merge pull request #302 from ceph/wip-8850
tell us if you are really not running any more Zack Cerza
08:28 PM Revision 4686115c (ceph): add another call to use "ps aux"
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
08:17 PM Revision ec342d81 (ceph): tell us if you are really not running any more
Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com> Alfredo Deza
07:26 PM Revision d700076a (ceph): mon: s/%%/%/
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:07 PM Revision bf232ba8 (ceph): Merge pull request #2163 from ceph/wip-filestore-omap
clean up KeyValueDB instantiation
Reviewed-by: Haomai Wang <haomaiwang@gmail.com>
Reviewed-by: Josh Durgin <josh.dur...
Sage Weil
06:53 PM Revision 0f8929a6 (ceph): cls_rgw: fix object name of objects removed on object creation
Fixes: #8972
Backport: firefly, dumpling
Reported-by: Patrycja Szabłowska <szablowska.patrycja@gmail.com>
Signed-off...
Yehuda Sadeh
06:12 PM Revision 1c7c2eb4 (ceph): nuke: allow nuking vpm hosts
Skip the console check, though.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:57 PM devops Bug #8976: httpd on RHEL7 (RHEL repo) incompatible with mod_fastcgi (ceph repo)
Also, when trying to enable the httpd ceph pkg with systemctl:
systemctl enable httpd
httpd.service is not a nat...
Marcelo Giles
05:22 PM devops Bug #8976 (Resolved): httpd on RHEL7 (RHEL repo) incompatible with mod_fastcgi (ceph repo)
On a RHEL7 system
yum install httpd mod_fastcgi
systemctl start httpd
Apache fails to start with the folowin...
Marcelo Giles
05:52 PM Revision 061c8e93 (ceph): librbd: enable rbd cache by default; writethrough until flush
This is recommended and safe. Avoid some user confusion.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
05:12 PM Bug #8947 (Need More Info): Writing rados objects with max objects set for cache pool crashed osd
can you attach the complete logs? all three osds claim to have hit an assert, but the assert message isn't in the lo... Sage Weil
05:12 PM Revision 4e1405e7 (ceph): erasure-code: create default profile if necessary
After an upgrade to firefly, the existing Ceph clusters do not have the
default erasure code profile. Although it may...
Loïc Dachary
04:59 PM rbd Bug #8920 (Pending Backport): rbd/singleton/{all/formatted-output.yaml} fails on trusty due to wh...
Sage Weil
01:43 PM rbd Bug #8920 (Fix Under Review): rbd/singleton/{all/formatted-output.yaml} fails on trusty due to wh...
Sage Weil
04:36 PM Bug #8776: osd: runaway memory on dumpling
it's all here:... Sage Weil
02:49 PM Bug #8969 (Fix Under Review): PerfCounters.SinglePerfCounters failure on i386
Sage Weil
10:31 AM Bug #8969 (Resolved): PerfCounters.SinglePerfCounters failure on i386
[ RUN ] PerfCounters.SinglePerfCounters
test/perf_counters.cc:111: Failure
Value of: msg
Actual: "{"test_perfcount...
Sage Weil
02:29 PM Bug #8628 (Resolved): Bad ceph_osd_op.extent union access in ReplicatedPG::do_osd_ops
commit:58212b1245373b6f015cbff11844d33a900bf3cb Sage Weil
02:19 PM Bug #8628 (Rejected): Bad ceph_osd_op.extent union access in ReplicatedPG::do_osd_ops
ceph_osd_op_uses_extent(op.op) guards the references ot the extent view of the union Sage Weil
02:13 PM Bug #8717: teuthology: valgrind leak checks broken for osd (at least)
Sage Weil
02:12 PM Bug #8717 (Resolved): teuthology: valgrind leak checks broken for osd (at least)
Sage Weil
02:12 PM Bug #8777 (Can't reproduce): osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == log...
Sage Weil
02:11 PM Bug #8595: osd: client op blocks until backfill starts (dumpling)
Sage Weil
02:02 PM Bug #8595 (In Progress): osd: client op blocks until backfill starts (dumpling)
Sage Weil
01:59 PM Bug #8714 (Fix Under Review): we do not block old clients from breaking cache pools
https://github.com/ceph/ceph/pull/2172 Sage Weil
01:55 PM Revision ec2f9490 (ceph): kcephfs: thrash mds too
And shorten names.
Signed-off-by: Sage Weil <sage@redhat.com>
Sage Weil
01:46 PM Bug #8974 (Can't reproduce): osd crashed with merge_log assert due to removal of isds
Even I got same asserts in one of the osds, when removed one osd from each node in a ceph cluster of 3 osd nodes ( 5 ... Sahana Lokeshappa
01:31 PM devops Bug #8850: ceph-deploy tests fail during tar due to file changed; incomplete shutdown?
an initial take on getting more information on what is going on:

https://github.com/ceph/teuthology/pull/302/files
Alfredo Deza
12:47 PM devops Bug #8850: ceph-deploy tests fail during tar due to file changed; incomplete shutdown?
I initially thought that the ceph daemon was still running but according to upstart docs, this output:... Alfredo Deza
01:29 PM Revision 49181796 (ceph): task/mpi: Explicit check for version mismatch
Instead of proceeding to have mpiexec fail or hang,
do an up-front check that the MPI version is the same
on all of t...
John Spray
11:53 AM Feature #8973 (New): Add support for collecting usage information by namespace
As of now there is no simple way to determine how much data is being used by a particular namespace. Customers curren... Tyler Brekke
11:36 AM rgw Bug #8972 (Resolved): rgw: bucket index log wrong object name in multipart completion
When completing a multipart upload operation, when removing the parts from the index the entries that are logged in t... Yehuda Sadeh
11:27 AM rgw Bug #8971 (Duplicate): rgw: s3 test failures with civetweb
teuthology logs are copied to ubuntu@mira023.front.sepia.ceph.com:/home/ubuntu/civetweb_s3
config.yaml:...
