Activity
From 04/09/2014 to 05/08/2014
05/08/2014
- 10:35 PM Revision d34cc1e7 (ceph): Merge pull request #1772 from ceph/wip-8169
- rgw: calculate user manifest
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 09:58 PM Revision 5986f746 (ceph): :doc Ceph OSD is standard name
- This is a method of standardizing the usage of OSD so that "Ceph OSD"
is the daemon, and OSD maintains its industry s... - 09:29 PM Revision ddc2e1a8 (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... - 09:19 PM Revision 589b639a (ceph): osd/ReplicatedPG: carry CopyOpRef in copy_from completion
- There is a race with copy_from cancellation. The internal Objecter
completion decodes a bunch of data and copies it ... - 08:50 PM Revision aff119ac (ceph): Merge pull request #1791 from ceph/wip-8011
- ReplicatedPG: block scrub on blocked object contexts
Reviewed-by: Sage Weil <sage@inktank.com> - 08:37 PM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- Well, there were number of upgrades 0.72.2 --> 0.78 --> 0.79 --> 0.80~rc1 --> 0.80
to give you the impression how mu... - 08:35 PM Revision d4e67ff3 (ceph): ReplicatedPG::recover_backfill: do not update last_backfill prematurely
- Previously, we would update last_backfill on the backfill peer to
backfills_in_flight.empty() ? backfill_pos :
bac... - 08:35 PM Revision d620b13c (ceph): ReplicatedPG: add empty stat when we remove an object in recover_backfill
- Subsequent updates to that object need to have their stats added
to the backfill info stats atomically with the last_... - 08:13 PM devops Bug #7627: ceph-disk: does not start daemons properly under systemd
- Opened https://bugzilla.redhat.com/show_bug.cgi?id=1095988
- 07:57 PM Revision fced0562 (ceph): rgw: don't error out on empty owner when setting acls
- Fixes: #6892
Backport: dumpling, emperor
s3cmd specifies empty owner field when trying to set acls on object
/ bucket... - 07:46 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- I built DKMS modules [libceph,ceph,rbd] based on HEAD ("ceph: reserve caps for file layout/lock MDS requests") of "fo...
- 09:57 AM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- 07:11 AM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Dmitry, a patch in #8300 should fix this.
- 07:41 PM Revision db4ccb04 (ceph): ReplicatedPG: block scrub on blocked object contexts
- Fixes: #8011
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com> - 06:47 PM Revision b4508a08 (ceph): Merge pull request #253 from ceph/wip-7510-wusui
- Further clarify 'too many values to unpack' error.
- 05:57 PM Revision 6fbf98bb (ceph): Further clarify 'too many values to unpack' error.
- Many errors in yaml configurations cause ValueError to get thrown
with the message 'too many values to unpack.' A pr... - 05:52 PM Revision 3152faf7 (ceph): osd/osd_types: add last_force_op_resend to pg_pool_t
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:13 PM Revision fc263c3f (ceph): Merge pull request #1778 from ceph/wip-7157
- ceph-disk: fix list for encrypted or corrupt volume
Reviewed-by: Alfredo Deza <alfredo.deza@inktank.com> - 04:07 PM Revision c3e3a132 (ceph): Merge pull request #1789 from ceph/wip-jcsp-clang
- misc. cleanups from clang warnings
Reviewed-by: Sage Weil <sage@inktank.com> - 04:00 PM Revision fbeb298d (ceph): Merge pull request #1777 from ceph/wip-6966
- ceph-disk: partprobe before settle when preparing dev
Reviewed-by: Alfredo Deza <alfredo.deza@inktank.com> - 03:52 PM Revision 0f196265 (ceph): ceph-disk: partprobe before settle when preparing dev
- Two users have reported this fixes a problem with using --dmcrypt.
Fixes: #6966
Tested-by: Eric Eastman <eric0e@aol.... - 03:47 PM Revision 2e530771 (ceph): Merge pull request #1788 from ceph/wip-da-sca-20140507
- Fix some issues from SCA
- 03:44 PM Bug #8301: Suicide Timeout on Cache Tier OSDs
- above was on 24c5ea8df040da0889be7ab1a9985ae03ee68d9e
- 03:37 PM Bug #8301: Suicide Timeout on Cache Tier OSDs
- i is only 1344, I don't yet see how this situation leads to a hang. The value of new_trim_to is odd.
I've pushed ... - 03:35 PM Bug #8301: Suicide Timeout on Cache Tier OSDs
- This backtrace is consistent with the hangs I saw in some logs generated in a subsequent run.
(gdb) bt
#0 Replic... - 03:33 PM Revision df94b8de (ceph): Merge pull request #1790 from ceph/wip-krbd-fixes
- Two minor krbd fixes
Reviewed-by: Sage Weil <sage@inktank.com> - 03:08 PM Revision 56902320 (ceph): rbd-fuse.c: remove ridiculous linebreak
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 03:05 PM Revision 7a3724b0 (ceph): rbd-fuse.c: fix indentation
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 02:59 PM Revision 8101f980 (ceph): rbd-fuse.c: fix -Wmissing-field-initializers
- Init image_name with NULL to fix:
rbd_fuse/rbd-fuse.c:57:63: warning: missing field 'image_name' initializer
[-Wmis... - 02:22 PM rbd Bug #8318 (Can't reproduce): "rbd: create error" in upgrade:dumpling-dumpling-testing-basic-plana...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-07_19:15:07-upgrade:dumpling-dumpling-testing-basi...
- 02:17 PM Revision f1d953e0 (ceph): krbd: match new with delete, not free()
- struct krbd_ctx is allocated with new, use delete to get rid of it.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktan... - 02:17 PM Revision 65ca867e (ceph): krbd: fix sysfs path in the comment
- It's "/sys/bus/rbd/devices/<id>", but libudev works with devices and
not busses, so it's really "/sys/devices/rbd/<id... - 02:09 PM Revision 082367e8 (ceph): rbd.cc: init 'snap_protected' to fix -Wconditional-uninitialized
- Init 'snap_protected' with false to fix:
rbd.cc:544:35: warning: variable 'snap_protected' may be uninitialized
whe... - 02:04 PM Revision 0d01563f (ceph): rbd-fuse.c: init 'rbd' in open_rbd_image()
- Init 'rbd' in open_rbd_image() with NULL and add a check for
'rbd' before dereference it to fix:
rbd_fuse/rbd-fuse.c... - 01:54 PM Revision cfc885fa (ceph): ObjectCacher::_wait_for_write(): init 'bool done'
- Init 'bool done' with 'false' to fix:
osdc/Objecter.h:915:27: warning: implicit conversion los: variable 'done'
may... - 01:49 PM Bug #8011 (Pending Backport): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || s...
- 01:36 PM Bug #8011 (Fix Under Review): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || s...
- 01:47 PM Revision 8322878c (ceph): Objecter::calc_target(): init best_locality with 0
- Init best_locality to fix:
osdc/Objecter.cc:1519:26: warning: variable 'best_locality' may be
uninitialized when us... - 01:36 PM Bug #8104: OSD: changing min_size to larger than the acting set does not make the PG go inactive
- 01:35 PM Bug #8162: osd: dumpling advances last_backfill prematurely
- wip-8261-dumpling, going to do a suite run before merging.
- 01:23 PM Bug #7999 (In Progress): osd: pgs share info that hasn't been persisted
- 11:45 AM Bug #7999: osd: pgs share info that hasn't been persisted
- ubuntu@teuthology:/a/samuelj-2014-05-06_14:42:22-rados-wip-sam-testing-testing-basic-plana/240025/remote
- 12:54 PM rgw Fix #6892 (Resolved): rgw: ignore empty owner in set acl api
- cherry picked into dumpling
- 09:56 AM rgw Fix #6892 (Pending Backport): rgw: ignore empty owner in set acl api
- 12:28 PM Revision f0231ef3 (ceph): mon: Fix % escaping (\% should be %%)
- Clang's -Wpedantic points this out.
Signed-off-by: John Spray <john.spray@inktank.com> - 12:28 PM Revision 447335aa (ceph): os/FileJournal: remove unused attribute
- Clang:
os/FileJournal.h:224:8: warning: private field 'is_bdev' is not used
[-Wunused-private-field]
Signed-off-by: ... - 12:28 PM Revision 13750a1d (ceph): rgw: Remove trailing ; from fn definitions
- Clang:
warning: extra ';' after member function
definition [-Wextra-semi]
Signed-off-by: John Spray <john.spray@inkt... - 12:28 PM Revision 8584b406 (ceph): fragtree: remove dead code
- Signed-off-by: John Spray <john.spray@inktank.com>
- 12:28 PM Revision 6b15ce1c (ceph): fragtree: remove unused and broken verify()
- This fn had a while(1) with no break: if anyone
had called it it would block forever.
Caught by clang's "function 'v... - 12:28 PM Revision 3fd87127 (ceph): encoding: make .size() to __u32 cast explicit
- Caught by clang warning that this is a conversion
from "unsigned long" to "unsigned int" which can
lose precision. H... - 12:28 PM Revision d85b8faf (ceph): mds: Remove redundant 'using namespace std'
- This simply was not being used, and triggered
a clang warning.
Signed-off-by: John Spray <john.spray@inktank.com> - 11:26 AM rgw Feature #8316 (New): Ceilometer support for RGW Swift statistics
- Customer requests support for Ceilometer statistics collection against RadosGW when used as SWIFT object store for Op...
- 10:54 AM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- 10:34 AM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- Discussed in standup and decided on alternate approach:
epoch_t last_force_op_resend; ///< last epoch in which w... - 10:29 AM Bug #8305 (In Progress): objecter, osd: pool overlay change should trigger op resend
- 10:48 AM devops Bug #8298 (Resolved): missing emperor ceph package
- Thanks for the update :-)
- 10:12 AM devops Bug #8298: missing emperor ceph package
- Sorry this wasn't updated. Alfredo mentioned that it looks like maybe the jenkins repo was not updated correctly and ...
- 10:03 AM devops Bug #8298: missing emperor ceph package
- Now it contains the ceph package. This is strange....
- 07:33 AM devops Bug #8298: missing emperor ceph package
- It is back:...
- 10:46 AM Bug #8315: osd: watch callback vs callback funky
- I could not reproduce it on re-run logs are in /a/teuthology-2014-05-07_19:15:07-upgrade:dumpling-dumpling-testing-b...
- 10:43 AM Bug #8315 (Resolved): osd: watch callback vs callback funky
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-05-07_19:15:07-upgrade:dumpling-dumpling-testing-basi...
- 10:17 AM rgw Bug #8311: No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-bas...
- Additional logs from fresh runs - http://pulpito.front.sepia.ceph.com/teuthology-2014-05-07_19:15:07-upgrade:dumpling...
- 07:54 AM rgw Bug #8311 (Resolved): No pool name error in ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-...
- Logs rae in http://qa-proxy.ceph.com/teuthology/ubuntu-2014-05-06_21:02:54-upgrade:dumpling-dumpling-testing-basic-vp...
- 09:15 AM devops Bug #6966 (Pending Backport): ceph-disk: prepare --dmcrypt failing
- 09:00 AM devops Bug #6966 (Resolved): ceph-disk: prepare --dmcrypt failing
- Merged to master with hash 0f196265f049d432e399197a3af3f90d2e916275
- 08:21 AM devops Bug #6966: ceph-disk: prepare --dmcrypt failing
- I installed wip-6966 on a test cluster and was able to build and use dmcrypted OSDs.
Thanks! - 08:09 AM Revision b4b79ebb (ceph): remove superfluous second semicolons at end of lines
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 07:38 AM Revision 1214257a (ceph): msg: fix some -Wextra-semi warnings
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 07:34 AM Revision 9ad60428 (ceph): crush/builder.c: remove some unreachable return statements
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 07:28 AM rgw Bug #8293: admin api get quotas
- Tested on our newly updated Firefly cluster and same result. I get a 200 with no JSON in the body.
=== User ===
... - 07:08 AM Linux kernel client Bug #8300 (Resolved): Regression in 3.14: "No such device or address" reading file content
- Great, this patch is in 3.15-rc1 ("crush: fix off-by-one errors in total_tries refactor"). I'll make sure it gets in...
- 07:04 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- Yes, your patch fixes the problem. Thank you very much for looking into this!
- 06:40 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- OK, please try the attached patch (on top of 3.14.2) and see if it fixes the problem.
- 07:01 AM Support #8310: Most pgs stuck stale, no osds reporting them, repair ineffective
- You'll generally have better luck with stuff like this on the mailing list. But I see that your PGs aren't mapping to...
- 05:08 AM Support #8310: Most pgs stuck stale, no osds reporting them, repair ineffective
- I mentioned "repair ineffective" without detail. Specifically, I have tried pg repair on all stale pgs, osd scrubs, o...
- 05:03 AM Support #8310 (Closed): Most pgs stuck stale, no osds reporting them, repair ineffective
- After trying to resolve an issue with pgs stuck in cleaning, I restarted osds and most of the pgs in the cluster now ...
- 03:36 AM Revision 3f837254 (ceph): Merge pull request #1742 from ceph/wip-multimds
- Wip multimds
- 03:29 AM Revision 1f600602 (ceph): mds: properly wake up dentry waiters after fragmenting dirfrag
- When active MDS wants to fragment a replica dirfrag, it should set
the 'replay' parameter of MDCache::adjust_dir_frag... - 03:29 AM Revision 34e27e46 (ceph): mds: remove unused MMDSCacheRejoin::{MISSING,FULL}
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 03:29 AM Revision 3ca0d018 (ceph): mds: switch flushing ScatterLock to dirty ScatterLock after cache rejoin
- Otherwise the flushing flag may confuse Locker::eval_gather() if MDS later
imports lock's parent inode.
Signed-off-b... - 03:29 AM Revision 5fa2bae3 (ceph): mds: choose MIX state if replica of scatterlock is in MIX state
- After ScatterLock::infer_state_from_strong_rejoin() set scatterlock
to LOCK_MIX state, don't change the scatterlock t... - 02:58 AM Revision 727ad648 (ceph): client: refactor _lookup; fix NULL dentry case
- Return ENOENT for a valid NULL dentry in our cache. Restructure _lookup
to avoid duplicating some code.
Signed-off-... - 02:58 AM Revision 627e644c (ceph): client: do not manually clean up on unlink/rmdir
- The reply handler will do this in a safe, ordered fashion.
Signed-off-by: Sage Weil <sage@inktank.com> - 02:58 AM Revision 8f3409d1 (ceph): client: unlink dentry on traceless rmdir, unlink reply
- This used to be handled in _unlink() and _rmdir() even when a trace was
present in the reply, but this is cleaner.
S... - 02:58 AM Revision 635607ff (ceph): client: skip insert_trace on safe requests
- Only do this for the first reply.
Signed-off-by: Sage Weil <sage@inktank.com>
Signed-off-by: Yan, Zheng <zheng.z.yan... - 01:14 AM Revision bca32ef5 (ceph): Merge pull request #252 from ceph/wip-fsx-krbd
- rbd_fsx: expose krbd and related fsx options
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 01:13 AM Revision ff3987d4 (ceph): Merge pull request #1766 from ceph/wip-fsx-krbd
- krbd mode for librbd_fsx
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 01:10 AM Revision ce852b29 (ceph): Revert "ReplicatedPG: block scrub on blocked object contexts"
- This reverts commit e66f2e36c06ca00c1147f922d3513f56b122a5c0.
Reviewed-by: Sage Weil <sage@inktank.com>
0f3235d46c8f... - 01:01 AM Revision f1d412c3 (ceph): doc: Common graph used in 2 quick start files
- The graph in quick-ceph-deploy.rst applies to
quick-start-preflight.rst.
The graph in deploy seems more complete, so ...
05/07/2014
- 11:53 PM Feature #8307 (Resolved): Creating a pool with erasure code allows me to create invalid ec profil...
- Lets much around with the cmdline
[root@storage ~]# ceph osd pool create ec 128 128 erasure erasure-code-k=9 erasu... - 10:14 PM Revision db8873b6 (ceph): rgw: fix stripe_size calculation
- Fixes: #8299
Backport: firefly
The stripe size calculation was broken, specifically affected cases
where we had manif... - 10:14 PM Revision b1805e74 (ceph): Merge pull request #1780 from ceph/wip-8299
- rgw: fix stripe_size calculation
Reviewed-by: Sage Weil <sage@inktank.com> - 10:04 PM Revision 0e685c68 (ceph): rgw: send user manifest header field
- Fixes: #8170
Backport: firefly
If user manifest header exists (swift) send it as part of the object
header data.
Sig... - 10:03 PM Revision e0fb2e63 (ceph): rgw: cut short object read if a chunk returns error
- Fixes: #8289
Backport: firefly, dumpling
When reading an object, if we hit an error when trying to read one of
the ra... - 10:03 PM Revision 6a06f320 (ceph): Merge pull request #1776 from ceph/wip-8289
- rgw: cut short object read if a chunk returns error
Reviewed-by: Sage Weil <sage@inktank.com> - 09:53 PM Revision 328665db (ceph): rgw: send user manifest header field
- Fixes: #8170
Backport: firefly
If user manifest header exists (swift) send it as part of the object
header data.
Sig... - 09:53 PM Revision 7f5de5d0 (ceph): Merge pull request #1773 from ceph/wip-8170
- rgw: send user manifest header field
Reviewed-by: Sage Weil <sage@inktank.com> - 09:47 PM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- Sage Weil wrote:
> I think cache mode changes will cause similar problems. Let's add a pg_pool_t epoch_t that indic... - 02:34 PM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- Greg Farnum wrote:
> I don't think this lets us handle arbitrary changes in the overlay system. Consider two clients... - 02:27 PM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- I don't think this lets us handle arbitrary changes in the overlay system. Consider two clients a and b, a cache OSD,...
- 02:19 PM Bug #8305: objecter, osd: pool overlay change should trigger op resend
- I think cache mode changes will cause similar problems. Let's add a pg_pool_t epoch_t that indicates the last policy...
- 02:15 PM Bug #8305 (Resolved): objecter, osd: pool overlay change should trigger op resend
- If the client is sending ops a, b, c, d, and a map is received changing the overlay, ordering can break. For example...
- 09:47 PM Revision 20383e35 (ceph): client: add asok command to kick sessions that were remote reset
- Fixes: #8021
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
(cherry picked from commit 09a1bc5a4601d356b9cc69be854... - 09:46 PM Revision d1307631 (ceph): vstart.sh: fix client admin socket path
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:27 PM Revision cdb0fac2 (ceph): client: add asok command to kick sessions that were remote reset
- Fixes: #8021
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
(cherry picked from commit 09a1bc5a4601d356b9cc69be854... - 09:23 PM Revision 14ebe9b5 (ceph): osd: throttle snap trimmming with simple delay
- This is not particularly smart, but it is *a* knob that lets you make
the snap trimmer slow down. It's a flow and a ... - 09:23 PM Revision 4e5e41de (ceph): osd: throttle snap trimmming with simple delay
- This is not particularly smart, but it is *a* knob that lets you make
the snap trimmer slow down. It's a flow and a ... - 09:22 PM Revision 4a91196b (ceph): osd: throttle snap trimmming with simple delay
- This is not particularly smart, but it is *a* knob that lets you make
the snap trimmer slow down. It's a flow and a ... - 09:02 PM Revision 3ba2e228 (ceph): mon/MonClient: remove stray _finish_hunting() calls
- Callig _finish_hunting() clears out the bool hunting flag, which means we
don't retry by connection to another mon pe... - 09:01 PM Revision 383f6440 (ceph): osd/ReplicatedPG: fix trim of in-flight hit_sets
- We normally need to stat the hit_set to know how many bytes to adjust the
stats by. If the hit_set was just written,... - 09:01 PM Revision ef35448e (ceph): osd/ReplicatedPG: fix whiteouts for other cache mode
- We were special casing WRITEBACK mode for handling whiteouts; this needs to
also include the FORWARD and READONLY mod... - 07:49 PM Revision 5ae93dd2 (ceph): Merge pull request #32 from ceph/wip-8284
- Reviewed-by: Samuel Just <sam.just@inktank.com>
- 07:45 PM Revision 0ee409b6 (ceph): osd: Remove classic scrub code since Argonaut osd can't join
- Fixes: #7553
Signed-off-by: David Zafman <david.zafman@inktank.com> - 07:18 PM Revision 81c74182 (ceph): ECUtil.h: clarify calculation with braces
- Fix for cppcheck issue:
[src/osd/ECUtil.h:61]: (style) Clarify calculation
precedence for '%' and '?'.