Tamilarasi muthamizhan
10:35 AM Bug #8970 (Won't Fix): Injectargs - inconsistent parsing of bool values
Hi all,
ceph version 0.80.1 (a38fe1169b6d2ac98b427334c12d7cf81f809b74) on Ubuntu 14.04 LTS
This is how I am able ...
Peter Vilhan
10:19 AM Feature #8960 (Fix Under Review): filestore: store backend type persisently
https://github.com/ceph/ceph/pull/2163 Sage Weil
10:17 AM Bug #8601: erasure-code: default profile does not exist after upgrade
"rebased and repushed":https://github.com/ceph/ceph/pull/1990 Loïc Dachary
09:37 AM Bug #8966 (Closed): ceph.conf "osd pool default size = 2" not working
the config option needs to go in the [global] section, not [default] (which is never used for anything) Sage Weil
04:31 AM Bug #8966: ceph.conf "osd pool default size = 2" not working
Recognized the failure with the command "ceph osd dump". There the pools had always the size 3 (default). Christoph Pedro
04:29 AM Bug #8966 (Closed): ceph.conf "osd pool default size = 2" not working
Version
ceph-deploy: 1.5.9
ceph 0.80.5
Ceph.config:...
Christoph Pedro
09:03 AM Documentation #8875: `ceph-deploy new` needs to be called for every node, not just the admin one
It appears I was able to get further this time, the steps are below.
Key difference is, when I did ceph-deploy new I...
Bobby Yakov
06:20 AM Documentation #8875: `ceph-deploy new` needs to be called for every node, not just the admin one
Hi Alfredo,
Nodes were cleaned out, will re-run install today and get you the log files.
In the mean time, it appea...
Bobby Yakov
06:17 AM Bug #8922: ceph-deploy mon create fails to create additional monitoring nodes.
ceph-deploy new cwtcph001
ceph-deploy install cwtcph001 cwtcph002 cwtcph003
ceph-deploy mon create cwtcph001 cwtcph...
Bobby Yakov
06:16 AM Revision 5f65b4db (ceph): os/FileJournal: When dump journal, using correctly seq avoid misjudging...
In func FileJournal::dump, it always using seq=0 as last-seq and it can
misjudge the journal corrupt.
Signed-off-by:...
Ma Jianpeng
05:32 AM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
Ilya Dryomov wrote:
> What do you mean by "I can't explain why only one machine is affected" above? Do you have oth...
Dmitry Smirnov
12:27 AM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
What do you mean by "I can't explain why only one machine is affected" above? Do you have other similar boxes/setups... Ilya Dryomov
04:45 AM Revision 7b169a0f (ceph): rocksdb backend optimization
Signed-off-by: xinxin shu <xinxin.shu@intel.com> xinxin shu
04:45 AM Revision cbd0043c (ceph): add rocksdb support
Signed-off-by: xinxin shu <xinxin.shu@intel.com> Xinxin Shu
04:45 AM Revision 6eea02c1 (ceph): add rocksdb submodule
Sage Weil
04:45 AM Revision 101954c1 (ceph): get a stable rocksdb (3.0)
Signed-off-by: xinxin shu <xinxin.shu@intel.com> xinxin shu
04:45 AM Revision 708b5b87 (ceph): add --with-librocksdb-static configure options
with this options, ceph will build rocksdb from submodule
Signed-off-by: xinxin shu <xinxin.shu@intel.com>
xinxin shu
04:43 AM Revision 5d5902a6 (ceph): os/KeyValueStore: rename osd_keyvaluedb -> keyvaluestore_backend
This option should be keyvaluestore_*, not osd_*.
Clean up the backend instantiation.
Signed-off-by: xinxin shu <xi...
Sage Weil
04:30 AM Revision 1a5dea72 (ceph): os/FileStore: use generic KeyValueDB::create(); store omap_backend in s...
The only time we set this is at mkfs time. Thereafter, we stick with the
same backend.
Signed-off-by: Sage Weil <sa...
Sage Weil
04:28 AM Revision a2a36192 (ceph): mon/MonitorDBStore: use generic KeyValueDB::create()
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
04:28 AM Revision 86a0b9dd (ceph): os/KeyValueDB: make compaction interface generic
Signed-off-by: Sage Weil <sage@inktank.com> Sage Weil
04:28 AM Revision 4bf929ef (ceph): os/KeyValueDB: generic create(), test_init()
Let us create an implemenetation by name. Include a test_init() method
that will instantiate an instance and verify ...
Xinxin Shu
04:21 AM Revision e1418724 (ceph): config: allow unsafe setting of config values
Allow callers to set config values even when there is no observer.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil
03:49 AM Revision 2f9fe027 (ceph): mailmap: Red Hat names normalization
Signed-off-by: Loic Dachary <loic@dachary.org> Loïc Dachary
02:01 AM rgw Bug #8383: Upload part of one object passed with incorrect upload id or incorrect object id in S3...
Hi,sage,
Sure!
I use S3 API to do this test....
Jingjing Zhao
01:28 AM CephFS Bug #8961 (Won't Fix): du [directory] vs du -b [directory] size doubles
cephfs tracks recursive directory stats. A directory's size is space used by files underneath the directory. If you d... Zheng Yan

07/29/2014

11:53 PM Revision 3cfda577 (ceph): doc/release-notes: v0.83
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:23 PM Revision 6e6fc23c (ceph): Merge tag 'v0.83'
v0.83 Sage Weil
10:55 PM Revision 440c820c (ceph): Merge pull request #2161 from ceph/wip-jcsp-test
Reviewed-by: Greg Farnum greg@inktank.com John Spray
10:33 PM Revision 18ea2a86 (ceph): rgw: fix crash in swift CORS preflight request
Fixes: #8586
This fixes error handling, in accordance with commit 6af5a537 that fixed
the same issue for the S3 case...
Yehuda Sadeh
10:27 PM Revision 79357060 (ceph): Merge pull request #53 from ceph/wip-leaks
use notcmalloc to fix leak checking Sage Weil
10:26 PM Revision fec5b9cb (ceph): Merge pull request #70 from ceph/wip_add_v0.80.5
Added new ceph version Sage Weil
10:25 PM Revision 3350343d (ceph): Merge pull request #69 from ceph/wip_add_new_client1
Added client.1 in order to separate rgw tests to avoid failures on creat...
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
10:25 PM Revision 1717183a (ceph): Added new ceph version
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
10:18 PM Revision c916cabe (ceph): Added client.1 in order to separate rgw tests to avoid failures on crea...
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
09:41 PM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
This problem remains to be very painful... Average frequency is one crash per day. Less than 24 hours ago I had two c... Dmitry Smirnov
09:38 PM Bug #8863: osd: second reservation rejection -> crash
i used this command reimport the crushmap, bug osd still crash shaojun ruan
01:19 PM Bug #8863: osd: second reservation rejection -> crash
try this:
ceph osd getcrushmap -o cm
ceph osd setcrushmap -i cm
and then see if you can reproduce it after t...
Sage Weil
03:41 AM Bug #8863: osd: second reservation rejection -> crash
osd reject the other osd's backfill request twice probably because the space is full, then the request one crashed shaojun ruan
03:27 AM Bug #8863: osd: second reservation rejection -> crash
*scenario:*
1. 3-replica
2. space is nearlly full(some osd >96%)
We guess the reason is osd continuously receivi...
shaojun ruan
09:35 PM Revision 6d01f11a (ceph): Merge pull request #67 from ceph/wip_fix_emperor-x
Fixed -x in the suite, this fixed http://tracker.ceph.com/issues/8862 Warren Usui
09:32 PM Revision 07665ec4 (ceph): mds: introduce explicit DaemonState instead of int
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
09:32 PM Revision c283ad4b (ceph): mds: handle replaying old format journals
To get back to the reformatting procedure that otherwise
occurs during MDLog::open, introduce an MDLog::reopen call
t...
John Spray
09:32 PM Revision 6bb3aeaf (ceph): mds: remove some rogue "using namespace std;"
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
09:32 PM Revision 6b004f19 (ceph): mds: refactor MDS boot
* Make boot_start private.
* Define boot stages in enum, replace int with type.
* Merge steps 0 and 1, 0 always fell ...
John Spray
09:32 PM Revision 6be80873 (ceph): mds: separate inode recovery queue from MDCache
Refactor to:
* have somewhere to put some logic for doing
background recovery in future.
* trim a few lines from th...
John Spray
09:32 PM Revision e5870889 (ceph): mds: remove unused purge_prealloc_ino
Signed-off-by: John Spray <john.spray@redhat.com> John Spray
09:32 PM Revision 6832ec04 (ceph): mds: make MDS::replay_done clearer
... and add some assertions.
Signed-off-by: John Spray <john.spray@redhat.com>
John Spray
09:14 PM Revision 5d2ce7fd (ceph): Fixed -x in the suite, this fixed http://tracker.ceph.com/issues/8862
Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com> Yuri Weinstein
09:11 PM Revision 0d70989a (ceph): python-ceph: require libcephfs.
Signed-off-by: Sandon Van Ness <sandon@inktank.com> Sandon Van Ness
08:43 PM Revision b8687a54 (ceph): Fix package_version_for_hash() call
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> Zack Cerza
08:42 PM Revision 78ff1f0a (ceph): 0.83
Jenkins
08:40 PM Revision aa5f21ce (ceph): Merge pull request #2159 from ceph/wip-undump
tools/cephfs: fuller header in dump/undump
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
08:27 PM Revision 06c47361 (ceph): Remove reference from mkcephfs.
A bit of colission from spec changes for the rhel7/ceph-common
changes and alfredo's pull request for wip-die-ceph-mk...
Sandon Van Ness
07:52 PM Bug #8886: Miss some folders in PG's folder
Hi, Samuel,
First, I correct my word " it should be stored in the DIR_3 at third level", actually it miss the DIR_...
Jingjing Zhao
01:43 PM Bug #8886: Miss some folders in PG's folder
Can you add a find . on that pg directory? Also, does this happen reliably? Also, on what version did you reproduce... Samuel Just
07:36 PM Revision 54330a0a (ceph): Merge pull request #2156 from ceph/wip-upstart-nfile
upstart/ceph-osd.conf: bump nofile limit up by 10x
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
07:33 PM Revision 4045b2e8 (ceph): doc/release-notes: typo
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:30 PM CephFS Bug #8962: kcephfs: client does not release revoked cap
I saw similar hang a few weeks ago. In that case, all OSDs were down, the MDS couldn't submit log event. Zheng Yan
03:05 PM CephFS Bug #8962 (Resolved): kcephfs: client does not release revoked cap
several instances where the mds tries to revoke a cap (Ls and Fs have been observed so far) and the client doesn't re... Sage Weil
07:23 PM Revision df1bad8f (ceph): doc/release-notes: v0.80.5 release notes
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
07:18 PM CephFS Bug #8964: kcephfs: client does not resend requests on mds restart
Zheng Yan
07:18 PM CephFS Bug #8964: kcephfs: client does not resend requests on mds restart
probably fixed by https://github.com/ceph/ceph-client/commit/967166011221589288348b893720d358150176b9 Zheng Yan
05:40 PM CephFS Bug #8964: kcephfs: client does not resend requests on mds restart
mds log and the client kern.log with debug cranked up:... Sage Weil
05:39 PM CephFS Bug #8964 (Resolved): kcephfs: client does not resend requests on mds restart
i have a bunch of hung requests,... Sage Weil
07:16 PM Revision 4aca6c0d (ceph): utf8: export encode_utf8() and decode_utf8()
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
(cherry picked from commit 49fc68cf8c3122c878ea9503c9c74d7046bc9c6f)
Yehuda Sadeh
06:47 PM Feature #8965 (New): Improve threading for ObjectCacher
The ObjectCacher currently use a single global lock for all state. Break this down to improve multithread performanc... Haomai Wang
06:16 PM Revision 9461d8e6 (ceph): Merge remote-tracking branch 'gh/next'
Sage Weil
05:43 PM Revision ffa1f520 (ceph): rgw: dump prefix unconditionally
As part of issue #8858, and to be more in line with S3, dump the Prefix
field when listing bucket even if bucket is e...