Signed-off... - 06:20 PM Revision bb170c1b (ceph): Merge pull request #249 from ceph/wip-8284
- rados.py: Add pool_snaps option for ceph_test_rados test command
- 06:05 PM Revision 499b29a3 (ceph): Merge pull request #1783 from guangyy/folder-merge-doc
- Update doc to reflect the bahavior change for filestore_merge_threshold setting.
Reviewed-by: Samuel Just <sam.just@... - 06:05 PM Revision 7b1b553d (ceph): Merge pull request #1784 from ceph/wip-da-cleanup-includes
- Cleanup some included headers
Reviewed-by: Samuel Just <sam.just@inktank.com> - 06:01 PM Revision 13f54b7d (ceph): PG::start_peering_interval: use check_new_interval for same_interval_since
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 05:57 PM Bug #8239: log [WRN] : slow request 30.404834 seconds old, received at 2014-04-26 04:05:56.539287...
- per irc conversation with sjusthm, about active+clean+scrubbing+deep pg stuck state. attaching my logs and other info
- 04:57 PM devops Feature #8306 (Resolved): separate ceph.rpm into ceph and ceph-common
- 04:52 PM Revision 5752d76e (ceph): rgw_acl_swift.h: fix #define header guard
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 04:02 PM Revision 8059c9fb (ceph): rgw_rest_metadata.cc: fix -Wparentheses-equality
- Fix for:
warning: equality comparison with extraneous parentheses
[-Wparentheses-equality]
Signed-off-by: Danny Al-... - 04:00 PM Revision 8a0c0163 (ceph): ReplicatedPG.cc: fix -Wparentheses
- Fix for:
warning: using the result of an assignment as a condition
without parentheses [-Wparentheses]
Signed-off-... - 03:32 PM Revision a0f59df1 (ceph): test_rgw_manifest.cc: fix VLA of non-POD element type
- Use vector to fix:
test/rgw/test_rgw_manifest.cc:184:20: error: variable length array
of non-POD element type 'RGWOb... - 03:13 PM rgw Bug #8299 (Resolved): rgw: broken range read with new style manifest
- 03:03 PM rgw Bug #8289 (Resolved): rgw: memory not freed during in-progress read (dumpling)
- 02:52 PM rgw Bug #8170 (Resolved): rgw: missing manifest response header when reading swift user manifest object
- 02:01 PM Bug #8278 (Resolved): monclient: failure to retry after ill-timed connection reset during auth
- 02:00 PM Bug #8283 (Resolved): osd: hit_set_trim cannot stat recently written hitsets
- 01:59 PM Bug #8296 (Resolved): ./test/osd/RadosModel.h: 855: FAILED assert(0 == "racing read got wrong ver...
- 01:55 PM Revision b105a07a (ceph): rbd_fsx: expose krbd and related fsx options
- Expose
-K (enable krbd mode) through 'krbd',
-Z (use direct IO) through 'direct_io',
-U (disable randomized striping... - 01:47 PM Revision 6c49d6e1 (ceph): Merge pull request #1775 from ceph/wip-rbd-clang
- fix clang-analyzer warnings in rbd and objectcacher
Reviewed-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 01:34 PM Revision 3d280d6b (ceph): Merge pull request #1782 from xinglin/coverity-fixes
- test/libcephfs/test.cc: free cmount structure before return
Reviewed-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 01:30 PM Revision ab9de9c0 (ceph): test_librbd_fsx: wire up O_DIRECT mode
- Wire up O_DIRECT mode (-Z) for krbd, to have a workaround for possible
problems with BLKDISCARD leaving stale entries... - 01:30 PM Revision 817985b4 (ceph): test_librbd_fsx: align temporary buffers allocated in check_clone()
- check_clone() allocates temporary good_buf and temp_buf with malloc(),
which is not good enough for krbd with O_DIREC... - 01:30 PM Revision 8d41f86f (ceph): test_librbd_fsx: update usage
- Update usage to include all options and flags.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com> - 01:30 PM Revision d13e32e2 (ceph): test_librbd_fsx: move prterrcode() and simple_err()
- Move prterrcode() and simple_err() so that all printing functions are
close together.
Signed-off-by: Ilya Dryomov <i... - 01:30 PM Revision 7df50ecd (ceph): test_librbd_fsx: align temp_buf by readbdy instead of writebdy
- temp_buf is used for reads, so align it by readbdy.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com> - 01:30 PM Revision 3513ba0a (ceph): test_librbd_fsx: use posix_memalign() to allocate aligned buffers
- Use posix_memalign() to allocate good_buf and temp_buf, which must be
writebdy and readbdy aligned respectively. Usi... - 01:30 PM Revision d63808ed (ceph): test_librbd_fsx: make resizes sector-size aligned
- In preparation for krbd mode support, change check_trunc_hack() to
resize to a sector-size aligned value. The kernel... - 01:30 PM Revision fef984b9 (ceph): test_librbd_fsx: add holebdy option
- In preparation for krbd mode support, provide an option to specify
alignment for discards. The kernel will reject di... - 01:30 PM Revision d5daf718 (ceph): test_librbd_fsx: add a flag to disable randomized striping
- In preparation for krbd mode support, introduce an option to disable
randomized striping. The kernel as of 3.15 does... - 01:30 PM Revision 421e6c56 (ceph): test_librbd_fsx: add krbd mode support
- Add krbd mode support (-K) to test krbd in the same way librbd is
tested. This introduces a dependency on libkrbd an... - 01:30 PM Revision c4a764cc (ceph): test_librbd_fsx: fix a bug in docloseopen()
- docloseopen() always opens $iname image. This is bad, because the
image we had opened could have been something like... - 12:53 PM Feature #7553: Remove classic scrub
- 12:52 PM Feature #8284 (Resolved): Add --pool_snaps rados tests to teuthology
- teuthology:
ea3bef1e1dbf21a6117dea906f1e500db2b6af76
ceph-qa-suite:
99e67abc947f3ebb87bbfdf8032c9c2136e12de7 - 12:22 PM Bug #8229: 0.80~rc1: OSD crash (domino effect)
- I mean stopped on "host1", physically moved to "host2" and started there. Crush map will be updated automatically and...
- 09:52 AM Bug #8229: 0.80~rc1: OSD crash (domino effect)
- What do you mean by brought up on another server?
- 12:21 PM Revision 99400f82 (ceph): osdmaptool.cc: cleanup included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:21 PM Revision a5e0d802 (ceph): monmaptool.cc: cleanup included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:17 PM Revision 537385cc (ceph): ceph_osdomap_tool.cc: cleanup included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:15 PM Revision d57561a8 (ceph): ceph_monstore_tool.cc: cleanup included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:14 PM Revision e2e3d1d0 (ceph): ceph_filestore_tool.cc: remove not needed includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:12 PM Revision ea6df887 (ceph): ceph_kvstore_tool.cc: cleanup includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:12 PM Revision 86206098 (ceph): ceph_filestore_dump.cc: cleanup includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:03 PM Revision 16e86aef (ceph): mon_store_converter.cc: remove not needed includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:03 PM Revision 2c33ace3 (ceph): dupstore.cc: remove not needed include of <iostream>
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:02 PM Revision b1f4cd45 (ceph): rest_bench.cc: remove not needed includes, re-order includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:56 AM Revision bc166f3a (ceph): psim.cc: remove not used includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:55 AM Revision de252c11 (ceph): scratchtool.c: remove not needed include of <pthread.h>
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:55 AM Revision 4da88945 (ceph): radosacl.cc: remove include of <iostream>, re-order includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:53 AM Revision 24a047ec (ceph): ceph_conf.cc: cleanup includes, remove not needed headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:53 AM Revision c0dcd231 (ceph): ceph_authtool.cc: remove not needed includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision e66aec6e (ceph): ceph_monstore_tool.cc: remove twice included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision 71b340a0 (ceph): ceph_osdomap_tool.cc: remove some twice included headers
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision a8a2b564 (ceph): AuthMonitor.cc: remove twice included header, resorted includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision ddb9ce0e (ceph): MDSMonitor.cc: remove twice included header, resorted includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision 8ab3232a (ceph): rgw_admin.cc: remove twice included header, resort includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:27 AM Revision e2b4d417 (ceph): ceph_monstore_tool.cc: remove not needed includes
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 10:23 AM devops Feature #8303 (Rejected): ceph-extra packages for newer Ubuntu versions
- http://ceph.com/packages/ceph-extras/debian/dists/
It's missing packages for Saucy (13.10) and Trusty (14.04). The... - 10:19 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
- Sam:
This was tested with debug osd = 20, debug filestore = 20, debug ms = 1.
Unfortunately some of the other d... - 09:55 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
- You need to reproduce this with debug osd = 20, debug filestore = 20, debug ms = 1.
- 08:55 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
- Well, the OSDs are supposed to throttle incoming client requests in order to prevent this, but they're generally bett...
- 07:54 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
- Greg:
That very well may be the case. What I am seeing is that if each test in the set of tests is restricted to ... - 07:13 AM Bug #8301: Suicide Timeout on Cache Tier OSDs
- Sounds like you're just putting too many ops into the cluster too quickly. Is there anything making you think differe...
- 06:04 AM Bug #8301 (Resolved): Suicide Timeout on Cache Tier OSDs
- This appears to happen in both firefly and master. During 4MB random write tests on an RBD volume with a Ceph cache ...
- 09:54 AM Bug #8228 (Can't reproduce): 0.80~rc1: OSD crash: segfault in libtcmalloc.so.4.1.2
- In that case, marking can't reproduce. Let us know if it recurs.
- 09:54 AM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- It should take something rather extreme to create inconsistent objects in the first place. Did anything interesting ...
- 09:50 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- yes, we build our own kernels, so patching/testing is possible.
- 09:12 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- Hi Markus,
Judging by debug output, I'm assuming you can build your own kernels? - 08:02 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- dmesg shows nothing special without debuggung enabled. i attached debug output of kernel as well as the osdmap. can i...
- 07:25 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- I should note that the MDS is behaving fine according to that log; Zheng thinks there's been a regression in the CRUS...
- 07:12 AM Linux kernel client Bug #8300: Regression in 3.14: "No such device or address" reading file content
- Are there any messages in dmesg on the affected node? Do you have debugfs enabled?
- 04:28 AM Linux kernel client Bug #8300 (Resolved): Regression in 3.14: "No such device or address" reading file content
- On kernel 3.14.2 and ceph 0.72.2, reading from some files gives the error message "No such device or address". Kernel...
- 09:41 AM Revision f9a91f2b (ceph): Update doc to reflect the bahavior change for filestore_merge_threshold...
- Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
- 09:08 AM Bug #8232: Race condition during messenger rebind
- Hmm, so this is also following a mark_down_all. That would certainly explain why we no longer have any data on our pe...
- 08:52 AM devops Bug #7311 (Closed): GPG/packaging failures
- this is no longer happening. Will re-open if it does so again.
- 06:06 AM Revision 523619b0 (ceph): Merge pull request #1532 from ceph/wip-fast-dispatch
- fast dispatch
This series adds an ms_fast_dispatch interface to the Messenger/Dispatcher, designed so that you can di... - 05:23 AM Revision 5e5a0867 (ceph): Merge remote-tracking branch 'origin/master' into wip-fast-dispatch
- Conflicts:
src/osd/OSD.cc - 04:44 AM Revision 0d3cdb9f (ceph): test/libcephfs/test.cc: free cmount structure before return
- call ceph_shutdown to free cmount structure before return
Signed-off-by: Xing Lin <xinglin@cs.utah.edu> - 02:06 AM Revision 25d2469f (ceph): client: leave NULL dentry in place on ENOENT during lookup
- If we get a NULL lookup result, unlink the inode but leave the dentry
in place.
Signed-off-by: Sage Weil <sage@inkta... - 02:06 AM Revision 334c43f5 (ceph): client: avoid blindly removing dentries
- MetaRequests may have references to these dentries. Instead of removing
them and tearing down the directory, just un... - 02:06 AM Revision 8fa5408f (ceph): client: handle traceless rename in insert_trace, not verify_reply_trace
- The insert_trace() logic is about managing local cache consistency; the
verify_reply_trace() is about retrying reques... - 02:06 AM Revision cc65c392 (ceph): client: add debugging around traceless reply failures
- Tracking down #5021
Signed-off-by: Sage Weil <sage@inktank.com> - 12:07 AM Revision 2dff44a7 (ceph): Merge pull request #33 from ceph/wip-8297-wusui
- 2-workload testrgw needs to be sequential.
- 12:04 AM Revision 545d8ad1 (ceph): rgw: extend manifest to avoid old style manifest
- In case we hit issue #8269 we'd like to avoid creating an old style
manifest. Since we need to have parts that use di...
05/06/2014
- 11:55 PM Revision 9968b938 (ceph): rgw: fix stripe_size calculation
- Fixes: #8299
Backport: firefly
The stripe size calculation was broken, specifically affected cases
where we had manif... - 11:50 PM Revision 4a3728d1 (ceph): 2-workload testrgw needs to be sequential.
- Fixes: 8297
Signed-off-by: Warren Usui <warren.usui@inktank.com> - 11:10 PM Revision 6c2b1732 (ceph): mds: handle export freezen race
- handle following sequence of events:
- subtree becomes frozen, C_MDC_ExportFreeze is queued.
- export is cancelled
- ... - 11:10 PM Revision 7d1fd669 (ceph): mds: maintain auth bits during replay
- Objects' STATE_AUTH bits are set when replaying EImportStart event.
MDCache::trim_non_auth_subtree() clear objects' S... - 11:10 PM Revision 5b86a13c (ceph): mds: send dentry unlink message to replicas of stray dentry
- stray dentry may have more replicas than the unlinked dentry has.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 11:10 PM Revision 7a066f88 (ceph): mds: include authpinned objects in remote authpin request
- Server::handle_slave_auth_pin() may drop old authpins if it encounters
object that is not authpinable. So it is bette... - 11:10 PM Revision 6d6d1889 (ceph): mds: fix frozen inode check in MDCache::handle_discover()
- When MDCache::handle_discover() encounters a frozen dirfrag, it should
proceed if the dirfrag is being merged, but th... - 11:10 PM Revision f386e163 (ceph): mds: pre-allocate inode numbers less frequently
- no need to refill the pre-allocated inode numbers each time an inode
number is used.
Signed-off-by: Yan, Zheng <zhen... - 11:10 PM Revision c4f0f051 (ceph): mds: tolerate bad sessionmap during journal replay
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 11:10 PM Revision a85bf8c9 (ceph): mds: remove mdsdir in the final step of shutdown MDS
- Otherwise we may get bad subtree map if we restart the MDS before
the shutdown process finishes.
Signed-off-by: Yan,... - 11:10 PM Revision a2caea7c (ceph): mds: clear aborted flag before rollback slave requests
- There is a special case that the MDRequest needs to be preserved after
rolling back slave rename. The preserved MDReq... - 11:10 PM Revision f7541067 (ceph): mds: fix-up inode's fragstat/rstat according its dirfrags
- Extend the code that fixup inode's fragstat/rstat to handle multiple
dirfrags
Signed-off-by: Yan, Zheng <zheng.z.yan... - 11:10 PM Revision 4e844c94 (ceph): mds: encode dirfrag base in cache rejoin ack
- Makes sure recovering MDS get uptodate fragstat/rstat for subtree root
dirfrags. it's required the codes that fix-up ... - 11:10 PM Revision 5283d80d (ceph): mds: ignore stale rstat/fragstat when splitting dirfrag
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 11:10 PM Revision da173949 (ceph): mds: fix root and mdsdir inodes' rsubdirs
- inode rstat accounts inode itself.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 11:10 PM Revision 6e3501bd (ceph): mds: fix _rollback_repair_dir()
- _rollback_repair_dir() may increase dirfrag's rfiles/rsubdirs twice.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 11:10 PM Revision 22abd7b0 (ceph): mds: cancel fragmenting dirfrags when cluster is degraded
- when cluster is degraded, acquiring locking can take long time.
It is not good to keep dirfrags in frozen state for a... - 11:10 PM Revision a09070ab (ceph): mds: allow negetive rstat
- When splitting dirfrag, delta rstat is always added to the first new
dirfrag. Ancestors of the dirfrag may have nagti... - 10:55 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- No, only rbd.ko and libceph.ko.
- 10:45 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- In theory I could but I have so little time for this that I can't promise anything...
Let's see if issue comes back ... - 10:29 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Can you build your own debug kernels? If so, I'll come up with a debug
patch to at least try to isolate the problem. - 10:11 PM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- No idea how to replicate on healthy cluster.
Problem seems to be gone for now...
- 01:34 PM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- Can you reproduce with debug osd = 20, debug filestore = 20, debug ms = 1?
- 10:09 PM Bug #8229: 0.80~rc1: OSD crash (domino effect)
- Sorry, I can't reproduce on healthy cluster and users won't forgive me for potential downtime...
I think it could be... - 01:20 PM Bug #8229: 0.80~rc1: OSD crash (domino effect)
- Can you reproduce with
debug osd = 20
debug filestore = 20
debug ms = 1
on all osds?
? - 10:03 PM Bug #8228: 0.80~rc1: OSD crash: segfault in libtcmalloc.so.4.1.2
- I only had it once, it never happened again as far as I'm aware...
No ideas what could be the cause.
I suspect that... - 01:09 PM Bug #8228: 0.80~rc1: OSD crash: segfault in libtcmalloc.so.4.1.2
- Is there a straightforward way to reproduce this?
- 10:01 PM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
- Many thanks, Zheng. Looks like it is commit commit:09a1bc5.
- 02:27 AM CephFS Bug #8291 (Resolved): 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
- current master branch contains fix for this issue (use "kick_stale_sessions" admin socket command)
- 01:33 AM CephFS Bug #8291: 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
- Another MDS have no problem reconnecting to client:...
- 01:29 AM CephFS Bug #8291 (Resolved): 0.80: fuse client hangs after wake-up from suspend until restart of active MDS
- When computer with CephFS mounted using ceph-fuse wakes-up from suspend the @/mnt/ceph@ mount is not responding (i.e....
- 09:06 PM Revision 09beebe3 (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... - 09:00 PM Revision e7df73dd (ceph): osd: Prevent divide by zero in agent_choose_mode()
- Fixes: #8175
Backport: firefly
Signed-off-by: David Zafman <david.zafman@inktank.com>
Signed-off-by: Sage Weil <sage... - 09:00 PM Revision 022d467b (ceph): osd, common: If agent_work() finds no objs to work on delay 5 (default)...
- Add config osd_agent_delay_time of 5 seconds
Honor delay by ignoring agent_choose_mode() calls
Add tier_delay to logg... - 09:00 PM Revision 6a55c3bc (ceph): osd/ReplicatedPG: agent_work() fix next if finished early due to start_max
- Backport: firefly
Signed-off-by: David Zafman <david.zafman@inktank.com>
(cherry picked from commit 9cf470cac8dd4d8f... - 08:29 PM Revision 14650b28 (ceph): PG: only complete replicas should count toward min_size
- Backport: emperor,dumpling,cuttlefish
Fixes: #7805
Signed-off-by: Samuel Just <sam.just@inktank.com>
Signed-off-by: S... - 08:18 PM Revision bd8e026f (ceph): rgw: don't allow multiple writers to same multiobject part
- Fixes: #8269
Backport: firefly, dumpling
A client might need to retry a multipart part write. The original thread
mi... - 08:08 PM Revision b7134c9a (ceph): Merge pull request #1774 from ceph/wip-8296
- osd/ReplicatedPG: fix whiteouts for other cache mode
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:06 PM Revision 1b899148 (ceph): Fix clone problem
- When clone happened, the origin header also will be updated in GenericObjectMap,
so the new header wraper(StripObject... - 08:03 PM Bug #8232: Race condition during messenger rebind
- > Just to be clear, there is a client and server concept in the messenger, but in the case of the OSDs they are peers...
- 12:00 PM Bug #8232: Race condition during messenger rebind
- Guang Yang wrote:
> After understanding more of the messenger code, I found the above analysis was wrong in terms of... - 05:18 AM Bug #8232: Race condition during messenger rebind
- After understanding more of the messenger code, I found the above analysis was wrong in terms of the reconnecting log...
- 07:59 PM Revision 091d1fe4 (ceph): Revert "Revert "Clean up remote.py and misc.py changes.""
- This reverts commit 02504c3fd27d788e2e446369015b14cbf259a8d2.
- 07:59 PM Revision 36b07b8a (ceph): Use SFTPClienti get for long reads/writes
- Modified remote.py to use the paramiko SFTPClient get
method to extract long files (mostly tar files) from
the remote... - 07:59 PM Revision 8bed6ab6 (ceph): FIx mktemp dir and redundant Paramiko connecting.
- Use previously initialized connection for sftp_get calls.
Use local directory for tarball temp file location. - 07:59 PM Revision 01cf3671 (ceph): Fix linter errors
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:59 PM Revision a1838b2a (ceph): Rewrite most file-retrieval functions
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:59 PM Revision 38578b80 (ceph): Revert "Restrict paramiko to old versions for now"
- This reverts commit c86df77aa68dc5331b98df3fa038faee77c578ad.
- 07:59 PM Revision 02684870 (ceph): Revert "Revert "Handle raw data I/O.""
- This reverts commit 0c8a3e2334631f5fc20cb7933d0005353ea6a46e.
- 07:06 PM Revision 7b1eec94 (ceph): Use longer varchar for locked_by in DB.
- Signed-off-by: Sandon Van Ness <sandon@inktank.com>
- 07:03 PM Revision 03b0d1cf (ceph): rgw: cut short object read if a chunk returns error
- Fixes: #8289
Backport: firefly, dumpling
When reading an object, if we hit an error when trying to read one of
the ra... - 06:39 PM Revision 2d5d3097 (ceph): Pipe: wait for Pipes to finish running, instead of just stop()ing them
- Add a stop_and_wait() function that, in addition to closing the Pipe and killing
its socket, waits for any fast_dispa... - 06:21 PM Revision 6ec99f7a (ceph): librbd: check return value during snap_unprotect
- This would only fail if the header object was corrupted, so it's
unlikely to occur in practice.
Signed-off-by: Josh ... - 06:18 PM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
- Bumped pri, assigned to Ilya for investigation
- 05:10 PM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
- Seeing this with master during RBD tiering tests only on the cache tier OSDs. Eventually during heavy RBD random 4MB...
- 06:11 PM Revision 6f2eddaa (ceph): ObjectCacher: remove useless assignment
- left is not read after the break. Caught by clang-analyzer.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 06:01 PM Revision 3e387d62 (ceph): osd/ReplicatedPG: fix whiteouts for other cache mode
- We were special casing WRITEBACK mode for handling whiteouts; this needs to
also include the FORWARD and READONLY mod... - 05:49 PM Revision 650051cd (ceph): Merge pull request #1601 from ceph/wip-7576
- osd: prevent pg map epochs from lagging too far behind
Reviewed-by: Samuel Just <sam.just@inktank.com> - 05:30 PM Revision ea3bef1e (ceph): rados.py: Add pool_snaps option for ceph_test_rados test command
- Fixes: #8284
Signed-off-by: David Zafman <david.zafman@inktank.com> - 05:12 PM Revision 2b48e52c (ceph): Merge pull request #1748 from onlyjob/docs
- sample.ceph.conf update:
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 05:10 PM Revision 9c0e92f0 (ceph): Merge pull request #1653 from ceph/wip-7499
- rgw, radosgw-admin: bucket link uses bucket instance id now
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 05:04 PM rgw Bug #8269 (Fix Under Review): rgw: corrupted multipart object
- 09:05 AM rgw Bug #8269 (In Progress): rgw: corrupted multipart object
- 05:03 PM rgw Bug #8289 (Fix Under Review): rgw: memory not freed during in-progress read (dumpling)
- 09:04 AM rgw Bug #8289 (In Progress): rgw: memory not freed during in-progress read (dumpling)
- 05:00 PM Revision 5cc56860 (ceph): rgw: send user manifest header field
- Fixes: #8170
Backport: firefly
If user manifest header exists (swift) send it as part of the object
header data.
Sig... - 04:56 PM rgw Bug #8299 (Fix Under Review): rgw: broken range read with new style manifest
- 04:13 PM rgw Bug #8299: rgw: broken range read with new style manifest
- This will trigger only when dealing with manifests that describe multiple parts.
- 03:40 PM rgw Bug #8299 (Resolved): rgw: broken range read with new style manifest
- 04:48 PM Feature #8284: Add --pool_snaps rados tests to teuthology
- 04:45 PM Feature #7553 (Fix Under Review): Remove classic scrub
- 04:34 PM Revision bdd1b5ac (ceph): Merge pull request #251 from ceph/wip-8295
- rgw: fix indentation for cache_pools
- 04:22 PM Revision 4ad23dc5 (ceph): rgw: fix indentation for cache_pools
- Fixes: #8295
Signed-off-by: Sage Weil <sage@inktank.com> - 02:20 PM rgw Bug #8251: radosgw-agent does not sync objects uploaded to recreated buckets
- It appears this is due the replica log api using the bucket name only, not the bucket-instance which is given to it b...
- 02:12 PM Revision f31e3ee0 (ceph): Merge pull request #1768 from daniel-j-h/code_quality
- Variable length array of std::strings (not legal in C++) changed to std::vector<std::string>
Reviewed-by: Sage Weil ... - 02:04 PM Revision e65a9da9 (ceph): Revert "Fix installation into user home directory, broken by d3f0c0b"
- This reverts commit 7539281037ce7a755ac8661ecb15aea32e5f79f6.
This breaks mount.fuse.ceph installation. - 02:03 PM Revision b78644e7 (ceph): 0.80
- 02:01 PM devops Bug #6746 (Resolved): ceph-release rpm not playing well with yum-plugin-priorities
- you can now specify the repository information when installing and just apply whatever priority you need in the confi...
- 02:01 PM Bug #8175 (Resolved): Some values of target_max_objects for tiering will crash OSDs
- 02:01 PM Bug #8113 (Resolved): agent_work can be continuously rescheduled during recovery while most objec...
- 01:58 PM devops Bug #7157 (Duplicate): ceph-disk list fails in encrypted disk setup
- 01:57 PM devops Bug #7227 (Can't reproduce): ceph-create-keys upstart script fails
- 01:52 PM devops Bug #7605 (Resolved): statup script /etc/init.d/ceph has incorrect slash
- commit:4cf9a73bacb73706fff66110528b733e9ec80b21
- 01:48 PM devops Bug #6216 (Resolved): rpm missing package when junit not installed
- 01:48 PM devops Bug #5479 (Resolved): Append our built packages with some sort of inktank/ceph identifier
- 01:45 PM devops Bug #7004 (Rejected): qemu: rhel and centos qemu packages should depend on librbd
- 01:43 PM devops Bug #7552 (Resolved): dregs of mkcephfs still live on
- 01:42 PM devops Bug #8095 (Resolved): [chef] No version specified, and no candidate version available for libmpic...
- 01:41 PM Bug #8058 (Resolved): "LibRadosTierECPP.FlushWriteRaces" failed in upgrade:dumpling-x:parallel-fi...
- 01:40 PM Bug #8180 (Duplicate): osd.3 crashed in upgrade:dumpling-x:stress-split-firefly-distro-basic-vps
- 01:40 PM Bug #8204 (Duplicate): "timed out waiting for admin_socket to appear after osd.5 restart" in upgr...
- 01:40 PM devops Bug #6966 (Fix Under Review): ceph-disk: prepare --dmcrypt failing
- https://github.com/ceph/ceph/pull/1777
- 01:38 PM devops Bug #8298 (Resolved): missing emperor ceph package
- thanks for the quick fix ! Not sure what it was but it was fixed within minutes which is impressive ;-)...
- 01:33 PM devops Bug #8298 (Resolved): missing emperor ceph package
- The ceph package is missing from ...
- 01:37 PM Bug #8164 (Duplicate): "[ERR] 4.15 0 tried to pull" in upgrade:dumpling-x:stress-split-firefly---...
- 01:37 PM Bug #8206 (Duplicate): "osd.4 ...[ERR] : 3.14 push" in upgrade:dumpling-x:stress-split-firefly-di...
- 01:36 PM Bug #8207 (Duplicate): "[ERR] 3.6 missing primary copy.." in upgrade:dumpling-x:stress-split-fire...
- 01:36 PM devops Bug #8151 (Rejected): Perms on /etc/ceph/ceph.client.admin.keyring wrong on some nodes after install
- 600 looks correct to me?
- 01:35 PM devops Bug #5778 (Resolved): gitbuilders use cryptopp instead of nss libraries
- 01:30 PM Bug #7779 (Resolved): osd: object file can have too many xattrs, get E2BIG
- 01:30 PM Bug #7976 (Duplicate): 4.8 missing primary copy of ..., unfound (dumpling)
- 01:29 PM Bug #5818: leveldb 1.12: hang on shutdown (mon)
- 01:27 PM Bug #5818 (Duplicate): leveldb 1.12: hang on shutdown (mon)
- This is related to #5847, but I can't link them due to limitations in RedMine
- 01:29 PM Bug #8021 (Duplicate): osd: ENOENT on clone on dumpling
- #8162
- 01:28 PM Bug #7805 (Resolved): emperor can go active with < min_size non-incomplete peers since we check a...
- 01:25 PM Bug #8104: OSD: changing min_size to larger than the acting set does not make the PG go inactive
- make sure this is consistent with check_new_interval()
- 01:24 PM Bug #8103 (Won't Fix): pool has too few PGs warning misleading when using cache pools
- no simple way to avoid this false positive but still maintain this warning. and it is useful.
- 01:21 PM Bug #8036 (Can't reproduce): levedb: throws std::bad_allow on 14.04
- 01:21 PM Bug #8176 (Resolved): Change target_max_objects/target_max_bytes has no immediate effect
- 01:18 PM Bug #7744 (Need More Info): osd: assert(last_e.version.version < e.version.version)
- we need a full log to see how this happens on dumpling.
Brian, to work around this and get your osd up, you need t... - 01:15 PM Bug #7576 (Resolved): osd: large skew in pg epochs (dumpling)
- 11:05 AM Bug #7576 (Pending Backport): osd: large skew in pg epochs (dumpling)
- 01:13 PM Bug #7916 (Can't reproduce): ceph_test_rados got ENOENT on ec pool + thrashing
- 01:13 PM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- 01:10 PM Bug #8123 (Can't reproduce): OSD: received operation against clone which was not backfilled (but ...
- 01:10 PM Bug #8199 (Resolved): rados unit test failure: LibRadosTwoPoolsECPP.FlushTryFlushRaces hang
- 01:07 PM Bug #8237 (Resolved): ceph centos 6 repo broken
- 01:07 PM Bug #8272 (Duplicate): leveldb 1.12.0 broken on el6
- This duplicates #5847, but I can't link them due to limitations in RedMine
- 01:07 PM Bug #7986 (Can't reproduce): 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/2...
- 01:06 PM Bug #8296 (Pending Backport): ./test/osd/RadosModel.h: 855: FAILED assert(0 == "racing read got w...
- 10:32 AM Bug #8296 (Resolved): ./test/osd/RadosModel.h: 855: FAILED assert(0 == "racing read got wrong ver...
- ...
- 01:06 PM Bug #8282 (Resolved): ceph-objectstore-test segv
- 11:33 AM Bug #8283 (Pending Backport): osd: hit_set_trim cannot stat recently written hitsets
- 10:54 AM Revision cdbbf86f (ceph): doc: Fixed artifacts from merge.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:54 AM Revision a31b9e9c (ceph): doc: Added sudo to setenforce. Restored merge artifact.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:53 AM Revision 51582722 (ceph): doc: Added erasure coding and cache tiering notes. Special thanks to Lo...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:43 AM Bug #8273 (Duplicate): leveldb 1.12.0 broken on el6
- 09:42 AM devops Feature #5847: Build own versions of most recent leveldb for all supported platforms.
- 09:08 AM rgw Bug #8194: rgw: test_region_copy_object fails with erasure coding
- commit at ceph-qa-suite:ee69c7a4f3b425377bd20b307d674ecc70904c0b
- 09:07 AM rgw Bug #8194 (Resolved): rgw: test_region_copy_object fails with erasure coding
- 08:18 AM Bug #8294 (Rejected): "Error EINVAL" in upgrade:dumpling-dumpling-testing-basic-vps
- Logs are in /a/yuriw/231774...
- 07:51 AM Revision 08a4e888 (ceph): Variable length array of std::strings (not legal in C++) changed to std...
- Signed-off-by: Daniel J. Hofmann <daniel@trvx.org>
- 06:28 AM rgw Bug #8293 (Resolved): admin api get quotas
- I have been trying to implement the different admin api quota commands. I am getting a 200 answer but no body when r...
- 03:32 AM devops Bug #8292 (Resolved): ceph-disk prepare output not explicit on too small disk
- While trying to use "ceph-disk prepare /dev/sdb", where /dev/sdb has only a 1Go partition,
we get this output
STD... - 12:20 AM Revision d158c156 (ceph): Merge pull request #250 from ceph/wip-fix-thrasher
- ceph_manager: reset osd weights to 1 when waiting for clean
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 12:05 AM Revision 33b4bfc7 (ceph): ceph_manager: reset osd weights to 1 when waiting for clean
- If we leave the weights adjusted, we can get PGs stuck in a remapped state
because we are probabilistically rejecting...
05/05/2014
- 11:55 PM Revision 38408f6b (ceph): Merge pull request #1770 from ceph/wip-8290
- client: check snap_caps in Incode::is_any_caps()
Reviewed-by: Sage Weil <sage@inktank.com> - 11:46 PM Revision ae434a35 (ceph): client: check snap_caps in Incode::is_any_caps()
- Fixes: #8290
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 11:26 PM Revision 7f3a4206 (ceph): Merge pull request #1764 from eile/master
- Fix installation into user home directory, broken by d3f0c0b
Reviewed-by: Sage Weil <sage@inktank.com> - 11:05 PM devops Bug #7598: ceph-disk-activate error with ceph-deploy
- I faced this issue consistently in my setup.
ubuntu@ip-10-15-16-160:~$ sudo ceph-deploy osd activate ip-10-15-16-1... - 10:39 PM Linux kernel client Feature #190: krbd: DISCARD support
- Definitely agree with Kyle. Due to this, and after finding out that an XFS fstrim within QEMU reports success but doe...
- 10:29 PM Revision e1277ba6 (ceph): OSD: move the peer_epoch and map sharing infrastructure into OSDService
- None of this code requires OSD-internal data or acquring locks from
anybody else.
Signed-off-by: Greg Farnum <greg@i... - 10:29 PM Revision b038f0c5 (ceph): OSD: rename share_map_incoming and share_map_outgoing
- share_map_incoming -> share_map
share_map_outgoing -> share_map_peer
Signed-off-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision 4bf20afc (ceph): SimpleMessenger: Don't grab the lock when sending messages if we don't ...
- We'd like it if sending a message didn't require any global locks, but the
submit_message() function conditionally ne... - 10:29 PM Revision 9028f95e (ceph): OSD: Juggle the locking when resurrecting a PG
- Don't hold the old PG's lock in _create_lock_pg. Instead, just copy the
necessary data bits into a holding location. ... - 10:29 PM Revision 62b2d43a (ceph): OSD: remove dead comment
- enqueue_op no longer requires holding the osd_lock.
Signed-off-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision fd2b57ea (ceph): OSD: enable ms_fast_dispatch
- We've been setting it up, now this patch actually adds a fast path for osd ops
which bypasses the osd_lock and should... - 10:29 PM Revision fccf1c70 (ceph): OSD: do not take the pre_publish_lock in connection utility functions
- They loop back around for local connections and deadlock, so we use the
map reservation mechanism instead.
TODO: actu... - 10:29 PM Revision 2ec92c76 (ceph): OSD: scan for dropped PGs in consume_map instead of advance_map
- We have to wait until after we know that nobody will be adding ops for
newly-dead PGs to the list. While we're moving... - 10:29 PM Revision 938feb49 (ceph): OSD: move the {boot,up,bind} epochs into OSDService
- Provide interfaces around setting and retrieving them, instead of accessing
them directly with a lock.
Signed-off-by... - 10:29 PM Revision 399e67f8 (ceph): OSD: pass a pointer to last_sent_epoch instead of the whole Session
- We don't use any other part of the Session, and this interface will
be easier to move out of the OSD class.
Signed-o... - 10:29 PM Revision 0fbaa160 (ceph): OSD: move should_share_map and share_map_incoming to OSDService
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision 0ffdeab9 (ceph): OSD: fix a few map sharing bugs
- 1) do not share OSD maps with peers that already have them
2) do not share maps with oneself
Signed-off-by: Greg Far... - 10:29 PM Revision 9fba69a1 (ceph): OSD: allow build_incremental_map_msg to fail on lookups
- Since we're now building incremental map messages out-of-band with doing
other map updates now, we need to tolerate l... - 10:29 PM Revision 6c98e36f (ceph): OSD: add an op threadpool GenContext workqueue
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision ebdc0970 (ceph): OSD: use the async workqueue to send OSDMap updates on dropped ops
- Check whether we actually want to send a map in-line, and if we do, create
a GenContext which does so and put that in... - 10:29 PM Revision 5268e51b (ceph): OSD: don't share_map_incoming() directly from handle_replica_op()
- Let the op_tp handle it, or our C_SendMap callback in the op_gen_wq.
Signed-off-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision b2187ac9 (ceph): OSD: use an OSDMapRef& and require the Session* in _share_map_incoming
- You can pass in a NULL Session*, but both callers do that; and using
an OSDMapRef& reduces shared_ptr copies.
Signed... - 10:29 PM Revision b53cec43 (ceph): OSD: add _should_share_map function
- Just copy _share_map_incoming and rip out all the parts that actually
update data structures.
Signed-off-by: Greg Fa... - 10:29 PM Revision 667769c6 (ceph): OSD: simplify _share_map_incoming based on _should_share_map()
- Also, remove the bool return code since nobody looks at it.
Signed-off-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision 1e3c4959 (ceph): OSD: add a Session::sent_epoch_lock
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision 276a4fe4 (ceph): OSD: change Session handling around _share_map_incoming
- Move responsibility for the reference up to _share_map_incoming's caller,
and start using the Session::sent_epoch_loc... - 10:29 PM Revision d78988bf (ceph): OSD: refactor handle_op error handling cases
- We move our map version-checking code earlier (to dispatch_op) and refactor
our other fail-to-dispatch cases. This is... - 10:29 PM Revision c97f9683 (ceph): OSD: share map updates in the op_tp threads instead of the main dispatc...
- Sharing maps can require disk accesses and things. We don't want to do that
in our fast path, so do it in OSD::dequeu... - 10:29 PM Revision 9d8c797e (ceph): OSD: Push responsibility for grabbing pg_map_lock up to callers of _rem...
- The atomicity requirements of other systems prevent us dropping the PG lock
inside that function, and the PG lock is ... - 10:29 PM Revision 767e94ac (ceph): OSD: shard heartbeat_lock
- heartbeat_need_update must be protected independently in order to avoid
a loop with the pg_map_lock and the PG::_lock... - 10:29 PM Revision a94a64d9 (ceph): OSD: protect access to boot_epoch, up_epoch, bind_epoch
- We need to access these members in some call chains via fast_dispatch,
where they're otherwise unprotected.
Signed-o... - 10:29 PM Revision 2f97f477 (ceph): OSD: protect state member with a Spinlock
- This member was previously protected by the osd_lock (although setting
SHUTDOWN was synchronized with the heartbeat l... - 10:29 PM Revision 812c6723 (ceph): OSD::_share_map_incoming: pass osdmap in explicitly
- We'll want to be able to use this method without the osd_lock. Note
that we can't do so yet -- we call send_increment... - 10:29 PM Revision b199194d (ceph): OSD::send_incremental_map: use service superblock so we can avoid locki...
- TODO: make it actually safe by dealing with build_incremental_map_msg()
Signed-off-by: Samuel Just <sam.just@inktank... - 10:29 PM Revision 9835866e (ceph): OSD: use safe params in map-sharing functions
- We were previously using unprotected access to OSD members.