Yehuda Sadeh
05:41 PM Revision 6df17a92 (ceph): rgw: list extra objects to set truncation flag correctly
Otherwise we end up returning wrong truncated value, and no data on the
next iteration.
Signed-off-by: Yehuda Sadeh ...
Yehuda Sadeh
05:41 PM Revision 77c20879 (ceph): rgw: account common prefixes for MaxKeys in bucket listing
To be more in line with the S3 api. Beforehand we didn't account the
common prefixes towards the MaxKeys (a single co...
Yehuda Sadeh
05:41 PM Revision 7068d859 (ceph): rgw: add NextMarker param for bucket listing
Partially fixes #8858.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
(cherry picked from commit 924686f0b6593deff...
Yehuda Sadeh
05:41 PM Revision 9604425b (ceph): rgw: improve delmited listing of bucket
If found a prefix, calculate a string greater than that so that next
request we can skip to that. This is still not t...
Yehuda Sadeh
05:34 PM Revision a949a55b (ceph): Merge branch 'origin/wip-osd-leaks'
Reviewed-by: Greg Farnum <greg@inktank.com> Greg Farnum
05:33 PM Revision b287c677 (ceph): rgw: don't try to wait for pending if list is empty
Fixes: #8846
Backport: firefly, dumpling
This was broken at ea68b9372319fd0bab40856db26528d36359102e. We ended
up ca...
Yehuda Sadeh
05:14 PM Revision b354631e (ceph): Merge pull request #62 from ceph/wip-standbyreplay
suites/fs: add a standbyreplay configuration John Spray
05:11 PM Revision fd6e53a5 (ceph): ceph_deploy: use new ship_utilities location
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 828bb2a6ea934b8788570ca3f9871a637dbd3c70)
Sage Weil
03:55 PM Feature #8960: filestore: store backend type persisently
Sage Weil
10:27 AM Feature #8960 (Resolved): filestore: store backend type persisently
Sage Weil
03:32 PM rgw Bug #8586 (Fix Under Review): Missing Swift API Header causes RadosGW to segfault
Yehuda Sadeh
03:06 PM RADOS Bug #8963 (Resolved): erasure coding crush rulset breaks rbd kernel clients on non-ec pools on Ub...
On a fresh install using ceph-deploy on Ubuntu 14.04 creating any erasure coded pool breaks rbd clients on linux 3.13... Greg Dahlman
03:02 PM Bug #8726 (Resolved): (firefly command on dumpling issue?) Error "'adjust-ulimits ceph-coverage /...
commit:fcc0b2451b47793a64fc4cd4675fef667a4a5b45 in ceph-qa-suite.git Josh Durgin
02:31 PM Bug #8628: Bad ceph_osd_op.extent union access in ReplicatedPG::do_osd_ops
This was fixed in 58212b1. Adam Crume
02:28 PM devops Bug #6091 (Won't Fix): centos build should use redhat-rpm-config for debuginfo packages
Sage Weil
02:28 PM devops Bug #5819 (Won't Fix): redhat-rpm-config package needed for debuginfo packages
Sage Weil
02:26 PM devops Bug #7181 (Rejected): debian 7 wheezy init.d script will not start OSDs not corresponding to a mo...
touch /var/lib/ceph/osd/*/sysvinit Sage Weil
02:26 PM devops Bug #6937 (Resolved): udev: OSD using dmcrypt aren't automatically started
Sage Weil
02:25 PM devops Bug #6453 (Won't Fix): libapache2-mod-fastcgi Packages for Debian Squeeze have incorrect dependen...
Sage Weil
02:25 PM devops Bug #6158: selective sync of ceph precise dependencies from havana cloud archive
Note: Talk to neil about this. Sandon Van Ness
02:22 PM devops Bug #8602 (Rejected): ceph fedora package is missing erasure code libraries
redoing (redid?) these packages Sage Weil
02:22 PM Bug #8711 (Resolved): Error "ceph --format=json-pretty osd lspools" is "unrecognized command" in ...

Oops, this should have been closed already...
John Spray
01:51 PM Bug #8711: Error "ceph --format=json-pretty osd lspools" is "unrecognized command" in cuttlefish
Probably best to change the test to cope? Samuel Just
02:21 PM devops Bug #7598 (Can't reproduce): ceph-disk-activate error with ceph-deploy
Sage Weil
02:19 PM devops Bug #8581 (Can't reproduce): DNS issues when resolving hosts
Sage Weil
02:17 PM devops Bug #8734: EPEL / Ceph.com package priority issues
ceph-deploy sets the priority; other users will need to do so themselves.
perhaps that can be mentioned in the doc...
Sage Weil
02:15 PM devops Bug #5283 (Won't Fix): Ceph-deploy can't handle /dev/disk/by-* device paths
Sage Weil
02:06 PM devops Bug #7627 (Resolved): ceph-disk: does not start daemons properly under systemd
commit:3e0d9800767018625f0e7d797c812aa44c426dab Sage Weil
02:01 PM Documentation #8875: `ceph-deploy new` needs to be called for every node, not just the admin one
Can you paste the whole output of ceph-deploy? Alfredo Deza
01:58 PM Bug #6141 (Can't reproduce): OSDs crash on recovery
Samuel Just
01:52 PM Bug #8673 (Resolved): s3tests.functional.test_s3.test_multipart_upload failed in teuthology-2014-...