Unfortunately, this does not make them completely safe: ... - 10:29 PM Revision 09bf5e80 (ceph): msgr: change the delay queue flushing semantics
- Since we're doing fast_dispatch out of the delay queue, we don't want to
flush while holding the pipe lock. Instead, ... - 10:29 PM Revision 5abbbfeb (ceph): OSDService: add osdmap reservation mechanism
- The goal here is to be able to get "reserved" refs
to next_map, and ensure that pgs won't see a newer
map until the r... - 10:29 PM Revision 475d8319 (ceph): OSD: add a RWLock pg_map_lock
- If we're going to dispatch ops without grabbing the osd lock, we need
something else to protect the pg map (and it'll... - 10:29 PM Revision 6d533492 (ceph): OSD: pass osdmap to handle_op and handle_replica_op
- We need a map to process them, and we don't want to
take the OSD lock to access one. (And we can't just
use the servi... - 10:29 PM Revision eb30f88c (ceph): OSD: add session waiting_for_map mechanisms
- This will replace the existing waiting_for_osdmap mechanism
with a per-session wait list.
Signed-off-by: Samuel Just... - 10:29 PM Revision 37553183 (ceph): OSD: remove wake_all_pg_waiters
- We shouldn't need this -- we check the pg waiters list on each
map.
Signed-off-by: Samuel Just <sam.just@inktank.com... - 10:29 PM Revision 00d36f6e (ceph): OSD: wake_pg_waiters atomically with pg_map update
- Also, call enqueue_op directly rather than going back
through the entire dispatch machinery.
Be sure to grab the pg l... - 10:29 PM Revision ec163579 (ceph): OSD: replace handle_pg_scan, handle_pg_backfill with handle_replica_op
- Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision e0ac34a0 (ceph): OSD: remove unused push_wq
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision dd3d023a (ceph): OSD: rename gen_wq, schedule_work, and PG_QueueAsync to include "recovery"
- These all hook into the recovery thread pool and need to make that obvious.
Signed-off-by: Greg Farnum <greg@inktank... - 10:29 PM Revision 1379c031 (ceph): OSD: remove never-activated while loop from send_incremental_map
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision a62db614 (ceph): DispatchQueue: factor out pre_dispatch and post_dispatch
- Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision 4e20ce19 (ceph): Messenger,DispatchQueue: add ms_fast_dispatch mechanism
- This adds a Dispatcher interface allowing the implementation
to accept ms_fast_dispatch calls for some messages witho... - 10:29 PM Revision 69fc6b2b (ceph): msgr: enable fast_dispatch on local connections
- We do two things:
1) Call ms_handle_fast_connect() when setting up the local connection, so
the Dispatcher can set up... - 10:29 PM Revision 816b10ed (ceph): RWLock: assert pthread function return values
- Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision 78f310d8 (ceph): PG: constify the init() function params
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:29 PM Revision 37fac29c (ceph): OSD::_share_map_incoming: line wrap debug output
- Formatting only.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:29 PM Revision 63cc1ec1 (ceph): OSD: add handle_osd_map debug output
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 10:26 PM Revision e68422b3 (ceph): Merge pull request #1769 from ceph/wip-doc-cache-tier
- doc: Support for cache tiering.
- 10:14 PM Revision ea34f48c (ceph): Merge remote-tracking branch 'origin/wip-doc-radosgw-80'
- Conflicts:
doc/radosgw/config.rst - 09:57 PM Revision d7e04cc8 (ceph): TrackedOp: remove the init_from_message function
- I'm not sure why we ever had this instead of just doing things in the
subclass constructor, and the semantics around ... - 09:57 PM Revision 6a559a5d (ceph): TrackedOp: introduce a _dump_op_descriptor function
- Use this instead of direct access to the Message underneath when dumping
the TrackedOp.
Signed-off-by: Greg Farnum <... - 09:57 PM Revision 2e674dea (ceph): TrackedOp: rename arrived_at to initiated_at, specify when constructed
- Instead of relying on the message's get_recv_stamp, take a timestamp
when the TrackedOp is constructed. Rename get_re... - 09:57 PM Revision 5a3efda7 (ceph): TrackedOp: introduce an _unregistered() function to let implementations...
- Right now, the OpRequest uses it to clean up Message payload data.
Signed-off-by: Greg Farnum <greg@inktank.com> - 09:57 PM Revision 95fc551a (ceph): TrackedOp: do not track a Message
- Give it to the OpRequest (currently, the only TrackedOp implementation).
Signed-off-by: Greg Farnum <greg@inktank.com> - 09:57 PM Revision e2b62bc3 (ceph): TrackedOp: do not require a Message when creating new Ops
- Further parameterize the template to allow passing in an arbitrary parameter,
and move all the Message-based event ma... - 09:53 PM Revision 3c9529ad (ceph): Merge pull request #1760 from ceph/wip-8283
- osd/ReplicatedPG: fix trim of in-flight hit_sets
Reviewed-by: Samuel Just <sam.just@inktank.com> - 09:48 PM Bug #7744: osd: assert(last_e.version.version < e.version.version)
- I'm getting this as well on a single OSD (osd.3) that mounts fine, but does not start. Same precise assertion failure...
- 07:50 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- I had another incident of "misdirected client" during recovery after replacement of another OSD...
All pools were ... - 07:35 PM Revision fc3318ed (ceph): doc: Fix hyperlink.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:34 PM Revision a7e72193 (ceph): doc: Index update and librados.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:33 PM Revision fcbc5fa6 (ceph): doc: Quotas for Admin Ops API.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:33 PM Revision e97b56eb (ceph): doc: New Admin Guide for Ceph Object Storage.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:18 PM Revision 99e67abc (ceph): rados/thrash: Add pool_snaps variants
- Fixes: #8284
Signed-off-by: David Zafman <david.zafman@inktank.com> - 06:10 PM Revision c86df77a (ceph): Restrict paramiko to old versions for now
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:04 PM rgw Bug #8289: rgw: memory not freed during in-progress read (dumpling)
- Could very well be related to #8269, in which we don't handle nicely the cancellation.
- 10:22 AM rgw Bug #8289: rgw: memory not freed during in-progress read (dumpling)
- I tend to think that what we see is that we read the data off RADOS much quicker than the client reading the data bac...
- 04:58 PM CephFS Bug #8290 (Resolved): ESTALE when grepping a snapshot via ceph-fuse
- 04:46 PM CephFS Bug #8290: ESTALE when grepping a snapshot via ceph-fuse
- 04:28 PM CephFS Bug #8290: ESTALE when grepping a snapshot via ceph-fuse
- /teuthology-2014-05-02_23:01:27-fs-master-testing-basic-plana/230880/
- 04:20 PM CephFS Bug #8290 (Resolved): ESTALE when grepping a snapshot via ceph-fuse
- ...
- 03:03 PM Revision 0c8a3e23 (ceph): Revert "Handle raw data I/O."
- This reverts commit 257e1459fa064701d789f0ad54384bb80b45e6d9.
- 03:03 PM Revision 02504c3f (ceph): Revert "Clean up remote.py and misc.py changes."
- This reverts commit 74eff43ee1a2b73159277370cfa9d194e42bf49c.
- 02:08 PM devops Feature #8120: RHEL7 GA kernel build
- Could do... or we could build out the new repo layout and use this as the first package in there?
- 10:17 AM devops Feature #8120 (In Progress): RHEL7 GA kernel build
- The kmod packages built for libceph/rbd for the rc-1 kernel (3.10.0-121). I seem to recall we were publishing these i...
- 11:32 AM Feature #7553 (In Progress): Remove classic scrub
- 11:24 AM Bug #8285 (Resolved): Thrasher throws exception when finishing up
- a723ddf5dbc8d78ca8c3cd843a1ffa2af4292cdd
- 07:54 AM Revision 75392810 (ceph): Fix installation into user home directory, broken by d3f0c0b
- Signed-off-by: Stefan Eilemann <Stefan.Eilemann@epfl.ch>
- 06:28 AM Revision 8e1e4ba3 (ceph): marginal/multimds: fuse_default_permissions = 0 for ceph-fuse
- This can reduce the test time becuase it avoids sending getattr
request whenever the kernel checks inode permission.
... - 04:20 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- kernel 3.14.2 looks good so far. i will keep an eye on it.
- 03:57 AM Bug #8232: Race condition during messenger rebind
- If the above analysis makes sense, here is the pull request: https://github.com/ceph/ceph/pull/1765
- 03:55 AM Bug #8232: Race condition during messenger rebind
- ...
05/04/2014
- 09:26 AM rgw Bug #8289: rgw: memory not freed during in-progress read (dumpling)
- Do we know if the object is a multi-part upload or not? And is it chunked or a single RADOS object? This might only b...
05/03/2014
- 11:07 PM rgw Bug #8289: rgw: memory not freed during in-progress read (dumpling)
- in this particular case the request is getting 503 after 30 seconds (the fastcgi timeout?).
- 10:53 PM rgw Bug #8289: rgw: memory not freed during in-progress read (dumpling)
- workaround is to limit the number of threads so that only a handful of such large reads can be in progress at once; t...
- 10:52 PM rgw Bug #8289 (Resolved): rgw: memory not freed during in-progress read (dumpling)
- observing memory growing linearly during a read on a large object. once the read completes, memory is freed.tcmalloc...
- 10:14 PM Revision 24c5ea8d (ceph): osd: check blacklisted clients in ReplicatedPG::do_op()
- OSD checks if client is blacklisted only when receiving OSD request.
It's possible that OSD request's sender get blac... - 10:13 PM Revision b4dfd3d5 (ceph): Merge pull request #1740 from ceph/wip-8155
- mon: OSDMonitor: disallow nonsensical cache-mode transitions
Reviewed-by: Sage Weil <sage@inktank.com> - 10:11 PM Revision 491cfdb3 (ceph): Merge pull request #1763 from ceph/wip-blacklist
- Wip blacklist
Backport: firefly
Reviewed-by: Sage Weil <sage@inktank.com> - 09:36 PM Revision f92677c5 (ceph): osd: check blacklisted clients in ReplicatedPG::do_op()
- OSD checks if client is blacklisted only when receiving OSD request.
It's possible that OSD request's sender get blac... - 08:50 PM rgw Bug #8288 (Duplicate): rgw tests cause osd wrongly marked down
- See this mostly with valgrind. See
teuthology-2014-05-02_23:00:58-rgw-master-testing-basic-plana/230836
teutholo... - 05:13 PM Bug #7996: 0.78: OSD is not suspend-friendly (unresponsive cluster on OSD crash)
- Sage Weil wrote:
> I suspect that the mon on that machine is the key factor at play here.
No MON doesn't matter. ... - 04:41 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- After recovery to "active+clean" cluster became operational again. I/O errors on RBD devices are gone for now.
Could... - 03:12 PM Feature #8155 (Pending Backport): Disallow changing cache_mode in nonsensical ways
- 02:59 PM Revision c64b67b5 (ceph): ceph-object-corpus: rebase onto firefly corpus
- Signed-off-by: Sage Weil <sage@inktank.com>
- 02:57 PM Revision 077e6f86 (ceph): ceph-object-corpus: v0.80-rc1-35-g4812150
- Signed-off-by: Sage Weil <sage@inktank.com>
- 02:41 PM Bug #8287 (Rejected): Package rebuild needed for trusty for dumpling on trysty
- These are old point releases of Dumpling - we don't need packages for them on Trusty.
- 08:49 AM Bug #8287 (Rejected): Package rebuild needed for trusty for dumpling on trysty
- Logs are in http://qa-proxy.ceph.com/teuthology/ubuntu-2014-05-02_22:15:51-upgrade:dumpling-dumpling-testing-basic-pl...
- 02:41 PM Revision af8a5298 (ceph): Merge pull request #1762 from yuyuyu101/wip-8282
- Fix clone problem
Backport: firefly
Reviewed-by: Sage Weil <sage@inktank.com> - 02:24 PM Revision 794c9465 (ceph): ceph_manager: fix float stringification
- Signed-off-by: Sage Weil <sage@inktank.com>
- 01:49 PM Revision ca116a3e (ceph): Merge pull request #1752 from ceph/wip-da-SCA-fixes-20140501
- Various fixes from SCA
Reviewed-by: Sage Weil <sage@inktank.com> - 01:42 PM Revision 00868514 (ceph): Merge pull request #1755 from eile/master
- Fix out of source builds
Reviewed-by: Sage Weil <sage@inktank.com> - 10:02 AM Revision 8bd4e582 (ceph): Fix out of source builds
- Signed-off-by: Stefan Eilemann <Stefan.Eilemann@epfl.ch>
- 09:34 AM Bug #8113 (Pending Backport): agent_work can be continuously rescheduled during recovery while mo...
- 09:34 AM Bug #8175 (Pending Backport): Some values of target_max_objects for tiering will crash OSDs
- 07:40 AM Bug #8282 (Pending Backport): ceph-objectstore-test segv
- 04:55 AM Revision 3aee1e0f (ceph): Fix clone problem
- When clone happened, the origin header also will be updated in GenericObjectMap,
so the new header wraper(StripObject... - 04:18 AM Revision a723ddf5 (ceph): ceph_manager: fix typo
- From ce7fa1839f4b3e3db675b2d68a2bb57849f58c1e. Tested this time.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewe... - 02:42 AM Revision fd970bbc (ceph): mon: OSDMonitor: disallow nonsensical cache-mode transitions
- Fixes: 8155
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com> - 02:26 AM Revision 3e314584 (ceph): Merge pull request #1735 from ceph/wip-8113
- Reviewed-by: Samuel Just <sam.just@inktank.com>
- 12:06 AM Revision 7a46469f (ceph): Merge pull request #248 from ceph/wip-thrash-osd-weights
- ceph_manager: randomly reweight in osds
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 12:06 AM Revision 94e6c085 (ceph): rgw: don't allow multiple writers to same multiobject part
- Fixes: #8269
A client might need to retry a multipart part write. The original thread
might race with the new one, t... - 12:04 AM Revision d2d7b941 (ceph): cache-snaps.yaml: set target_max_objects to test snap flush/evict
- Signed-off-by: Samuel Just <sam.just@inktank.com>
05/02/2014
- 11:52 PM Revision 4aa93dd1 (ceph): Merge pull request #1698 from ceph/wip-snapmapper-debug
- osd/SnapMapper: debug
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:48 PM Revision 2700ebf8 (ceph): Merge pull request #1694 from ceph/wip-throttle-snap-master
- osd: throttle snap trimmming with simple delay
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:43 PM Revision b45adc98 (ceph): Merge pull request #1759 from dachary/wip-mailmap
- DNM: mailmap updates
Reviewed-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 11:41 PM Revision 72fdd557 (ceph): osd/ReplicatedPG: fix trim of in-flight hit_sets
- We normally need to stat the hit_set to know how many bytes to adjust the
stats by. If the hit_set was just written,... - 11:36 PM Revision 84728058 (ceph): Revert "ReplicatedPG: block scrub on blocked object contexts"
- This reverts commit e66f2e36c06ca00c1147f922d3513f56b122a5c0.
Reviewed-by: Sage Weil <sage@inktank.com> - 11:35 PM Revision b7d31e5f (ceph): osd, common: If agent_work() finds no objs to work on delay 5 (default)...
- Add config osd_agent_delay_time of 5 seconds
Honor delay by ignoring agent_choose_mode() calls
Add tier_delay to logg... - 11:35 PM Revision f47f8679 (ceph): osd: Prevent divide by zero in agent_choose_mode()
- Fixes: #8175
Backport: firefly
Signed-off-by: David Zafman <david.zafman@inktank.com>
Signed-off-by: Sage Weil <sage... - 11:35 PM Revision fe0031d9 (ceph): rados.cc: fix typo in help output
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision 70a4a73d (ceph): SimpleLock.h: remove twice included osd_types.h
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision be5a99d6 (ceph): SimpleLock.h: remove unused private function clear_more()
- Remove unused private function clear_more(), it's replaced by
try_clear_more().
Signed-off-by: Danny Al-Gaaf <danny.... - 11:35 PM Revision 1e7eb1a1 (ceph): osd_types.h: pass eversion_t by reference to operator<<
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision 8fad144c (ceph): PGBackend::be_compare_scrubmaps(): pass pgid by reference
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision 296b8ed0 (ceph): PG::read_info(): pass 'const coll_t coll' by reference
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision 8bf039d0 (ceph): Dumper::dump_entries(): reduce scope of 'got_data'
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision b05e04ec (ceph): Dumper::dump_entries(): remove not needed variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:35 PM Revision 4fe31c1b (ceph): linux_fiemap.h: remove twice included int_types.h
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:34 PM Revision 9a716d88 (ceph): rgw_bucket.cc: return error if update_containers_stats() fails
- In case need_stats is set on rgw_read_user_buckets() and the
update_containers_stats() call fails with !-ENOENT, not ... - 11:32 PM Revision ce7fa183 (ceph): ceph_manager: randomly reweight in osds
- Signed-off-by: Sage Weil <sage@inktank.com>
- 10:55 PM Revision 3f1d7f5e (ceph): mon/PGMonitor: set tid on no-op PGStatsAck
- The OSD needs to know the tid. Both generally, and specifically because
the flush_pg_stats may be blocking on it.
F... - 10:54 PM Revision 4e0eaa95 (ceph): mon/OSDMonitor: share latest map with osd on dup boot message
- If we get a dup boot message, share the newer maps with the osd so that
they know they are living in the past.
Fixes... - 10:54 PM Revision 89044a6d (ceph): mon/PGMonitor: set tid on no-op PGStatsAck
- The OSD needs to know the tid. Both generally, and specifically because
the flush_pg_stats may be blocking on it.
F... - 10:43 PM Revision 5a6ae2a9 (ceph): mon/PGMonitor: set tid on no-op PGStatsAck
- The OSD needs to know the tid. Both generally, and specifically because
the flush_pg_stats may be blocking on it.
F... - 10:43 PM Revision 2e6b2486 (ceph): mon/OSDMonitor: share latest map with osd on dup boot message
- If we get a dup boot message, share the newer maps with the osd so that
they know they are living in the past.
Fixes... - 10:42 PM Revision 77a6f0ae (ceph): mon/MonClient: remove stray _finish_hunting() calls
- Callig _finish_hunting() clears out the bool hunting flag, which means we
don't retry by connection to another mon pe... - 10:42 PM Revision d0245947 (ceph): mailmap: Florent Bautista affiliation
- and name normalization
Signed-off-by: Loic Dachary <loic@dachary.org> - 10:35 PM Revision 61a2f064 (ceph): mailmap: Warren Usui name normalization
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 10:35 PM Revision 7b192f7d (ceph): mailmap: Guang Yang name normalization
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 09:22 PM Revision 331869a0 (ceph): Merge pull request #1754 from nereocystis/hardware-to-glossary
- doc: Include links from hardware-recommendations to glossary
- 07:26 PM Bug #8175 (Resolved): Some values of target_max_objects for tiering will crash OSDs
- f47f867952e6b2a16a296c82bb9b585b21cde6c8
- 07:26 PM Bug #8113 (Resolved): agent_work can be continuously rescheduled during recovery while most objec...
- b7d31e5f5952c631dd4172bcb825e77a13fc60bc
- 07:24 PM Bug #8285: Thrasher throws exception when finishing up
My test passed after I used teuthology from the commit before "randomly reweight in osds"- 06:47 PM Bug #8285 (Resolved): Thrasher throws exception when finishing up
2014-05-02 18:36:50,164.164 INFO:teuthology.task.thrashosds:joining thrashosds
2014-05-02 18:36:50,165.165 ERROR:t...- 07:15 PM Fix #8286 (Resolved): 0.80~rc1: `crushtool` crash
- ...
- 06:54 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- I re-built OSD.4 from scratch and re-added it.
While data is flowing back to OSD.4 I run `badblocks` and found read ... - 06:29 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- That was the first thin I did (setting weight of OSD.7 back to 1) but it didn't help.
All OSDs have reweight value 1... - 05:57 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Can you try setting hte osd weights all to 1.0? (ceph osd reweight N 1) My best guess is that the bug is in that co...
- 05:43 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- A little update: `badblocks` show multiple "bad blocks" or RBD devices.
Ceph logs the following on every failed read... - 03:58 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Forgot to mention that shortly before error appear I did...
- 03:51 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Damn it just happened again, roughly at the same time when just one OSD went down on healthy cluster.
Linux-3.14.2... - 05:48 PM Support #7609: http://tracker.ceph.com/account/register returns 500 Internal error
- Saw that a couple of month ago and forgot to create a tracker.
It's a lang issue. I just tried to register from a ... - 03:33 PM Support #7609: http://tracker.ceph.com/account/register returns 500 Internal error
- Just signed out and hit Register, and got a page as normal. Don't know what might be causing the issue. I see a 500...
- 05:03 AM Support #7609: http://tracker.ceph.com/account/register returns 500 Internal error
- +1 today. User "anton" on IRC has it too.
- 05:26 PM Feature #8284 (Resolved): Add --pool_snaps rados tests to teuthology
After adding --pool-snaps option to ceph_test_rados in 3ce407800d4187645b1d99ec533df69d6fc0a1f2, we should add pool...- 04:48 PM Revision 5844c23e (ceph): Bump paramiko to 1.12.0
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:40 PM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- I encountered this issue before. please try 3.14 kernel, I think I haven't encountered in 3.14 kernel.
- 11:52 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- There are some outstanding osd requests, as i can see. Restarting the corresponding OSDs unfroze the hanging tasks.
... - 11:36 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- is there hang OSD request ? (in /sys/kernel/debug/ceph/*/mdsc)
- 08:11 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- On deadlocked node, i have only the cephfs kernel client running and no OSD. Memory is plenty available and I did not...
- 07:54 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- There is an interesting article which details why this happens for nfs, and what they're doing to fix the problem the...
- 07:09 AM CephFS Bug #7474: Kernel oops with cephfs [ceph_write_begin -> *x8 -> wait_on_page_read]
- I have encountered the same issue on v3.12.17. Is there already a patch available for this one?
[Fri May 2 15:03:... - 04:14 PM Bug #8283 (Fix Under Review): osd: hit_set_trim cannot stat recently written hitsets
- 04:08 PM Bug #8283 (Resolved): osd: hit_set_trim cannot stat recently written hitsets
- ...
- 04:11 PM Feature #7792: leveldb 1.12.0 for rhel
- 1.12 is gone from the dumpling repo now
- 12:05 PM Feature #7792: leveldb 1.12.0 for rhel
- I'm concurring. C6 system, Ceph 0.72.2, after reboot oh the host I experienced crashing mon and hanging OSDs. Downgra...
- 09:44 AM Feature #7792: leveldb 1.12.0 for rhel
- Did you pull it? I still see 1.12 here: http://ceph.com/rpm-dumpling/el6/x86_64/
- 07:53 AM Feature #7792: leveldb 1.12.0 for rhel
- Have we made sure we are applying the patch mentioned here?
http://tracker.ceph.com/issues/6022#note-9 - 07:41 AM Feature #7792: leveldb 1.12.0 for rhel
- Re-opening because the new package is causing severe issues with the monitors.
- 07:36 AM Feature #7792: leveldb 1.12.0 for rhel
- Same here. Mon doesn't restart after upgrading leveledb.
- 07:19 AM Feature #7792: leveldb 1.12.0 for rhel
- pulling leveldb 1.12.0 for rhel made it impossible to use/deploy ceph on rhel6/el6 systems.
As already mentioned i... - 04:08 PM Revision 46628907 (ceph): sample.ceph.conf update:
- * corrected URLs.
* added [client] section.
* more options and descriptions.
* filestore settings were moved under... - 04:06 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- this triggered again on commit:c6ada53a146f3196e11f545cfc968fc21657aec6
0> 2014-05-02 11:26:12.053553 7fd1ee1... - 04:04 PM Bug #8282 (Resolved): ceph-objectstore-test segv
- ...
- 03:54 PM Bug #8280 (Resolved): mon: stats reply doesn't include tid if no-op
- 03:08 PM Bug #8280 (Resolved): mon: stats reply doesn't include tid if no-op
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-05-02_02:30:10-rados-master-testing-basic-plana/229240...
- 03:54 PM Bug #8279 (Resolved): mon: osd failed to boot after being thrashed down
- 02:58 PM Bug #8279 (Resolved): mon: osd failed to boot after being thrashed down
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-05-02_02:30:10-rados-master-testing-basic-plana/229236...
- 03:47 PM Documentation #8281 (Resolved): Documentation: Detailed explanation of ceph df output is non-exis...
- Right now ceph df output for pool usage is vague and there is no documentation explaining what the numbers means.
... - 03:42 PM Bug #8278 (Pending Backport): monclient: failure to retry after ill-timed connection reset during...
- 02:47 PM Bug #8278 (Fix Under Review): monclient: failure to retry after ill-timed connection reset during...
- 02:40 PM Bug #8278 (Resolved): monclient: failure to retry after ill-timed connection reset during auth
- ...
- 03:17 PM Bug #7986 (Need More Info): 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/20...
The obvious areas to look at is client delete and hit_set_persist() archive removal. The code in hit_set_persist()...- 02:19 PM Feature #8276 (Resolved): ceph-filestore-dump import-rados -p <pool> <archive>
- I'd also like to propose a mode where ceph-filestore-dump reinserts the contents of an exported pg via rados, somethi...
- 10:05 AM Linux kernel client Bug #8275 (New): krbd: 'rbd unmap' gets stuck
- This could be a libceph issue, but both Hannes and myself saw it on 'rbd unmap'.
From: Hannes Landeholm <hannes@ju... - 07:10 AM Bug #8272: leveldb 1.12.0 broken on el6
- duplicate to 8273
- 07:08 AM Bug #8272 (Duplicate): leveldb 1.12.0 broken on el6
- Hi,
since the resolving of #7798, issue #6022 needs to be reopened.
leveldb-1.12 breaks ceph-deploy on centos6/... - 07:09 AM Bug #8273 (Duplicate): leveldb 1.12.0 broken on el6
- Hi,
since the resolving of #7792, issue #6022 needs to be reopened.
leveldb-1.12 breaks ceph-deploy on centos6/... - 05:02 AM Bug #8241: XfsFileStoreBackend tries to set extsize but may get EINVAL
- Today pinguini (with a different nick "anton" this time) gave more information on IRC regarding this issue:...
- 04:34 AM CephFS Bug #8255: mds: directory with missing object cannot be removed
- Thanks, I might try that or make a new file system from scratch.
There are more than one issue mentioned in this t... - 04:04 AM Revision adf2ec43 (ceph): Merge pull request #30 from ceph/wip-8263
- Wip 8263
- 03:55 AM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- It's getting interesting: every time there are less errors found so apparently each iteration fixes some.
Also OSD s... - 02:01 AM Bug #8270: 0.80~rc1: OSD crash during replication after repair
- ...
- 12:42 AM Bug #8270 (Can't reproduce): 0.80~rc1: OSD crash during replication after repair
- ...
- 02:21 AM Revision 9cf470ca (ceph): osd/ReplicatedPG: agent_work() fix next if finished early due to start_max
- Backport: firefly
Signed-off-by: David Zafman <david.zafman@inktank.com> - 12:59 AM rgw Documentation #8271 (Closed): Document support for S3 multi-object delete
- The S3 API documentation contains no mention of rgw's support for multi-object delete.
- 12:31 AM Revision 9f1a9168 (ceph): osd/SnapMapper: pass snaps set by const ref
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:29 AM Revision 6105c355 (ceph): osd/SnapMapper: debug
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:25 AM Revision cf25bdf6 (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 ... - 12:25 AM Revision 49a3b222 (ceph): osd: ignore MarkMeDown message if we aren't in PREPARING_TO_STOP state
- If we aren't waiting for this, ignore it.
Signed-off-by: Sage Weil <sage@inktank.com> - 12:25 AM Revision 58ace1aa (ceph): osd: fix 'ack' to be 'request_ack' in MOSDMarkMeDown
- This field was passed along but always set to false. It did not seem to
indicate whether this was/wasn't an ack (the... - 12:25 AM Revision f0658090 (ceph): mon/OSDMonitor: do not reply to MOSDMarkMeDown if ack is not requested
- If a reply isn't requested, do not bother to send one. Note that old
clients did not request an ack, but we will inf... - 12:24 AM Revision 81e4c477 (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 ... - 12:08 AM Revision 48121505 (ceph): Merge pull request #1751 from ceph/wip-mds-shutdown
- mds: remove mdsdir in the final step of shutdown MDS
Reviewed-by: Sage Weil <sage@inktank.com> - 12:04 AM Revision c879e895 (ceph): doc: Include links from hardware-recommendations to glossary
- Included :term: in parts of hardware-recommendations so that glossary
links appear.
Signed-off-by: Kevin Dalley <kevi...
05/01/2014
- 11:06 PM Revision c6ada53a (ceph): Merge pull request #1749 from hufman/fix-typo-releasenotes-pyramind
- Fixes a very minor typo in the release notes
Reviewed-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 10:07 PM Revision cc043225 (ceph): mds: note MDiscoverReply encoding change in corpus
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:43 PM Feature #7784 (Resolved): mon osd down out interval = 0 should prevent ceph health from reporting ok
- 09:40 PM Revision e597068b (ceph): mds: remove mdsdir in the final step of shutdown MDS
- Otherwise we may get bad subtree map if we restart the MDS before
the shutdown process finishes.
Signed-off-by: Yan,... - 09:39 PM Revision 18334ea3 (ceph): rados/thrash: Fix workload of cache-agent-big
- Create a log of objects and operate on some of them
(Initial object creation counts against total operations specifie... - 09:36 PM Revision 27b276e6 (ceph): rgw: test with ec + cache pool
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:36 PM Revision c5da7b21 (ceph): rgw: option to create a cache pool
- 64mb for now!
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 09:03 PM Bug #8263 (Resolved): cache-agent-big.yaml ceph_test_rados params cause only writes
- 07:21 PM Bug #8263 (Fix Under Review): cache-agent-big.yaml ceph_test_rados params cause only writes
- 09:36 AM Bug #8263: cache-agent-big.yaml ceph_test_rados params cause only writes
- see how many objects it did create in the run time (10 minutes?), and maybe create a third of that and run it for lon...
- 08:52 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Understood, I'll report debug logs if it happen again. I only have simplest replicated 'rbd' pool. Read errors occurr...
- 02:31 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- If you can reproduce, it'd be useful to get osd logs with debug osd = 20, debug ms = 1 when the I/O errors occur. Did...
- 08:46 PM Revision 1f4a3e1f (ceph): mds: bump protocol
- In commit f689e5f049736bb0a0fa437e05936f6c1b9c1bb6 we change the encoding
and semantics for MDiscoverReply.
Signed-o... - 05:52 PM Revision b0803559 (ceph): Fix syntax of erasure coded pool creation
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 05:46 PM Revision d993b9ca (ceph): Merge pull request #1738 from ceph/wip-8147
- osd: automatically scrub PGs with invalid stats
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: David Z... - 05:17 PM Revision f74eea7b (ceph): Merge pull request #247 from ceph/requests-sessions
- Use a requests.Session object for retries instead of safe_while
- 04:59 PM Revision 1ac05fd1 (ceph): doc/release-notes: changelog link
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:48 PM Revision d1b93530 (ceph): Add branch name to job config
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:37 PM rgw Bug #8269 (Resolved): rgw: corrupted multipart object
- scenario:
- client uploads part
- upload stalls, client starts retransmission of that part
- rgw identifies f... - 04:25 PM Revision ab9645f9 (ceph): Add suite name to job config
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:25 PM Revision ba66c6ba (ceph): Add /build and /*.yaml to gitignore
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 03:36 PM Revision ffef20fe (ceph): doc/release-notes: final v0.67.8 notes and changelog
- Signed-off-by: Sage Weil <sage@inktank.com>
- 01:39 PM Feature #8147 (Pending Backport): osd: make split automatically trigger scrub
- 12:17 PM Bug #8099 (Duplicate): LibRBD.DiffIterateStress failure - extra extent in diff
- Looks like it hasn't occurred since then.
- 09:49 AM Bug #8099: LibRBD.DiffIterateStress failure - extra extent in diff
- 8091 was fixed; is this still an issue?
- 11:55 AM devops Feature #8039: move to libgoogle-perftools4
- 11:53 AM devops Feature #7947: Create separate ceph and ceph-common packages for EL6 and EL7 builds
- 11:44 AM devops Cleanup #7675 (Resolved): clean up Gary Lowell's WIP branches
- 11:42 AM devops Tasks #7230 (Resolved): Rebuild sync-agent packages for dumpling repo
- 11:18 AM Revision ffc58b4e (ceph): 0.67.8
- 10:46 AM Bug #8267 (Can't reproduce): Bad ceph command syntax sends to mon anyway
ceph osd pool create base 4 erasure teuthologyprofile
2014-04-18T16:57:19.734 DEBUG:teuthology.orchestra.run:Run...- 10:18 AM Bug #8114 (In Progress): "osd/RadosModel.h: 1055: FAILED assert" in upgrade:dumpling-x:stress-spl...
- 10:14 AM Bug #8011 (Resolved): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= s...
- 10:14 AM Bug #7995 (Duplicate): osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
- this seems to be the same as #7891
- 09:54 AM devops Bug #6966: ceph-disk: prepare --dmcrypt failing
- This is still failing in Firefly 0.80-rc1...
- 06:07 AM Revision 387b2974 (ceph): Merge pull request #1750 from nereocystis/doc-link-to-involved
- doc: documenting links to get-involved
- 05:25 AM Revision 0454962e (ceph): Fixes a very minor typo in the release notes
- pyramind -> pyramid
- 04:31 AM Revision 78b3c93d (ceph): doc: documenting links to get-involved
- Create a link from documenting-ceph so that it is easy to find the
github repository used for ceph.
Signed-off-by: K... - 01:46 AM Revision 79bf1a69 (ceph): Merge pull request #28 from ceph/wip_add_0688_release
- added new correctd tag 67.8
- 01:41 AM Revision 013a3b6c (ceph): added new correctd tag 67.8
04/30/2014
- 11:48 PM Revision 2e0befda (ceph): Merge pull request #27 from ceph/wip_add_0688_release
- added latest dumpling tag v0.68.8
- 11:35 PM Revision 4322ade6 (ceph): added latest dumpling tag v0.68.8
- 10:15 PM Revision 4b16b70c (ceph): Merge pull request #1743 from ceph/wip-mon-backports.dumpling
- mon: OSDMonitor: HEALTH_WARN on 'mon osd down out interval == 0'
Reviewed-by: Sage Weil <sage@inktank.com> - 10:13 PM Revision 0f3235d4 (ceph): ReplicatedPG: block scrub on blocked object contexts
- Fixes: #8011
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
(cherry pick... - 10:11 PM Revision e66f2e36 (ceph): ReplicatedPG: block scrub on blocked object contexts
- Fixes: #8011
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com> - 10:10 PM Revision 4bac8c7a (ceph): rados.h,ReplicatedPG: add CEPH_OSD_FLAG_ENFORCE_SNAPC and use on flush
- We need to ensure that even with pool snaps, we use the snapc provided in order
to ensure that the clones are written... - 10:09 PM Revision 3b6d262f (ceph): Merge pull request #1745 from ceph/wip-7941
- rados.h,ReplicatedPG: add CEPH_OSD_FLAG_ENFORCE_SNAPC and use on flush
Reviewed-by: Sage Weil <sage@inktank.com> - 10:08 PM Revision d9106ce5 (ceph): ECBackend::continue_recovery_op: handle a source shard going down
- get_min_avail_to_read_shards might return an error if there are
no longer enough sources to reconstruct the missing s... - 10:07 PM Revision ed464336 (ceph): Merge pull request #1744 from ceph/wip-8161
- ECBackend::continue_recovery_op: handle a source shard going down
Reviewed-by: Sage Weil <sage@inktank.com> - 10:06 PM Revision 87195d5f (ceph): ReplicatedPG: we can get EAGAIN on missing clone flush
- Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 060105c313c5b4a777c55f17115eeb95ebb17117) - 10:05 PM Revision d700d99f (ceph): ReplicatedPG: do not preserve op context during flush
- Any information stashed in the OpContext may be obsolete by the time we
actually mark the object clean. Instead, let... - 10:03 PM Revision 348e8c17 (ceph): Merge pull request #1746 from ceph/wip-8086
- Wip 8086
Reviewed-by: Sage Weil <sage@inktank.com> - 09:37 PM Revision aafed10c (ceph): rgw_common.cc: reduce scope of 'fpos' variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:34 PM Revision 49b810ff (ceph): rgw_admin.cc: remove unused string variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:31 PM Revision b9e612ca (ceph): PGBackend.cc: remove unused to_remove variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:29 PM Revision 5fec86ee (ceph): KeyValueStore.cc: remove unused variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:28 PM Revision 7928fe75 (ceph): rgw_op.cc: remove unused map variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:26 PM Revision 074161fc (ceph): rgw_main.cc: remove unused variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 09:21 PM Revision ae9a7d06 (ceph): rgw_main.cc: use static_cast instead of c-style
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:54 PM Feature #8265 (Resolved): config: make int parser accept K, M, G, T, P suffix and scale value acc...
- 08:51 PM Revision b99f6365 (ceph): buffer.cc: catch exception by reference
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision 5da8e0e8 (ceph): rados.cc: reduce scope of variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision 3ebbd998 (ceph): CDentry.cc: fix bool comparison using relational operator
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision e624085f (ceph): ObjectStore.h: pass const string parameter by reference
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision a84fed61 (ceph): crush/mapper.c: fix printf format for unsigned variable
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision e4b31094 (ceph): KeyValueStore: rename s/logger/perf_logger/
- [src/os/KeyValueStore.h:368] -> [src/os/ObjectStore.h:100]: (warning) The class
'KeyValueStore' defines member varia... - 08:51 PM Revision 55624288 (ceph): OSDMonitor.cc: prefer prefix ++operator for non-trivial iterator
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision ef0de7ac (ceph): OSDMap.cc: prefer prefix ++operator for non-trivial iterator
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision 46442ea9 (ceph): hitset.cc: fix format string to unsigned int
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 08:51 PM Revision 022e705f (ceph): SimpleLock.h: fix bool comparison using relational operator
- [src/mds/SimpleLock.h:287]: (warning) Comparison of a boolean value
using relational operator (<, >, <= or >=).
Sig... - 07:30 PM Revision a9d7aa35 (ceph): Refactor teuthology.beanstalk
- This architecture will make it easier to add new functionality.
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> - 07:30 PM Revision 041666b0 (ceph): Add --runs, to print only run names
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:30 PM Revision 3fa6271f (ceph): Calculate a timeout to use based on queue size
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 06:12 PM Bug #8264 (Rejected): "unknown op copy_from" error in ubuntu-2014-04-30_14:23:02-rados-dumpling-t...
- If you are seeing the copy-from failure, it is becuase you are using the master/firefly branch of teuthology or ceph-...
- 06:05 PM Bug #8264 (Rejected): "unknown op copy_from" error in ubuntu-2014-04-30_14:23:02-rados-dumpling-t...
- Logs are in http://qa-proxy.ceph.com/teuthology/ubuntu-2014-04-30_14:23:02-rados-dumpling-testing-basic-plana/224887/...
- 05:52 PM Revision 82a3668e (ceph): Merge pull request #1629 from ceph/wip-die-mkcephfs
- remove mkcephfs (merge post-firefly?)
Reviewed-by: Alfredo Deza <alfredo.deza@inktank.com> - 05:45 PM CephFS Bug #8255: mds: directory with missing object cannot be removed
- get inode number of 'epiphany' directory, then modify Server::_dir_is_nonempty_unlocked() and Server::_dir_is_nonempt...
- 05:12 PM CephFS Bug #8255: mds: directory with missing object cannot be removed
- FS was created on 0.72.2 then upgraded to 0.78, 0.79 following by 0.80~rc1.
Somehow journal was corrupted during clu... - 07:12 AM CephFS Bug #8255: mds: directory with missing object cannot be removed
- besides, I'm curious when was the fs created (which version)
- 05:36 PM Revision 8979eb39 (ceph): Merge pull request #1741 from ceph/wip-early-reply
- Wip early reply
- 05:33 PM Revision 21bbdf5d (ceph): mds: avoid adding replicas of target dentry to rename witnesses
- When the rename target dentry is NULL, we can use MDentryLink messages
instead of slave requests to update its replic... - 05:32 PM Revision 3a7d6684 (ceph): mds: allow early reply when request's witness list is empty
- Early reply should be Ok when there were slave requests, but all
of them were for acquiring locks.