Sage Weil
01:50 PM Bug #8654 (Resolved): Parsing /etc/lsb-release for OSD metadata is not portable
Sage Weil
01:49 PM Bug #8644 (Rejected): 624ae21833 breaks ceph-disk
Sage Weil
01:48 PM Bug #8852 (Won't Fix): submodules not cecking out the right branch, jerasure does not compile
workaround is to remove the dir then rerun the submodule command. we blame git! Sage Weil
01:47 PM Bug #8801 (Can't reproduce): Ceph monitors do not start after server restart
from teh logs the ceph-mon process was never started.. iw ould look in your /var/log/upstart logs? Sage Weil
01:37 PM Bug #8943 (Pending Backport): "ceph df" cannot show pool available space correctly
commit:04d0526718ccfc220b4fe0c9046ac58899d9dafc Sage Weil
01:34 PM Bug #8495 (Duplicate): osd: bad state machine event on backfill request
Sage Weil
01:29 PM Bug #8694 (Duplicate): OSD crashed (assertion failure) at FileStore::_collection_move_rename
#8733 Sage Weil
01:28 PM rgw Bug #8676: md5sum check failed during readwrite.py
I don't see anything wrong in the logs other than this:... Yehuda Sadeh
01:27 PM Bug #8753: PG::activate assert failed when recover finished
Has this happened since? Samuel Just
01:26 PM Bug #8865: cep osd setmaxosd doesn't check if osds exist
agreed Samuel Just
01:26 PM Bug #8752 (Can't reproduce): firefly: scrub/repair stat mismatch
Sage Weil
01:25 PM Bug #8752 (Resolved): firefly: scrub/repair stat mismatch
Sage Weil
01:06 PM CephFS Bug #8961 (Won't Fix): du [directory] vs du -b [directory] size doubles
Under cephfs using the kernel client, du -b shows an incorrect size.
I've also found that du --apparent-size shows...
Matt Hook
01:04 PM Bug #8717 (In Progress): teuthology: valgrind leak checks broken for osd (at least)
Sage Weil
01:03 PM Bug #8717 (Resolved): teuthology: valgrind leak checks broken for osd (at least)
Sage Weil
01:04 PM Revision 0bd4c862 (ceph): Merge pull request #2139 from ceph/wip-journal-header
os/FileJournal: Update the journal header when closing journal
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
01:03 PM Bug #8926 (Resolved): osd: invalid Message* deref in C_SendMap
Sage Weil
01:03 PM Bug #8924 (Resolved): osd: leaking local_connection under valgrind
Sage Weil
01:01 PM Revision 37eba045 (ceph): Merge pull request #2146 from ceph/wip-8932
ceph_test_rados_api_tier: do fewer writes in HitSetWrite
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
12:59 PM Messengers Bug #8880: msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Sage Weil
12:58 PM Revision 050ac875 (ceph): Merge pull request #2147 from ceph/wip-8931
osd: fix ops blocked by full cache tier dequeue
Reviewed-by: Greg Farnum <greg@inktank.com>
Gregory Farnum
12:24 PM Revision 38b73c67 (ceph): 0.80.5
Jenkins
12:11 PM Revision 0147c746 (ceph): task/mds_journal_migration: fix cleanup
Previously this was remounting in the background
which meant that if a workunit was executed immediately
afterwards i...
John Spray
10:42 AM rgw Bug #8632 (Resolved): rgw: bucket listing with delimiter doesn't scale well
backported to dumpling commit:9604425b86f5839a109faa1f396b0d114e9b9391 Yehuda Sadeh
09:36 AM rgw Bug #8632 (Pending Backport): rgw: bucket listing with delimiter doesn't scale well
in firefly, not dumpling yet Sage Weil
10:34 AM Revision 431ae2f5 (ceph): suites/fs: add a standbyreplay configuration
A place for any tests which should be run with
standbyreplay MDSs in play. Initially just
the mds_journal_migration ...
John Spray
10:31 AM rgw Bug #8846 (Resolved): radosgw on 0.80.4 crashes when doing a multi-part upload
Yehuda Sadeh
10:11 AM Bug #8532 (Can't reproduce): 0.80.1: OSD crash (domino effect), same as BUG #8229
Let us know if anything interesting comes up. Samuel Just
10:10 AM Bug #8229: 0.80~rc1: OSD crash (domino effect)
This bug described a whole bunch of unrelated problems, can you open a fresh bug? Samuel Just
10:01 AM Bug #8959: osd crashed in upgrade:dumpling-x-firefly---basic-vps suite
this sounds a bit like a problem we had a while back with hung IOs from the VMs? Sage Weil
08:40 AM Bug #8959: osd crashed in upgrade:dumpling-x-firefly---basic-vps suite
Seems the same crash in another tests, logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-07-28_11:48:15... Yuri Weinstein
08:36 AM Bug #8959 (Can't reproduce): osd crashed in upgrade:dumpling-x-firefly---basic-vps suite
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-07-28_11:48:15-upgrade:dumpling-x-firefly---basic-vps... Yuri Weinstein
09:41 AM CephFS Bug #8574: teuthology: NFS mounts on trusty are failing
I'm not sure if this is a different issue or a different system:... Greg Farnum
09:40 AM devops Support #8861: Deploying additional monitors fails.
I am also seeing this error when trying to add a new monitor. Same version of Ubuntu and Ceph. James Devine
09:38 AM rgw Bug #8735 (Can't reproduce): TestAccountNoContainers fail in Firefly upgrade:firefly-x:stress-split
Sage Weil
09:38 AM rgw Bug #8766: multipart minimum size error should be EntityTooSmall
Sage Weil
09:37 AM rgw Bug #8848 (Resolved): "adjust-ulimits: command not found" in upgrade:firefly-firefly-testing-basi...
Sage Weil
09:37 AM rgw Bug #8847 (Can't reproduce): "Error initializing cluster client" in upgrade:firefly-firefly-testi...
Sage Weil
09:34 AM Bug #8921 (Won't Fix): ceph pg dump <{summary|sum|delta|pools|osds|pgs|pgs_brief}> only work corr...
Sage Weil
09:33 AM rgw Bug #8864: radosgw help doesn't seem to display some debug options
there are others that we could add Sage Weil
09:32 AM rgw Bug #8864 (Resolved): radosgw help doesn't seem to display some debug options
Sage Weil
09:32 AM rgw Bug #6911 (Won't Fix): rgw test failure on the arm set up
Sage Weil
09:31 AM rgw Bug #8111 (Need More Info): /etc/init.d/ceph-radosgw for RHEL needs QA
isn't it /etc/init.d/radosgw?
Sage Weil
09:30 AM rgw Bug #8383 (Need More Info): Upload part of one object passed with incorrect upload id or incorrec...
Can you provide more detailed steps to reproduce? ideally, a new test in s3-tests.... :) Sage Weil
09:29 AM rgw Bug #7799 (Can't reproduce): Errors in upgrade:dumpling-x:stress-split-firefly---basic-plana suite
Sage Weil
09:25 AM rgw Bug #8311 (Resolved): No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-...
Sage Weil
09:25 AM rgw Bug #8784: rgw: completion leak
Sage Weil
09:23 AM rbd Bug #6695 (Won't Fix): Upgrade rbd failure in nightly tests. (mkdir --p ..)
Sage Weil
09:22 AM rbd Bug #5480 (Can't reproduce): libceph: unexpected old state in con_sock_state_change
Sage Weil
09:21 AM rbd Bug #8845: Flattening Clones of clone, results in command failure
fsx is now able to catch this one. Ilya Dryomov
09:19 AM rbd Bug #8845: Flattening Clones of clone, results in command failure
Josh Durgin
09:15 AM rbd Bug #8845: Flattening Clones of clone, results in command failure
Ilya Dryomov
09:21 AM rbd Bug #7693: virsh domblkinfo fails with 'Bad file descriptor'
https://bugzilla.redhat.com/show_bug.cgi?id=1124508 Sage Weil
09:17 AM rbd Bug #7620 (Can't reproduce): BUG: soft lockup - CPU#0 stuck for 23s!
Sage Weil
09:15 AM Linux kernel client Bug #8568 (New): libceph: kernel BUG at net/ceph/osd_client.c:885
Ilya Dryomov
09:10 AM Linux kernel client Bug #8568: libceph: kernel BUG at net/ceph/osd_client.c:885
Ilya Dryomov
09:14 AM rbd Bug #8709: stale size reported by ioctl(BLKGETSIZE64) after librbd_resize() returns
The problem has been traced to http://tracker.ceph.com/issues/8806. Keeping this around to re-test after it gets fixed. Ilya Dryomov
09:11 AM Bug #8439 (Won't Fix): ceph-osd crashing often
see 0.80.x Sage Weil
09:10 AM Bug #8445 (Won't Fix): osd not starting anymore
0.78 had lots of issues; see 0.80.x Sage Weil
09:01 AM rbd Bug #8318 (Can't reproduce): "rbd: create error" in upgrade:dumpling-dumpling-testing-basic-plana...
Sage Weil
09:01 AM rbd Bug #8715 (Can't reproduce): "ceph_test_librbd_fsx: invalid option -- 'h'" error in teuthology-20...
Sage Weil
06:57 AM CephFS Feature #7759 (Resolved): journal-tool: roll in resetter/dumper from MDS
... John Spray
06:56 AM CephFS Feature #7761 (Resolved): journal-tool: forwards-search through corrupt regions
... John Spray
06:55 AM CephFS Feature #7763: journal-tool: import
... John Spray
06:54 AM CephFS Feature #7763 (Resolved): journal-tool: import
This was done when undump was merged into cephfs-journal-tool John Spray
06:51 AM CephFS Bug #8773 (Resolved): failing cephfs set_layout tests
Test is retired and unsafe behaviour (data pool default to 0) is disabled in master. John Spray
06:07 AM CephFS Bug #8811 (Resolved): Journal corruption during upgrade to 0.82 with standby-replay daemons
This got fixed 11 days ago, but was never marked closed. Merged in commit:b9463e3497cc1f2a1bab0838430a4402d8c88af0 Greg Farnum
05:59 AM Bug #8932 (Resolved): rados api test hang on HitSetWrite
Merged to master in commit:37eba045ec78f2ea8f9000c6b158e20808d29fb2 Greg Farnum
05:56 AM Bug #8931 (Pending Backport): failed write reply order from ceph_test_rados
Merged to master in commit:050ac87530c2637f097e07b5373115721303f07c Greg Farnum
01:42 AM Revision e99acf98 (ceph): OSD: add require_same_peer_inst(OpRequestRef&,OSDMap&) helper
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
01:39 AM Revision e179e922 (ceph): OSD: introduce require_self_aliveness(OpRequestRef&,epoch_t) function
Take the self-aliveness checks out of require_same_or_newer_map() and use
the new function for that and for require_u...
Greg Farnum
12:18 AM Revision f36cffc9 (ceph): unittest_crush_wrapper: fix build
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
12:06 AM Revision 7f913dcd (ceph): Merge pull request #2150 from ceph/wip-libs
don't link everything with blkid, udev, and boost_threads Dan Mick

07/28/2014

11:28 PM Revision 3f3da880 (ceph): Update README
Yuri Weinstein
11:24 PM Revision f2b74226 (ceph): Update README
Yuri Weinstein
11:14 PM Revision 7dcb8bd6 (ceph): dumpling-firefly-x: document structure of suite
Hopefully this makes it easier to digest and maintain
than reading yaml fragments.
Signed-off-by: Josh Durgin <josh....
Josh Durgin
10:47 PM Bug #8944: Ceph daemon bad asok used in connection with cluster
wip-8944 created, but gitbuilders are having enough problems I'm not submitting a PR yet Dan Mick
02:11 PM Bug #8944 (Fix Under Review): Ceph daemon bad asok used in connection with cluster
Adding the global args to the invocation of ceph-conf seems to resolve this. Dan Mick
12:41 PM Bug #8944: Ceph daemon bad asok used in connection with cluster
oh....because --cluster on the cli ... yeah.