Signed-off-by: Ya... - 05:11 PM Revision b5d1dd8a (ceph): Merge pull request #1121 from ceph/wip-no-anchor
- mds: remove anchor table (merge post-firefly only)
Reviewed-by: Sage Weil <sage@inktank.com> - 05:10 PM Revision 0fa969b5 (ceph): Merge pull request #1670 from yuyuyu101/wip-test-clone
- Add clone test on store_test
Reviewed-by: Sage Weil <sage@inktank.com> - 04:42 PM Bug #8263 (Resolved): cache-agent-big.yaml ceph_test_rados params cause only writes
All 3 teuthology.log files I saw showed that the ceph_test_rados didn't do any delete, copy_from nor read even thou...- 04:37 PM Bug #7996: 0.78: OSD is not suspend-friendly (unresponsive cluster on OSD crash)
- I reproduced this problem on 0.80~rc1. On healthy cluster suspend of the machine with one OSD have very ill effect on...
- 04:13 PM Revision fb0944e2 (ceph): mon: OSDMonitor: HEALTH_WARN on 'mon osd down out interval == 0'
- A 'status' or 'health' request will return a HEALTH_WARN whenever the
monitor handling the request has the option set... - 03:08 PM Bug #7941 (Resolved): caching needs to be able to enforce snap context on flush even with pool snaps
- 03:06 PM Bug #8161 (Resolved): osd/ECBackend.cc: 475: FAILED assert(r == 0)
- 03:03 PM Bug #8068 (Resolved): try_flush_mark_clean can end up using a snapset from the past corrupting th...
- 12:07 PM Feature #7792 (Resolved): leveldb 1.12.0 for rhel
- 1.12 ended up just being pulled for fedora but there for centos/rhel. Added it to the various ceph repos and ceph-ext...
- 11:47 AM CephFS Documentation #8258: 0.80~rc1: outdated MDS man page
- Can you do this since you're working with that functionality right now? :)
- 10:31 AM Revision 41d93aab (ceph): mds: include authority of the source inode in rename witnesses
- rename updates source inode's ctime
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 06:13 AM Bug #8259: add a target to ensure new dependencies are installed before builds
- make install-deps perhaps? This would be even more useful if done in
a way that will allow us to centralize our lis... - 05:29 AM Bug #8259 (Resolved): add a target to ensure new dependencies are installed before builds
- While trying to build Ceph from master today, Jenkins failed:...
- 04:31 AM CephFS Bug #8257: 0.80~rc1: MDS segmentation fault
- This is the return value of Dumper::init not being checked in ceph_mds.cc.
This case is fixed on the wip-journal-t... - 01:01 AM Bug #8229: 0.80~rc1: OSD crash (domino effect)
- Interesting experiment: on degraded cluster I take down one OSD and bring it up on another host.
Three other OSDs cr... - 12:51 AM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- Perhaps it's too late to provide maps -- I had RBD read errors on three clients while cluster was severely degraded (...
04/29/2014
- 11:11 PM CephFS Documentation #8258 (Resolved): 0.80~rc1: outdated MDS man page
- ...
- 11:07 PM CephFS Bug #8257 (Resolved): 0.80~rc1: MDS segmentation fault
- The following command reproduce MDS segmentation fault:...
- 11:06 PM Bug #8256 (Won't Fix): unhelpful ceph cli command --help
The usage on the left says <poolname> everywhere not making it clear where to put the <tierpool> name from the desc...- 10:55 PM CephFS Bug #8255 (Need More Info): mds: directory with missing object cannot be removed
- need more log to diagnose
truncate the mds log
execute "rm -rv /mnt/ceph/home/user/.config/epiphany"
update the ... - 10:13 PM CephFS Bug #8255 (Resolved): mds: directory with missing object cannot be removed
- MDS write the following line to it log over 14000 times per minute:...
- 09:41 PM Revision adc51e1c (ceph): Drop usage of safe_while
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 09:17 PM Revision ec72137e (ceph): Add a requests.Session object to ResultsReporter
- By default it is set up to retry requests 10 times
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> - 09:17 PM Revision ea9c034f (ceph): Use the new ResultsReporter.session object
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 09:17 PM Revision 61e469b6 (ceph): Remove unused timeout arg to ResultsReporter init
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 09:16 PM Revision 3f93d168 (ceph): Merge pull request #243 from ceph/wip-8116-wusui
- Wip 8116 wusui
- 09:14 PM Revision 257e1459 (ceph): Handle raw data I/O.
- Paramiko 1.13.0 checks data in the Channel and fails if
invalid UTF-8 characters are sent. The teuthology/misc.py
fu... - 09:14 PM Revision 74eff43e (ceph): Clean up remote.py and misc.py changes.
- Fixed method names to be non-redundant (remote_mktemp in remote is
now just mktemp, for example), and made some param... - 08:18 PM Revision 5339c1f2 (ceph): Changes so these are not installed and still removed
- 08:18 PM Revision 3faeb08d (ceph): When deleting all of a run's jobs, delete the run
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:07 PM Tasks #8252 (Rejected): Activation of cache tiers
- This seems like a dup of #7032.
- 05:11 PM Tasks #8252: Activation of cache tiers
- On the flip side, we should also add nightly tests where cache tiers are disabled (after being completely flushed) wh...
- 05:08 PM Tasks #8252 (Rejected): Activation of cache tiers
- It is unclear whether cache tiers can be enabled while the backing pool is active. Greg has suggested that if the a c...
- 07:30 PM Bug #8232: Race condition during messenger rebind
- Thanks Greg.
I upload the core dump onto dropbox so that you can access here - https://www.dropbox.com/s/35l92m13h... - 12:43 PM Bug #8232: Race condition during messenger rebind
- Oh, and no, there's not a good way to learn about the messenger except by code reading, sorry.
- 11:04 AM Bug #8232: Race condition during messenger rebind
- Guang Yang, this is interesting:...
- 10:49 AM Bug #8232: Race condition during messenger rebind
- I need to study this more, but it looks like we're not correctly handling message resending during repeated reconnect...
- 04:29 AM Bug #8232: Race condition during messenger rebind
- Following are some log snippets related with the ms between the two hosts before crash:...
- 06:39 PM Bug #8254 (Resolved): Not logging missing_on_shards properly
- ...
- 06:23 PM Revision 68b440d6 (ceph): osd: automatically scrub PGs with invalid stats
- If a PG has recnetly split and has invalid stats, scrub it now, even if
it has scrubbed recently. This helps the sta... - 05:03 PM Revision d01aa5bf (ceph): mon: OSDMonitor: return immediately if 'osd tier cache-mode' is a no-op
- Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
- 04:12 PM Revision 5a6b3516 (ceph): Make symlink of librbd to qemu's folder so it can detect it.
- Per issue #7293.
Signed-off-by: Sandon Van Ness <sandon@inktank.com>
(cherry picked from commit 65f3354903fdbdb81468... - 03:44 PM rgw Bug #8251 (Closed): radosgw-agent does not sync objects uploaded to recreated buckets
- If a bucket is deleted and one with the same name is recreated, subsequently uploaded objects to this bucket are no l...
- 03:31 PM Bug #8250 (Resolved): osd crashed "pthread lock: Invalid argument" in upgrade:dumpling-x:stress-s...
- Logs are is http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-28_20:35:04-upgrade:dumpling-x:stress-split-firefl...
- 03:01 PM Bug #7986 (In Progress): 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/2044 ...
- 02:56 PM Feature #7553: Remove classic scrub
- verify that argonaut cluster member are actually impossible
- 02:52 PM Bug #8235 (Rejected): osd/ReplicatedPG.cc: 8262: FAILED assert((data_included.empty() && data.len...
- emailed btrfs list about this one, seems to be btrfs shenanigans
- 02:46 PM Feature #7873 (Resolved): pg query: dump peer_info, peer_missing in all states
- 02:35 PM Feature #8215 (In Progress): OSD op instrumentation plan
- 02:35 PM Feature #7547 (Resolved): Basic docs for Cache Tiering functionality
- 02:34 PM Feature #8155 (In Progress): Disallow changing cache_mode in nonsensical ways
- 10:05 AM Feature #8155 (Fix Under Review): Disallow changing cache_mode in nonsensical ways
- https://github.com/ceph/ceph/pull/1740
- 02:31 PM Feature #6729 (Resolved): Make pg statistics less wrong after split
- 12:09 PM Revision 063b6a27 (ceph): Fixes #8050 Adds a cluster.yaml that is written by interactive task
- 12:09 PM Revision d71a8745 (ceph): These will likely go somewhere better before merge
- 12:09 PM Revision 1532af44 (ceph): Moves node: remote mapping to the internal task.
- 12:09 PM Revision ce778848 (ceph): Changes invocation of serialize_remote_roles to internal task to avoid ...
- 09:37 AM devops Feature #6020 (In Progress): radosgw-apache opinionated package
- So several problems here. The wip branch for actual ceph builds a package which depends on ceph versions of apache an...
- 09:25 AM rgw Documentation #7434 (Resolved): rgw: doc user/group quota
- 09:10 AM rgw Documentation #7434 (In Progress): rgw: doc user/group quota
- 09:22 AM rgw Tasks #8110 (Resolved): rgw: diagram for rgw notifications (zone object sync)
- 09:13 AM devops Feature #7960 (Resolved): backport rpm creation of /usr/lib64/qemu/librbd.so.1 symlink to dumpling
- commit:5a6b35160417423db7c6ff892627f084ab610dfe
- 09:09 AM rgw Feature #7932 (Resolved): Create design for object versioning, including subtasks and estimates
- 06:50 AM Bug #8242 (Duplicate): RPM repository version mismatch
- 02:20 AM Bug #8242 (Duplicate): RPM repository version mismatch
- While invoking 'ceph-deploy install controller01' or even while manually setting the release name, the script tries t...
- 05:58 AM Revision f689e5f0 (ceph): mds: remove discover ino
- Anchor table was the main user of MDCache::discover_ino(), it has
been removed. MDCache::discover_path() can replace ... - 01:11 AM Revision 913a5dd4 (ceph): mds: remove anchor table
- use backtrace instead of anchors to find/open remote inodes
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 01:11 AM Revision 82176000 (ceph): doc: Ensure fastcgi socket doesn't clash with gateway daemon socket.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 01:10 AM Revision 9c9b92f9 (ceph): doc: Verified RHEL configuration.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 01:09 AM Revision ec11bf7e (ceph): doc: Fixed inconsistent header.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 01:08 AM Revision 63b2964b (ceph): doc: Added rhel-6-server-optional-rpms repo.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 01:07 AM Revision 150a1de4 (ceph): Merge branch 'master' of https://github.com/ceph/ceph into wip-doc-radosgw
- 12:28 AM Revision f674f36f (ceph): Copy range using fiemap not entire length
- Under rbd usage, if a volume has tens of thousands of objects and each 4M
object only has several KB(run fio on this ... - 12:08 AM Revision 3920f40a (ceph): rbd-fuse: fix unlink
- The path contains a leading / that needs to be ignored.
Fixes: #8197
Signed-off-by: Josh Durgin <josh.durgin@inktank...
04/28/2014
- 11:52 PM Revision 915bd92f (ceph): Merge pull request #1701 from ceph/wip-libkrbd
- libkrbd convenience library
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 11:51 PM Revision a0867dbd (ceph): Merge pull request #1648 from ceph/wip-client-sleep
- Wip client sleep
Reviewed-by: Sage Weil <sage@inktank.com> - 10:57 PM Revision bab84d45 (ceph): Revert "valgrind.supp: be less picky about library versions"
- This reverts commit f895d16c9e2fd59aab446254e53480cdb91092a1.
- 10:57 PM Revision f261687f (ceph): valgrind: fix tcmalloc suppression for trusty
- Fixes: #8225
Signed-off-by: Sage Weil <sage@inktank.com> - 10:51 PM Revision b3db3a5f (ceph): Merge pull request #1709 from dachary/wip-brag
- brag : useability changes
Reviewed-by: Babu Shanmugam <anbu@enovance.com>
Reviewed-by: Josh Durgin <josh.durgin@inkt... - 10:36 PM Revision 9021b352 (ceph): Merge branch 'wip-rbd-invalidate'
- Reviewed-by: Sage Weil <sage@inktank.com>
- 10:33 PM Revision 818dde31 (ceph): Merge pull request #1737 from steveftaylor/add_rbd_fuse_image_restriction
- Added a new command line parameter (-i or --image=) that allows rbd-fuse...
Reviewed-by: Josh Durgin <josh.durgin@in... - 10:28 PM Revision 11e06061 (ceph): Merge pull request #1699 from chrisglass/python-api-cleanup
- Simple mechanical cleanups
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 10:18 PM Bug #8241 (Resolved): XfsFileStoreBackend tries to set extsize but may get EINVAL
- IRC user pinguini reports a problem with FSSETXATTR:...
- 10:15 PM Revision 5d340d26 (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... - 10:14 PM Revision e08b8b66 (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... - 09:31 PM Revision b1df2c37 (ceph): Changed the -i parameter to -r in order to avoid a conflict with a gene...
- 09:15 PM Revision a0271000 (ceph): rgw: fix url escaping
- Fixes: #8202
This fixes the radosgw side of issue #8202. Needed to cast value
to unsigned char, otherwise it'd get pa... - 09:13 PM Revision d8c5cc67 (ceph): Merge pull request #1652 from ceph/wip-5170-firefly
- Wip 5170 firefly
- 09:11 PM Revision 735a90a9 (ceph): rgw: fix url escaping
- Fixes: #8202
This fixes the radosgw side of issue #8202. Needed to cast value
to unsigned char, otherwise it'd get pa... - 09:11 PM Revision 8cc878e4 (ceph): Merge pull request #1734 from ceph/wip-8202
- rgw: fix url escaping
Reviewed-by: Sage Weil <sage@inktank.com> - 09:06 PM Revision 5544a51d (ceph): Merge pull request #1736 from ceph/wip-7500-wusui
- Fix s3 tests in the rgw workunit.
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com> - 08:56 PM Revision 9e3b8609 (ceph): Fix s3 tests in the rgw workunit.
- Make it possible to set RGW_PORT with ENV variable.
Fixes: 7500
Signed-off-by: Warren Usui <warren.usui@inktank.com> - 08:49 PM Revision 3ec00406 (ceph): Added a new command line parameter (-i or --image=) that allows rbd-fus...
- the mount directory. The purpose of this is to allow a single RBD to be "mounted" in userspace without opening (and l...
- 07:45 PM Revision 060105c3 (ceph): ReplicatedPG: we can get EAGAIN on missing clone flush
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 07:45 PM Revision d83b8f58 (ceph): ReplicatedPG: do not preserve op context during flush
- Any information stashed in the OpContext may be obsolete by the time we
actually mark the object clean. Instead, let... - 07:40 PM Revision a60e15af (ceph): doc/release-notes: v0.67.8 notes
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:23 PM Bug #8237: ceph centos 6 repo broken
- I managed to fix the problem. It's was because of ceph-deploy 1.4.0. It's installing the older packages and repo on n...
- 10:14 AM Bug #8237: ceph centos 6 repo broken
- I am not seeing any problems on our end. I did notice you were using the autobuild.asc but release packages (from you...
- 09:53 AM Bug #8237 (Resolved): ceph centos 6 repo broken
- Having issue install ceph on centos 6. I tried remove the older version and re-install. I tried on a different machin...
- 06:39 PM CephFS Bug #8201 (Resolved): client: (optionally) crash/exit if we are refused reconnect to the mds
- 06:02 PM Revision 2ac27d8f (ceph): Merge pull request #244 from ceph/wip-7199-wusui
- Wip 7199 wusui
- 06:00 PM Revision 694827bc (ceph): Allow scrubbing while thrashing
- Added ability to implement scrubbing while thrashing
(scrub_interval in config can be set to an interval
similar to h... - 05:39 PM Linux kernel client Bug #8226: 0.80~rc1: RBD read errors (ENXIO)
- RBD devices should never report I/O errors. Generally they block and retry instead of reporting an error to the upper...
- 05:33 PM Revision c511894e (ceph): Merge branch 'master' of https://github.com/ceph/ceph into wip-doc-radosgw
- 05:14 PM Revision bcf92c49 (ceph): rgw: fix url escaping
- Fixes: #8202
This fixes the radosgw side of issue #8202. Needed to cast value
to unsigned char, otherwise it'd get pa... - 05:08 PM rbd Bug #8197 (Resolved): Cannot unlink rbd images using rbd-fuse
- Thanks, added to master!
- 04:56 PM Feature #8147 (Fix Under Review): osd: make split automatically trigger scrub
- 04:42 PM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
- There were plenty of other timeouts before the suicide, and this sort of situation is not specific to ceph.parent bat...
- 03:11 PM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
- Can you get me the backtraces from the suiciding OSDs? That shouldn't be happening regardless of how many ops we dump...
- 03:01 PM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
- 1) I've had batches of ceph.parent ops pile up ever since ceph.parent was introduced (early 0.7x IIRC); I'm running 0...
- 12:58 PM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
- Hmm, there are several things going on here that you're conflating. Some questions and notes
1) Exploding a tarball ... - 07:40 AM CephFS Feature #8230: mds: new requests are not throttled by backend rados progress
- This change might turn out to have another unexpected benefit, although the reason for that might turn out to be a bu...
- 04:12 PM Revision 2cbe1dc0 (ceph): Only attempt to use sudo if necessary
- 03:56 PM Bug #8225 (Resolved): valgrind failures on trusty
- teuthology.git f261687f292df47b7a5296814480713c3c3d306f
- 03:36 PM Bug #8239: log [WRN] : slow request 30.404834 seconds old, received at 2014-04-26 04:05:56.539287...
- can you enable logging on osd.8 as well (sudo ceph --admin-daemon /var/run/ceph/ceph-osd.8.asok config set debug_ms 1...
- 03:05 PM Bug #8239: log [WRN] : slow request 30.404834 seconds old, received at 2014-04-26 04:05:56.539287...
- Here is the log from osd.0 after a recurrence.
- 02:29 PM Bug #8239 (Need More Info): log [WRN] : slow request 30.404834 seconds old, received at 2014-04-2...
- 02:28 PM Bug #8239 (Resolved): log [WRN] : slow request 30.404834 seconds old, received at 2014-04-26 04:0...
- empty cluster, waiting on more info
root@storage0:~# ceph --admin-daemon /var/run/ceph/ceph-osd.0.asok dump_ops_in... - 02:56 PM Feature #8227: RFE: introduce “back in a bit” osd state
- I can't really follow the story you're telling, but even if you are backfilling only changed objects, you still have ...
- 02:36 PM Feature #8227: RFE: introduce “back in a bit” osd state
- Greg, it looks like you're discussing difficulties related with the more elaborate plans for the log-only replica, wh...
- 11:23 AM Feature #8227: RFE: introduce “back in a bit” osd state
- You're missing the ways in which this differs from existing recovery mechanisms:
1) For this state, we would want to... - 02:39 PM devops Tasks #8240 (Resolved): Build 0.67.8 & 0.80 on RHEL7-RC
- Hope the Subject is sufficient..
- 02:10 PM rgw Bug #8202 (Resolved): rgw: failure to copy objects with chinese names
- 12:41 PM Bug #8232: Race condition during messenger rebind
- Actually Guang Yang, do you have a copy of the full backtrace you're concerned about? It's a little mangled from the ...
- 10:11 AM Bug #8232: Race condition during messenger rebind
- #6992 was backported, but there's not been a Dumpling release containing the backport yet. A new patch isn't needed, ...
- 08:59 AM Bug #8232 (Fix Under Review): Race condition during messenger rebind
- Greg, Please review the linked PR.
- 05:03 AM Bug #8232: Race condition during messenger rebind
- > A simple fix of the issue, is to invoke mark down all pipes from within the
> accepter thread before and after its... - 04:56 AM Bug #8232: Race condition during messenger rebind
- Please help to review the pull request - https://github.com/ceph/ceph/pull/1733
- 10:49 AM Feature #7873 (New): pg query: dump peer_info, peer_missing in all states
- 10:25 AM devops Bug #6726 (Resolved): Official packages do not appear to be available for Saucy
- Closing this out. We now have builds of dumpling, emperor, firefly and testing with saucy/trusty builds.