Dan Mick
12:40 PM Bug #8944: Ceph daemon bad asok used in connection with cluster
ceph uses ceph-conf --show-config-value admin_socket -n <name> and believes it; wonder why that's not working? Dan Mick
09:58 AM Bug #8944: Ceph daemon bad asok used in connection with cluster
Sage Weil
05:01 AM Bug #8944 (Resolved): Ceph daemon bad asok used in connection with cluster
Using @ceph --cluster clustername daemon mon.host1 config@ causes ... Szymon Zacher
10:46 PM Bug #8947: Writing rados objects with max objects set for cache pool crashed osd
Uploading crash dump Mallikarjun Biradar
01:45 PM Bug #8947: Writing rados objects with max objects set for cache pool crashed osd
Could not reproduce using vstart.sh on current master branch. I never saw a crash or bug report with that stack trace. David Zafman
10:08 AM Bug #8947: Writing rados objects with max objects set for cache pool crashed osd
I don't remember the details, but we were previously crashing with a 10-object limit anyway due to hit sets and such.... Greg Farnum
08:16 AM Bug #8947: Writing rados objects with max objects set for cache pool crashed osd
Test configuration:
No of osd nodes: 3
No of osd's : 4
No of monitors: 2
Kernel versions: 3.13.0-24-generic
No o...
Mallikarjun Biradar
08:15 AM Bug #8947 (Duplicate): Writing rados objects with max objects set for cache pool crashed osd
Setting target_max_objects parameter and writing rados object onto cache pool crashed osd.
History of operations o...
Mallikarjun Biradar
09:30 PM Revision 79c63166 (ceph): Merge pull request #2153 from ceph/wip-fsx-overlap
librbd API fix + wip-fsx-overlap
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Josh Durgin
09:08 PM Revision eb2f1ea2 (ceph): OSD: use OpRequestRef& for a few require_* functions
Signed-off-by: Greg Farnum <greg@inktank.com> Greg Farnum
06:41 PM Revision 7f12a0f4 (ceph): Merge pull request #2152 from xiaoxichen/fix_ceph_df
PGMonitor: fix bug in caculating pool avail space
Reviewed-by: Sage Weil <sage@redhat.com>
Sage Weil
06:41 PM Messengers Bug #8880 (Fix Under Review): msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_s...
New patches to split up the code more, as requested. :) Greg Farnum
10:56 AM Messengers Bug #8880 (In Progress): msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq fe...
Greg Farnum
05:38 PM Revision 1526546d (ceph): Remove reference from mkcephfs.
A bit of colission from spec changes for the rhel7/ceph-common
changes and alfredo's pull request for wip-die-ceph-mk...
Sandon Van Ness
04:42 PM Revision 9b037522 (ceph): Fix some style and checking issue
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com> Xiaoxi Chen
04:27 PM Revision 5773a374 (ceph): upstart/ceph-osd.conf: bump nofile limit up by 10x
This should ensure that we don't hit this limit on all but the very biggest
clusters. We seen it hit on a ~500 OSD d...
Sage Weil
04:22 PM Revision cb20b996 (ceph): Merge pull request #2154 from simon3z/master
init: add systemd service files
Reviewed-by: Alfredo Deza <alfredo.deza@inktank.com>
Reviewed-by: Sage Weil <sage@re...
Sage Weil
02:50 PM Revision d3e5961d (ceph): tools/cephfs: fuller header in dump/undump
There were two problems here:
* write_pos was modified through an undump/dump cycle,
because it was probed during...
John Spray
02:12 PM rgw Bug #8937 (Fix Under Review): rgw: broken large(-ish) objects
Yehuda Sadeh
02:10 PM rgw Feature #7774 (Resolved): rgw: cache decoded user and bucket info
This one has been merged in a while a go, at commit:82c547952dc9e7a3e9fab1264f5fdd903ab6973e. Yehuda Sadeh
01:59 PM Revision 671533c2 (ceph): move ship_utilities to install task (from ceph test)
No need to install ceph just to get daemon-helper, adjust-ulimits, etc..
These utilities are used by lots of tasks, n...
Sage Weil
01:07 PM Bug #8941 (Can't reproduce): DaemonConfig.SubstitutionLoop unit test goes haywire
nevermind, most recent occurrence was feb, so ignoring this. Sage Weil
01:02 PM rgw Feature #8956 (Resolved): rgw: support bucket notification
Yehuda Sadeh
11:32 AM Documentation #8955: doc refers to [default] section, don't think it exists
http://ceph.com/docs/master/start/quick-ceph-deploy/#create-a-cluster refers to the [default] section in the ceph.con... Dan Mick
11:31 AM Documentation #8955 (Resolved): doc refers to [default] section, don't think it exists
Dan Mick
09:53 AM Revision eb697dd9 (ceph): librbd: make rbd_get_parent_info() accept NULL out params
The C++ version of rbd_get_parent_info() allows passing NULL for parent
image name, image name and snapshot name out ...
Ilya Dryomov
09:53 AM Revision f6d1a920 (ceph): test_librbd_fsx: introduce rbd_image_has_parent()
A helper to check whether the image associated with the ctx has
a parent or not.
Signed-off-by: Ilya Dryomov <ilya.d...
Ilya Dryomov
09:53 AM Revision bb095ffd (ceph): test_librbd_fsx: randomize_parent_overlap
Truncate base images after they have been cloned from to cover more
code paths and make sure that clients look at sna...
Ilya Dryomov
09:53 AM Revision e183a4d9 (ceph): test_librbd_fsx: clone/flatten probabilities
Higher the clone probability to 8% and lower the probability of flatten
to 2%. This should give us longer parent cha...
Ilya Dryomov
09:47 AM Revision 04d05267 (ceph): PGMonitor: fix bug in caculating pool avail space
Currently for pools with different rules, "ceph df" cannot report
right available space for them, respectively. For d...
Xiaoxi Chen
09:21 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
I'm pretty sure it's the disabled lockdep that affects this. Our testing kernel is built with lockdep enabled, Ubunt... Ilya Dryomov
08:50 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
Hi Ilya,
I can reliably reproduce the error when running this generic kernel with no changes:
http://kernel.ubu...