- 08:19 AM Bug #8235 (Rejected): osd/ReplicatedPG.cc: 8262: FAILED assert((data_included.empty() && data.len...
- ubuntu@teuthology:/var/lib/teuthworker/archive/samuelj-2014-04-27_12:36:12-rados-wip-sam-testing-testing-basic-plana/...
- 07:07 AM Revision 27ec495a (ceph): Added Java Example
- 05:55 AM rgw Bug #8233 (Resolved): Installation & Documentation broken for Ubuntu Trusty 14.04 - rgw
- https://ceph.com/docs/master/install/install-ceph-gateway/
The repository configured by this process contains apac...
04/27/2014
- 07:46 PM Bug #8232 (Resolved): Race condition during messenger rebind
- When the system is in high load, we observed an assertion failure as below:
----------------------------------------... - 01:42 PM Revision 8f64b5c1 (ceph): Update librados-intro.rst
- 01:40 PM Revision b8aa58af (ceph): client: drop dirty/flushing caps if auth MDS' session is reset
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 01:40 PM Revision 70ab0793 (ceph): client: wake up cap waiters if MDS session is reset
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 01:40 PM Revision 3e41f92b (ceph): client: cleanup unsafe requests if MDS session is reset
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 01:40 PM Revision 09a1bc5a (ceph): client: add asok command to kick sessions that were remote reset
- Fixes: #8021
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 12:20 PM Feature #8231 (Resolved): ceph filestore dump improvements
- ceph-filestore-dump ... <pgid> <object> (get|set)-bytes <from> <to>
ceph-filestore-dump ... <pgid> <object> (get|set... - 12:15 PM Revision 998b365c (ceph): Changed the java code example
- 10:23 AM Bug #8180: osd.3 crashed in upgrade:dumpling-x:stress-split-firefly-distro-basic-vps
- And one more similar crash logs are in - http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-26_20:35:01-upgrade:d...
- 10:08 AM CephFS Feature #8230 (New): mds: new requests are not throttled by backend rados progress
- The use of the ceph.parent attribute as actual cephfs metadata, including its use for recovery and hard links, has tu...
- 06:48 AM CephFS Bug #8201 (Fix Under Review): client: (optionally) crash/exit if we are refused reconnect to the mds
- https://github.com/ceph/ceph/pull/1648/
- 01:51 AM Bug #8229 (Closed): 0.80~rc1: OSD crash (domino effect)
- Situation: degraded cluster recovering/remapping ~20% after replacing some OSDs.
During recovery I reboot two server...
04/26/2014
- 06:44 PM CephFS Bug #8211 (Resolved): 0.80~rc1: MDS failed to respawn
- 06:36 PM CephFS Bug #8200 (Resolved): failing kclient_workunit_kclient test
- fixed by commit f74d66a3ec1b62a663451083091ccb8341d721ec
- 04:23 PM Bug #8175 (Fix Under Review): Some values of target_max_objects for tiering will crash OSDs
- 03:10 PM Bug #8214: Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" in upgrade:dumplin...
- One more like this:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-25_20:35:06-upgrade:dumpling-x:stress-spl... - 12:49 PM Revision c1bf7dbb (ceph): Merge pull request #1729 from ceph/wip-7966
- readlink result in resapwn
Reviewed-by: Yan, Zheng <zheng.z.yan@intel.com> - 12:19 PM Feature #8195: shorten window of highest risk during recovery
- My proposal is much simpler than that, actually. In the simplest implementation possible, we'd just change the start...
- 12:18 PM Bug #8228: 0.80~rc1: OSD crash: segfault in libtcmalloc.so.4.1.2
- ...
- 12:13 PM Bug #8228 (Can't reproduce): 0.80~rc1: OSD crash: segfault in libtcmalloc.so.4.1.2
- OSD suddenly crashed on Debian GNU/Linux_3.14.1 x86_64; (~2 of ~12 GiB of RAM allocated).
From '/var/log/messages'... - 11:35 AM Feature #8227: RFE: introduce “back in a bit” osd state
- Err... I'm confused; it looks like we already do pretty much everything it would take to implement this feature. Sa...
- 09:44 AM Feature #8227: RFE: introduce “back in a bit” osd state
- We've discussed this sort of log-only replica at a few times in the past. It's conceptually simple, but unfortunately...
- 09:29 AM Feature #8227 (Resolved): RFE: introduce “back in a bit” osd state
- Sometimes I want to bring an osd down for a bit, say because it is slowing the cluster down, because I want to run so...
- 09:28 AM Bug #8180: osd.3 crashed in upgrade:dumpling-x:stress-split-firefly-distro-basic-vps
- I see it in more tests, e.g.:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-25_20:35:06-upgrade:dumpling-x:... - 09:14 AM Bug #7891: osd: leaked pg refs on shutdown
- ubuntu@teuthology:/a/teuthology-2014-04-25_02:30:12-rados-master-testing-basic-plana/214284/remote
- 05:53 AM CephFS Bug #7966 (Resolved): ceph-mds respawn doesn't always work
- 04:12 AM Linux kernel client Bug #8226 (Resolved): 0.80~rc1: RBD read errors (ENXIO)
- With Ceph_0.80~rc1 and Linux_3.14.1 I'm getting read errors on RBD devices.
from '/var/log/messages':... - 02:46 AM Revision 5d497826 (ceph): mds: terminate readlink result in resapwn
- readlink(2) does not null terminate the buffer; we need to do that.
Fixes: #7966
Signed-off-by: Sage Weil <sage@inkt...
04/25/2014
- 11:20 PM Revision 58d7640d (ceph): Merge pull request #1727 from ceph/wip-8193
- ceph_test_rados_api_tier: increase HitSetTrim timeouts
- 11:00 PM Revision 438b5789 (ceph): Merge pull request #1700 from xanpeng/patch-1
- Fix error in mkcephfs.rst
Signed-off-by: Xan Peng <xanpeng@gmail.com>
Reviewed-by: Sage Weil <sage@inktank.com> - 10:58 PM Revision 0062070e (ceph): Merge pull request #1725 from FlorentCoppint/master
- Skipping '_netdev' Debian fstab option
Reviewed-by: Sage Weil <sage@inktank.com> - 10:49 PM Revision d0f1806d (ceph): ceph_test_rados_api_tier: increase HitSetTrim timeouts
- ...so that they pass when they get unlucky with thrashing.
This will vastly decrease the probability of failure, but... - 09:21 PM Bug #8225 (Resolved): valgrind failures on trusty
- ...
- 08:22 PM Revision f102e494 (ceph): Post last_in_suite jobs, but delete when run
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:01 PM Revision 5de353e7 (ceph): Update unit test for Cluster.__repr__()
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:45 PM CephFS Bug #7966 (Fix Under Review): ceph-mds respawn doesn't always work
- 07:02 PM CephFS Bug #8211: 0.80~rc1: MDS failed to respawn
- Debian GNU/Linux x86_64 (i.e. amd64)
- 10:23 AM CephFS Bug #8211: 0.80~rc1: MDS failed to respawn
- Yeah. What environment are you running under?
- 01:38 AM CephFS Bug #8211 (Resolved): 0.80~rc1: MDS failed to respawn
- From the MDS log (timestamps dropped):...
- 04:36 PM Revision e6e28744 (ceph): Fix Cluster.__repr__()
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:20 PM Bug #8193 (Resolved): HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher ru...
- Request pulled
- 03:48 PM Bug #8193 (Fix Under Review): HitSetTrim test in test/librados/tier.cc needs to be skipped if thr...
- https://github.com/ceph/ceph/pull/1727
- 10:33 AM Bug #8193: HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher running
- I should have mentioned that there are 2 HitSetTrim tests as is typical.
- 03:50 PM Bug #8183 (Won't Fix): osd: In a tiered pool after successful removal request object still appear...
- This is as expected. The list_objects and rados ls are basically not coherent when you're using a cache pool. In th...
- 01:40 PM rgw Feature #418 (Duplicate): rgw: object versioning
- 01:31 PM rgw Feature #8224 (Resolved): rgw: test suite for object versioning
- 01:31 PM rgw Documentation #8223 (Closed): rgw: document object versioning
- 01:28 PM rgw Feature #8222 (Resolved): rgw: object versioning, object creation changes
- 01:25 PM rgw Feature #8221 (Resolved): rgw: object versioning swift support
- 01:24 PM rgw Feature #8220 (Resolved): rgw: object versioning, garbage collector changes
- 01:24 PM rgw Feature #8219 (Resolved): rgw: object versioning RESTful api
- list bucket versions
set bucket versioning
get bucket versioning
- 01:23 PM rgw Feature #8218 (Resolved): rgw: object versioning manifest changes
- 01:22 PM rgw Feature #8217 (Resolved): rgw: object versioning object overwrite / delete changes
- 01:22 PM rgw Feature #8216 (Resolved): rgw: object versioning objclass support
- 01:17 PM Feature #8215 (Closed): OSD op instrumentation plan
- We need a system for gaining insight on contended OSD resources and for understanding op handling bottlenecks.
syste... - 10:21 AM rbd Bug #7620: BUG: soft lockup - CPU#0 stuck for 23s!
- Test failed with similar error.
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-24_20:35:03-upg... - 10:18 AM Bug #8156: Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" in upgrade:dumplin...
- See #8214
- 10:17 AM Bug #8214 (Can't reproduce): Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" ...
- This look similar to #8156, but has no out of memory problem.
Logs are in http://qa-proxy.ceph.com/teuthology/teut... - 10:14 AM rgw Bug #8213 (Duplicate): RGW is creating empty pool names
- Yehuda says he's fixed several of these bugs, but they apparently haven't been backported to Dumpling, and there migh...
- 10:00 AM devops Feature #5397: terminate ceph-create-keys when its mon process dies
- I don't think this is an issue anymore is it?...
- 09:54 AM rgw Documentation #7434: rgw: doc user/group quota
- Should get merged into master today or tomorrow. The wip-doc-radosgw branch also conflates an update to the configura...
- 09:52 AM rgw Bug #8202 (In Progress): rgw: failure to copy objects with chinese names
- 09:50 AM rgw Bug #8194: rgw: test_region_copy_object fails with erasure coding
- 09:40 AM rgw Bug #7799 (Can't reproduce): Errors in upgrade:dumpling-x:stress-split-firefly---basic-plana suite
- 08:41 AM devops Bug #7617 (Resolved): ceph-deploy uninstall should document why it doesn't remove all relevant pa...
- Merged into ceph-deploy's master branch with hash: 4ba3fa3
- 07:20 AM Revision 9ac264a8 (ceph): Skipping '_netdev' Debian fstab option
- Signed-off-by: Florent Bautista <florent@coppint.com>
- 07:14 AM rbd Bug #8197 (Fix Under Review): Cannot unlink rbd images using rbd-fuse
- Josh - please review the patch in the text
- 06:53 AM Fix #8205 (In Progress): FileStore: properly fill in XATTR_NO_SPILL_OUT tag
- wip-xattr-spillout exists but is untested.
- 06:50 AM Bug #8212 (Resolved): Update Web docs for building ceph
- I attempted to build ceph using the instructions here:
http://ceph.com/docs/master/install/build-ceph/
The "./con... - 06:47 AM CephFS Bug #8200: failing kclient_workunit_kclient test
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-23_23:01:45-kcephfs-master-testing-basic-plana/212542/
- 04:58 AM Bug #8009 (Closed): librados failing tests for APILock
- Was unable to reproduce this with the original config.yaml (from the failed run), nor was I able to reproduce it manu...
04/24/2014
- 11:48 PM Revision 499adb1d (ceph): rados.h,ReplicatedPG: add CEPH_OSD_FLAG_ENFORCE_SNAPC and use on flush
- We need to ensure that even with pool snaps, we use the snapc provided in order
to ensure that the clones are written... - 09:29 PM Revision ee69c7a4 (ceph): rgw: update idle_timeout for rgw_s3tests_multiregion.yaml
- Fixes: #8194
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:49 PM rgw Fix #8210 (New): rgw: bulk deletes are slow
- Ceph deletions are a bit problematic. They didn't implement bulk deletes correctly. Instead of being faster, they are...
- 08:42 PM Revision 9b37398d (ceph): Merge pull request #1717 from dachary/wip-auid
- mon: add ceph osd pool set <pool> auid
Reviewed-by: Greg Farnum <greg@inktank.com> - 08:27 PM Revision e8b13f71 (ceph): Merge pull request #1724 from ceph/wip-uselocalgithubforqemu-wusui
- Use new git mirror for qemu-iotests
- 07:55 PM Revision ddf37d90 (ceph): Use new git mirror for qemu-iotests
- Fixes: 8191
Signed-off-by: Warren Usui <warren.usui@inktank.com> - 07:48 PM Revision 1885792c (ceph): ECBackend::continue_recovery_op: handle a source shard going down
- get_min_avail_to_read_shards might return an error if there are
no longer enough sources to reconstruct the missing s... - 05:56 PM Bug #8193: HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher running
This particular test case is timing sensitive. It doesn't make sense to run it when the thrasher is running. This ...- 04:27 PM rgw Bug #8194: rgw: test_region_copy_object fails with erasure coding
- I pushed an update to the test suite, should be ok now.
- 02:24 PM rgw Bug #8194: rgw: test_region_copy_object fails with erasure coding
- Cross region copy is just too slow due to ec backend. Apache ends up timing out. Need to increase the idle_timeout pa...
- 03:51 PM Bug #8207 (Duplicate): "[ERR] 3.6 missing primary copy.." in upgrade:dumpling-x:stress-split-fire...
- This cold be a duplicate of #7976
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-23_19:55:03-u... - 03:46 PM Revision af209851 (ceph): Don't push last_in_suite jobs to paddles
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 02:44 PM Bug #8206 (Duplicate): "osd.4 ...[ERR] : 3.14 push" in upgrade:dumpling-x:stress-split-firefly-di...
- This one was not reproduced on manual re-run.
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-2... - 02:25 PM Fix #8205 (Resolved): FileStore: properly fill in XATTR_NO_SPILL_OUT tag
- Right now, the only way the FileStore sets the XATTR_SPILL_OUT_NAME xattr to contain XATTR_NO_SPILL_OUT is when remov...
- 02:21 PM Bug #8204 (Duplicate): "timed out waiting for admin_socket to appear after osd.5 restart" in upgr...
- I could not reproduce it manually, but after consulting with devel still logging, so we can trace to similar race con...
- 01:15 PM Feature #8203 (Resolved): Replica setting values in df output
- The ability to see replica settings in ceph df. Potentially a warning when the replica is less than X in value, where...
- 01:01 PM devops Bug #7617 (Fix Under Review): ceph-deploy uninstall should document why it doesn't remove all rel...
- PR opened https://github.com/ceph/ceph-deploy/pull/182
- 10:06 AM devops Bug #7617 (In Progress): ceph-deploy uninstall should document why it doesn't remove all relevant...
- 12:50 PM Bug #8161: osd/ECBackend.cc: 475: FAILED assert(r == 0)
- 10:16 AM Bug #8161: osd/ECBackend.cc: 475: FAILED assert(r == 0)
- 12:29 PM rgw Bug #8202 (Resolved): rgw: failure to copy objects with chinese names
- From mailing list:...
- 11:30 AM CephFS Bug #8201 (Resolved): client: (optionally) crash/exit if we are refused reconnect to the mds
- currently we hang and there is no way for users of the fs to know that it is not going to unhang in the future.
- 11:29 AM CephFS Bug #8200: failing kclient_workunit_kclient test
- teuthology-2014-04-20_23:04:17-kcephfs-master-testing-basic-plana/206396/
- 11:27 AM CephFS Bug #8200 (Resolved): failing kclient_workunit_kclient test
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-22_23:05:45-kcephfs-firefly-testing-basic-plana/210650/
http:... - 10:04 AM Bug #7891: osd: leaked pg refs on shutdown
- ubuntu@teuthology:/a/sage-2014-04-23_18:03:07-rados-firefly-testing-basic-plana/211808
- 10:03 AM Bug #8199 (Resolved): rados unit test failure: LibRadosTwoPoolsECPP.FlushTryFlushRaces hang
- ubuntu@teuthology:/a/sage-2014-04-23_18:03:07-rados-firefly-testing-basic-plana/211806
- 09:55 AM rbd Bug #8184 (Fix Under Review): krbd: make sure we have latest osdmap on 'rbd map'
- wip-rbd-maposdmap-v2;
"rbd: make sure we have latest osdmap on 'rbd map'" on ceph-devel. - 09:02 AM Bug #7922: osd: multi-backfill reservation does not release on reject
- Kenneth Waegeman wrote:
> Is this fixed in 0.79 ? Or can I patch this myself? I seem to have this problem too
Yes... - 08:57 AM Bug #7922: osd: multi-backfill reservation does not release on reject
- Is this fixed in 0.79 ? Or can I patch this myself? I seem to have this problem too
- 08:29 AM Revision c0c2361b (ceph): brag : implement --verbose on client
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 08:18 AM Revision 70092110 (ceph): brag : document the zero argument behavior
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 08:18 AM Revision 2b16a818 (ceph): brag : meaningfull error messages
- To help figure out problems, include the error message in the output
when a submission fails.
Signed-off-by: Loic Da... - 08:15 AM rbd Feature #2467 (New): qemu: implement bdrv_invalidate_cache
- 06:45 AM rbd Bug #8197 (Resolved): Cannot unlink rbd images using rbd-fuse
- rbdfs_unlink is calling find_openrbd with the wrong path. The following patch fixes it.
diff --git a/src/rbd_fuse... - 05:59 AM devops Bug #7356: Kill all while loops that will never end....
- This is still an issue.
- 01:00 AM Revision 2708c3c5 (ceph): Merge remote-tracking branch 'gh/firefly'
- 12:58 AM rbd Bug #8178 (Resolved): 0.79: feature set mismatch, my 4a042a42 < server's 104a042a42, missing 1000...
- 12:58 AM rbd Bug #8178: 0.79: feature set mismatch, my 4a042a42 < server's 104a042a42, missing 1000000000
- In terms of features, 3.13 is almost 6 months old (3.13-rc1 was
released 5 months ago). But yeah, we should definit... - 12:55 AM Linux kernel client Feature #8196 (New): Document which features are supported by the kernel client
- Document which kernel supports what and the possible pitfalls, like
#8178 and a number of "feature set mismatch" thr... - 12:23 AM Revision d384d3a6 (ceph): Merge pull request #1720 from jdurgin/wip-list-children-test
- test_rbd.py: ignore children in cache pools
Reviewed-by: Sage Weil <sage@inktank.com>
04/23/2014
- 11:07 PM Revision 5b979766 (ceph): Merge pull request #1719 from ceph/wip-8168
- Wip 8168
Reviewed-by: Sage Weil <sage@inktank.com> - 09:11 PM Revision 39c1bfc4 (ceph): ReplicatedPG::do_op: don't return ENOENT for whiteout on snapdir read
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 09:11 PM Revision 83f89348 (ceph): ReplicatedPG::do_osd_ops: consider head whiteout in list-snaps
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 08:47 PM Revision 76a21389 (ceph): Merge pull request #1718 from ceph/wip-7882-wusui
- Support latest qemu iotest code
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 08:28 PM Revision a83aff54 (ceph): test_rbd.py: ignore children in cache pools
- This is necessary until http://tracker.ceph.com/issues/8187 is fixed.
Signed-off-by: Josh Durgin <josh.durgin@inktan... - 08:20 PM Revision aae16ab3 (ceph): mon: add ceph osd pool set <pool> auid
- When a pool is created with ceph osd pool create, the auid is not
inferred from the session auid and is set to zero. ... - 08:20 PM Revision 606e725e (ceph): Support latest qemu iotest code
- Modified qemu-iotests workunit script to check for versions
that use the latest qemu (currently only Trusty). Limit ... - 06:11 PM Feature #8195: shorten window of highest risk during recovery
In the current scheme since the primary runs through the objects in a hashed order it allows new writes before or a...- 04:53 PM Feature #8195 (New): shorten window of highest risk during recovery
- Say a 3-sized PG experienced failure of two OSDs, the second one failing when the first replacement was part-way thro...