Greg Wilson
08:39 AM Bug #8935: operations not idempotent when enabling cache
I think you're right that a per-object log would be needed to solve this problem — and I think that means we shouldn'... Greg Farnum
08:02 AM rgw Feature #8945 (Resolved): rgw: support swift /info api
Yehuda Sadeh
06:55 AM Bug #8938 (Resolved): OSD memory leak seen with fs-master-testing-basic/kernel_untar_build.sh
This was fixed at about the same time:... John Spray
06:42 AM CephFS Feature #7810 (In Progress): libcephfs: add a test that freezes + unfreezes a client, and then ve...
John Spray
05:27 AM Bug #8895: ceph osd pool stats (displayed incorrect values)
Negative & undefined values in counts objects:
*-5/0 objects degraded (-inf%)*
*-32/12 objects degraded (-266...
Andrey Matyashov
03:06 AM rgw Bug #8864: radosgw help doesn't seem to display some debug options
This should be closed with #8112 Abhishek Lekshmanan
02:48 AM Bug #8943 (Resolved): "ceph df" cannot show pool available space correctly
Currently when user have 2 pools with different ruleset and different root, basically they will use differen... Xiaoxi Chen
02:39 AM Revision 3695b255 (ceph): Merge pull request #2149 from yuyuyu101/wip-flush-set
Fix dup bh_write for TX state bh
Tested-by: Sage Weil <sage@redhat.com>
Reviewed-by: Haomai Wang <haomaiwang@gmail.c...
Sage Weil
12:37 AM Bug #8863: osd: second reservation rejection -> crash
Last week we've created a new cluster(all components use v0.80.4), continuously writing data until space is full, the... shaojun ruan

07/27/2014

11:58 PM Revision b08470f0 (ceph): configure.ac: link libboost_thread only with json-spirit
Signed-off-by: Sage Weil <sage@redhat.com> Sage Weil
11:45 PM Bug #8942 (Resolved): Bad JSON output in ceph osd tree
Hi,
JSON output for @ceph osd tree@ has bad format for stray array: every osd are printed in the same array element....
Szymon Zacher
10:41 PM Bug #8941 (Can't reproduce): DaemonConfig.SubstitutionLoop unit test goes haywire
... Sage Weil
10:31 PM Bug #8822: osd: hang on shutdown, spinlocks
saw this again, ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-07-27_02:30:01-rados-next-testing-basi... Sage Weil
10:28 PM Bug #8396: osd: message delayed in Session misdirected after split
very likely another instance, but i didn't look closely.... Sage Weil
10:20 PM Bug #8940 (Duplicate): 3.22s1 shard 0(2) missing ad166f62/benchmark_data_plana57_30491_object1036...
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-07-27_02:30:01-rados-next-testing-basic-plana/380335
...
Sage Weil
09:47 PM Bug #6003: journal Unable to read past sequence 406 ...
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-07-27_02:30:01-rados-next-testing-basic-plana/380261
...
Sage Weil
06:25 PM Revision 9d23cc6a (ceph): configure: don't link blkid, udev to everything
These are already explicitly called out for libkrbd; don't need them in
LIBS.
Signed-off-by: Sage Weil <sage@redhat....
Sage Weil
04:28 PM Revision 3e51cb67 (ceph): rados/singleton-nomsgr/all/alloc-hint: fix config syntax
Put in [osd] section.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2b4f7402007e6a81dce10f3b...
Sage Weil
02:32 PM Bug #8758: PGs get stuck in “replay”, but drop it upon osd restarts
As for the issue of losing replay states upon member osd restarts... Could the fix be as simple as not setting inter... Alexandre Oliva
01:44 PM Bug #8758: PGs get stuck in “replay”, but drop it upon osd restarts
Here's a patch that addresses the “stuck in replay” problem (but not the “replay is dropped after osd re-peering” one). Alexandre Oliva
11:21 AM Bug #8863 (Need More Info): osd: second reservation rejection -> crash
Sage Weil
11:20 AM Bug #8922 (Need More Info): ceph-deploy mon create fails to create additional monitoring nodes.
It sounds like the monitor names don't match the host names or something similar. Can you post the full sequence of ... Sage Weil
11:04 AM Revision cc7c9b2e (ceph): task/workunit: add sudo to dir delete
This failure to delete used to be hidden in
a catch-all exception handler that was
removed in 396dc073.
This was cau...
John Spray
05:37 AM Revision de9cfcaa (ceph): Only write bufferhead when it's dirty
The TX state bh should be skipped because the bh should be inflight. We only
need to write dirty bh. And TX and dirty...
Haomai Wang
05:36 AM Revision 1c26266d (ceph): ObjectCacher: fix bh_{add,remove} dirty_or_tx_bh accounting
tx buffers need to go on the bh_lru_rest as well, and removing erases
(not inserts) them into dirty_or_tx_bh.
Signed...
Josh Durgin
05:36 AM Revision 727ac1d0 (ceph): ObjectCacher: fix dirty_or_tx_bh logic in bh_set_state()
The else-if chain here was wrong. Handling dirty or tx buffers and
errors should be in independent conditions.
Signe...
Josh Durgin
05:33 AM Revision 5283cfee (ceph): Wait tx state buffer in flush_set
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> Haomai Wang
05:33 AM Revision d858fdc5 (ceph): Add rbdcache max dirty object option
Librbd will calculate max dirty object according to rbd_cache_max_size, it
doesn't suitable for every case. If user s...
Haomai Wang
05:33 AM Revision b8a56685 (ceph): Reduce ObjectCacher flush overhead
Flush op in ObjectCacher will iterate the whole active object set, each
dirty object also may own several BufferHead....
Haomai Wang
04:19 AM Revision 288908b3 (ceph): Revert "Merge pull request #2129 from ceph/wip-librbd-oc"
This reverts commit 74b386f03e4ca9970256db72c575589aea077534, reversing
changes made to 36265d0db0d7c0eb31d25a0f77ac2...
Sage Weil
 

Also available in: Atom