- 05:13 PM Bug #8193: HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher running
2014-04-22T16:26:11.096 INFO:teuthology.task.workunit.client.0.out:[10.214.131.16]: [ RUN ] LibRadosTierECPP.H...- 02:16 PM Bug #8193 (Resolved): HitSetTrim test in test/librados/tier.cc needs to be skipped if thrasher ru...
- Command failed on 10.214.131.16 with status 1: 'mkdir -p --
/home/ubuntu/cephtest/mnt.0/client.0/tmp && cd -... - 04:32 PM rgw Bug #8194 (Resolved): rgw: test_region_copy_object fails with erasure coding
- It looks like this has been failing since we added erasure coding pools, but this is the most recent one:
http://q... - 04:05 PM Bug #8168 (Resolved): osd: rbd_test.test_diff_iterate fails with a cache pool
- 03:51 PM Revision 4ed25fdb (ceph): Merge pull request #1714 from ceph/wip-fs-client
- two small fixes for client
Reviewed-by: Sage Weil <sage@inktank.com> - 03:07 PM Bug #8113 (Fix Under Review): agent_work can be continuously rescheduled during recovery while mo...
- 02:55 PM Bug #8066 (Duplicate): osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + fire...
- 02:55 PM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- Actually, osd.3 prematurely advanced last_backfill, was still on dumpling.
ubuntu@teuthology:/var/lib/teuthworker/... - 02:42 PM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- 2014-04-23 05:24:38.517382 7f7f6945d700 10 osd.0 911 dequeue_op 0x2b4d780 prio 127 cost 0 latency 0.000238 pg_backfil...
- 06:52 AM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-22_19:55:02-upgrade:dumpling-x:stress-split-firefly...
- 02:25 PM Bug #8192 (Duplicate): osd.0 crashed in upgrade:dumpling-x:stress-split-firefly---basic-plana
- Duplicate of 8180
- 02:11 PM Bug #8192 (Duplicate): osd.0 crashed in upgrade:dumpling-x:stress-split-firefly---basic-plana
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-22_19:55:02-upgrade:dumpling-x:stress-split-firefl...
- 11:10 AM Feature #8189 (New): ceph: display tier relationships visually
- Currently tiering information is only available via 'ceph osd dump'. It's hard for humans to discern the tiering/cach...
- 11:05 AM Bug #8185 (Won't Fix): ceph osd pool create does not set auid
- comments I made in irc:
I don't think we want to automatically set auids; they aren't expected in a lot of the infra... - 06:46 AM Bug #8185 (Fix Under Review): ceph osd pool create does not set auid
- "proposed fix":https://github.com/ceph/ceph/pull/1715
- 06:09 AM Bug #8185 (Won't Fix): ceph osd pool create does not set auid
- When a pool is created using the command line by a user associated with *auid*, the pool is not associated (owned) by...
- 11:04 AM Feature #8188 (Resolved): librados: interface to inspect pool properties
- Right now the only way to view pool properties about tiering (and probably a few other things) is via 'ceph osd dump'...
- 10:59 AM rbd Bug #8187 (Resolved): librbd: list_children() reports duplicates with cache pools
- list_children() and the internals of snap_unprotect() both go through all pools to check for children of a snapshot. ...
- 09:33 AM Revision 26517504 (ceph): rbd: add libkrbd convenience library
- Add libkrbd libtool convenience library to provide an interface for
mapping and unmapping rbd images programmatically... - 09:33 AM Revision 2521e73a (ceph): mount.ceph: switch to module_load()
- Implement modprobe() in terms of module_load() from common/module.h
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktan... - 09:33 AM Revision 0ba3960c (ceph): rbd: switch to libkrbd for 'rbd {map,showmapped,unmap}' operations
- Thanks to libkrbd, 'rbd map' now outputs the device node it mapped to
to stdout:
$ sudo rbd map foo
/dev/rbd... - 09:33 AM Revision 4238ffdc (ceph): doc: do not mention modprobe in rbd docs
- rbd binary will load rbd.ko itself, with appropriate options. Loading
it by hand with default options is undesirable... - 09:33 AM Revision 0c2b0fb8 (ceph): doc: 'rbd showmapped' doesn't need privileges
- No need to run 'rbd showmapped' with sudo.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com> - 09:33 AM Revision f6318545 (ceph): rbd: deprecate --no-settle option
- Waiting for udev has been the default for a while now, and, after
switching to libkrbd, is no longer an option. (lib... - 09:31 AM rbd Bug #8178: 0.79: feature set mismatch, my 4a042a42 < server's 104a042a42, missing 1000000000
- Dear Ilya,
You got the right impression but I didn't even mapped anything from new erasure pool when connected RBD... - 12:20 AM rbd Bug #8178: 0.79: feature set mismatch, my 4a042a42 < server's 104a042a42, missing 1000000000
- Hi Dmitry,
I'm assuming what you did is you created an EC pool, tried to map an
image out of the replicated pool,... - 02:29 AM Revision bad34e90 (ceph): client: check cap ID when handling cap export message
- handle following sequence of events:
- mds0 exports an inode to mds1. client receives the cap import
message from m... - 02:14 AM Revision 383d21dc (ceph): client: avoid releasing caps that are being used
- To avoid releasing caps that are being used, encode_inode_release()
should send implemented caps to MDS.
Signed-off-... - 12:42 AM Revision 3a2c8886 (ceph): rados: add ec and rep lost_unfound_delete tests
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 12:39 AM Revision e64d8314 (ceph): task/: add tests for ec and rep mark_unfound_lost delete
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 12:36 AM Revision d726251f (ceph): doc: Fix hyperlink to CRUSH maps.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:36 AM Revision a06f8667 (ceph): Merge pull request #1713 from ceph/wip-7439
- Wip 7439
Reviewed-by: Sage Weil <sage@inktank.com> - 12:31 AM Revision 6902e224 (ceph): doc: Added cache tiering settings to ceph osd pool set.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:30 AM Revision 0d964bc6 (ceph): doc: Added new cache tiering doc to index/TOC.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:30 AM Revision 44e4e3d5 (ceph): doc: Added new cache tiering doc to main docs.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
04/22/2014
- 11:17 PM rbd Bug #8184: krbd: make sure we have latest osdmap on 'rbd map'
- An attempt is in wip-rbd-maposdmap, Sage suggested the mon_get_version approach.
- 11:14 PM rbd Bug #8184 (Resolved): krbd: make sure we have latest osdmap on 'rbd map'
- ...
- 10:46 PM Revision 8350b6e4 (ceph): Bump psutil version requirement
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 10:43 PM Revision 2182815c (ceph): ReplicatedPG: handle ec pools in mark_all_unfound_lost
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 09:21 PM Revision 6769f4dc (ceph): 0.80-rc1
- 08:37 PM Revision 245923e7 (ceph): ReplicatedPG: enable mark_unfound_lost delete for ec pools
- revert is tricky to implement at this time for ec pools, so
we'll instead just implement delete for ec pools.
Fixes:... - 07:51 PM Bug #8183 (Won't Fix): osd: In a tiered pool after successful removal request object still appear...
What works:
After a removal a second removal returns ENOENT
After a removal the creation of another object with t...- 07:44 PM Revision 387110b1 (ceph): rados/singleton/all/cephtool: whitelist scrub vs split vs agent issue
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:41 PM Bug #8182 (Rejected): After rados bench on tiered pool can't remove objects
- 07:14 PM Bug #8182: After rados bench on tiered pool can't remove objects
- My trace is after a subsequent removal attempt which wouldn't be clear from the description. Now that I look at the ...
- 07:10 PM Bug #8182 (Rejected): After rados bench on tiered pool can't remove objects
- I'm based on the firefly branch with changes to the tiering agent code which shouldn't affect this test.
$ ./rados... - 07:40 PM Revision 47866fd2 (ceph): Merge pull request #1691 from ceph/wip-8139
- osd_types: pg_t: allow is_split to handle checks for splits prior to the most recent
Reviewed-by: Samuel Just <sam.j... - 07:35 PM Revision 9078513c (ceph): Fix for #8115
- Increase boot disk size per #8115 where monitors shut down due to
/ being full on vm machines.
Signed-off-by: Sandon... - 07:16 PM CephFS Bug #8177 (Duplicate): Client: seg fault in verify_reply_trace on traceless reply
- 07:11 AM CephFS Bug #8177: Client: seg fault in verify_reply_trace on traceless reply
- No idea if they are same.
- 06:52 AM CephFS Bug #8177: Client: seg fault in verify_reply_trace on traceless reply
- See #5021 and wip-5021; possibly related (or same)?
The wip-5021 worked okay except that it caused a crash with sm... - 06:59 PM Revision 009e8746 (ceph): qa/workunits/rbd/copy.sh: skip some tests when tiering is enabled
- The rados ls bit doesn't work.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@ink... - 06:59 PM Revision c0bff439 (ceph): qa/workunits/rbd/copy.sh: fix test
- I broke this in commit 9d64ac66082bd108ec3c2a74e2e77475b5564eae.
Signed-off-by: Sage Weil <sage@inktank.com>
Signed-... - 06:25 PM Revision 5daf5385 (ceph): ECBackend: use std::swap for boost::optional
- Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Samuel Just <sam.just@inktank.com> - 05:42 PM Feature #7439 (Resolved): EC: adapt unfound teuthology tests and add to nightly for EC
- 04:43 PM Revision 90040490 (ceph): rbd: use stringify() in options parsing routines
- Use stringify() in map_option_{uuid,ip,int}_cb() instead of essentially
open-coding it.
Signed-off-by: Ilya Dryomov ... - 04:43 PM Revision 070a8208 (ceph): configure: check for blkid/blkid.h header
- The check for the presence of blkid/blkid.h was missing.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com> - 04:43 PM Revision 944dd1c6 (ceph): Makefile: build common/secret.c with libtool
- Turn common/secret.c into a libtool convenience library, libsecret.la.
Currently it is build directly, twice: for mou... - 04:43 PM Revision ac9b461f (ceph): common: add module_{load,has_parameter}()
- Add two kernel module helpers: module_{module,has_parameter}(). They
are going to live in common/module.[ch].
Signe... - 04:43 PM Revision be081dbd (ceph): stringify: use ostringstream instead of stringstream
- Use ostringstream, as we don't need both input and output of the
stringstream in stringify().
Signed-off-by: Ilya Dr... - 04:02 PM Revision 6cb5ce86 (ceph): Merge pull request #1710 from ceph/wip-coverity
- a couple coverity fixes
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com> - 03:51 PM Bug #8168: osd: rbd_test.test_diff_iterate fails with a cache pool
- wip-8168 didn't fix it:...
- 02:59 PM Bug #8168: osd: rbd_test.test_diff_iterate fails with a cache pool
- Here's the osd/filestore log of one of the failing list-snaps calls:...
- 02:14 PM Bug #8168: osd: rbd_test.test_diff_iterate fails with a cache pool
- So the scenario that is failing (still fails on master) is that after the objects are deleted via rbd_discard(), diff...
- 03:37 PM Revision f244109c (ceph): Merge pull request #1711 from ceph/wip-coverity-respawn
- mds: make strncpy in ::respawn safer
Reviewed-by: Sage Weil <sage@inktank.com> - 03:31 PM Revision cac15c7d (ceph): mds: make strncpy in ::respawn safer
- Previous code assumed null terminated argv[0]
was not longer than PATH_MAX and the resulting
strncpy was not strictly... - 03:29 PM Revision b4eb5025 (ceph): osd/osd_types: RWState: initialize snaptrimmer_write_marker
- ** CID 1204295: Uninitialized scalar field (UNINIT_CTOR)
/osd/osd_types.h: 2716 in ObjectContext::RWState::RWState(... - 03:28 PM Revision 4e5f4420 (ceph): osdc/Objecter: drop unused field
- This as missed by 860d72770cdf092c027d50f4ee03bed76c975599.
** CID 1204296: Uninitialized scalar field (UNINIT_CTO... - 03:27 PM Revision 124a663a (ceph): doc/release-notes: a bit of prose about firefly
- Signed-off-by: Sage Weil <sage@inktank.com>
- 02:00 PM Bug #8180 (Duplicate): osd.3 crashed in upgrade:dumpling-x:stress-split-firefly-distro-basic-vps
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-21_20:35:06-upgrade:dumpling-x:stress-split-firefl...
- 01:11 PM Bug #8036 (In Progress): levedb: throws std::bad_allow on 14.04
- 01:10 PM Bug #7942 (Resolved): promote uses cloneid, but backend may have a different cloneid
- 01:08 PM Bug #7398 (Resolved): osd: ERANGE from clone
- 7e697b1bc2ffac086b6a24f97aba755401cd8c37
- 01:07 PM Bug #8067 (Duplicate): mon: enomem on vps, killed at ~800MB
- 01:06 PM Bug #8082 (Duplicate): hung recovery
- 01:05 PM Bug #7987 (Duplicate): osd: backfill/recovery makes no progress
- 12:41 PM devops Feature #7716 (Resolved): Build debug packages for EL6
- Verified the centos release buid machine is now including the debug packages with today's build (the build I schedule...
- 12:39 PM Bug #8139 (Resolved): osd/osd_types.cc: 398: FAILED assert(m_seed < old_pg_num)
- 11:57 AM devops Bug #7356 (Need More Info): Kill all while loops that will never end....
- Is this still an issue?
- 08:07 AM devops Bug #6726: Official packages do not appear to be available for Saucy
- Tom Verdaat wrote:
> Lauri Vant wrote:
> > When can we expect this to be resolved?
>
> Based on my understanding... - 06:48 AM devops Bug #6726: Official packages do not appear to be available for Saucy
- Note that the Trusty Ubuntu archive will contain 0.80 (it already contains 0.79) once it's released by Inktank.
- 05:44 AM devops Bug #6726: Official packages do not appear to be available for Saucy
- Lauri Vant wrote:
> When can we expect this to be resolved?
Based on my understanding of Sandon's replies above: ... - 04:26 AM Revision 66170f39 (ceph): osd/osd_types: pg_interval_t: dump primary
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:26 AM Revision 931ae6b8 (ceph): osd/osd_types: pg_interval_t: include up_primary in pg_interval_t
- Nothing uses this, but it triggers a new interval, which makes it confusing
when it is not recording in the interval ... - 04:26 AM Revision 18aded2e (ceph): osd/osd_types: pg_interval_t: include primaries in operator<<
- Also make up vs acting explicit.
Signed-off-by: Sage Weil <sage@inktank.com> - 04:26 AM Revision 000233f7 (ceph): osd: change in up set primary constitutes a peering interval change
- In several places, a change in the up_primary triggers a new peering
interval, but the palces that actually generate ... - 04:26 AM Revision 5562e26e (ceph): osd: use parent pgid (as appropriate) in generate_past_intervals()
- Feed in the ancestor pg_t (if any) when we are looking at intervals for
previous maps that may have preceded a recent... - 03:49 AM Revision 62301462 (ceph): Merge pull request #1651 from enovance/wip-brag
- Few bug fixes in ceph-brag
Reviewed-by: Sage Weil <sage@inktank.com> - 01:20 AM Revision 025ab9f4 (ceph): doc/release-notes: v0.80
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:48 AM rbd Bug #8178 (Resolved): 0.79: feature set mismatch, my 4a042a42 < server's 104a042a42, missing 1000...
- For some weeks I knew no troubles with RBD clients on Linux-3.13.10 x86_64.
Today after I created new erasure pool a...
04/21/2014
- 11:55 PM Bug #8113: agent_work can be continuously rescheduled during recovery while most objects are missing
- 11:53 PM Revision c80f128c (ceph): Merge pull request #1707 from ceph/wip-rbd-test
- rbd: fix tests for cache pools
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 11:47 PM Revision a80e66f9 (ceph): qa/workunit/rbd/import_export.sh: skip list-objects tests with tiering
- Listing objects isn't reliable with cache pools; skip that part of the
test if we see that rbd has tiering enabled.
... - 11:26 PM Revision 9d64ac66 (ceph): qa/workunit/rbd/copy.sh: do not delete/recreate rbd pool
- Among other things, it breaks when tiering is enabled.
Signed-off-by: Sage Weil <sage@inktank.com> - 10:43 PM Revision c3833d7c (ceph): doc: Fixed syntax to include 'pool'.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:31 PM Revision 8620bd2f (ceph): PG::PriorSet: consider lost osds in up_now for pcontdec
- Otherwise, the pg will remain down even as osds are marked lost.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:18 PM CephFS Bug #8177: Client: seg fault in verify_reply_trace on traceless reply
- 08:37 PM CephFS Bug #8177: Client: seg fault in verify_reply_trace on traceless reply
- Also /a/teuthology-2014-04-14_23:00:38-fs-master-testing-basic-plana/192241
- 08:31 PM CephFS Bug #8177 (Resolved): Client: seg fault in verify_reply_trace on traceless reply
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-17_23:01:49-fs-firefly-distro-basic-plana/199687/...
- 10:13 PM Revision f44e2c82 (ceph): Merge pull request #1703 from ceph/wip-7942
- Wip 7942
Reviewed-by: Sage Weil <sage@inktank.com> - 10:11 PM Revision 95394b60 (ceph): ReplicatedPG::do_op: check for blocked snapset obj
- Otherwise, we might use an invalid snapset in find_object_context.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:11 PM Revision 8259d874 (ceph): ReplicatedPG: in trim, grab w locks on obc and snapset_obc
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:11 PM Revision f3df5018 (ceph): ReplicatedPG: do not create whiteout clones
- First, make_writeable treats whiteout heads like snapdir for
cloning purposes. Second, to ensure that we send the co... - 10:11 PM Revision 0d5a5393 (ceph): ReplicatedPG: if we get ENOENT on clone, remove clone from snapset
- Fixes: #7916
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:11 PM Revision caa63565 (ceph): ReplicatedPG,rados: add CEPH_OSD_[COPY_FROM]_MAP_SNAP_TO_CLONE
- When promoting a clone, we want to use the provided snapid to specify
specify the clone id directly.
Signed-off-by: ... - 09:28 PM Revision bd39ecd6 (ceph): Merge pull request #1705 from ceph/wip-8124
- Wip 8124
Reviewed-by: Sage Weil <sage@inktank.com> - 09:18 PM Revision 2cb0bac6 (ceph): qa/workunits/cephtool/test.sh: make set pg_num test non-racy
- Loop while the pool is creating.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.... - 05:53 PM Revision e4a048c4 (ceph): ECMsgTypes::ECSubWrite: fix at_version indentation
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 05:53 PM Revision 7bb20115 (ceph): encoding: use unqualified name for encode/decode in boost::optional enc...
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 05:53 PM Revision 5821cc7e (ceph): osd/: propogate hit_set history with repop
- We don't actually send the whole info on each repop, just the log
entries, updated stats, and a few other bits. For ... - 05:53 PM Revision 16eccdd3 (ceph): PG,PGLog: update hit_set during peering
- Fixes: #8124
Signed-off-by: Samuel Just <sam.just@inktank.com> - 05:53 PM Revision f7e75880 (ceph): ReplicatedPG::agent_load_hit_sets: take ondisk_read_lock
- Otherwise, the hit_set might be not yet written due to a recently
completed recovery.
Signed-off-by: Samuel Just <sa... - 05:53 PM Revision 506dce84 (ceph): ReplicatedPG: do not use shard for hit_set object names
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 05:52 PM Revision ddf1e986 (ceph): osd: track the number of hit_set archive objects in a pg
- Also, use this value in agent_choose_mode instead of the max
number.
Related: #8124
Signed-off-by: Samuel Just <sam.... - 05:46 PM Revision 1fb90c94 (ceph): ReplicatedPG::hit_set_persist: clean up degraded check
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 05:21 PM Bug #8168: osd: rbd_test.test_diff_iterate fails with a cache pool
- The actual error is that the diff is nothing instead of a 512 byte extent being discarded. The error removing the ima...
- 11:24 AM Bug #8168 (Resolved): osd: rbd_test.test_diff_iterate fails with a cache pool
With a cache pool set up, test_rbd.test_diff_iterate, which removes the two snapshots it creates before removing ...- 04:55 PM Bug #8176 (Resolved): Change target_max_objects/target_max_bytes has no immediate effect
I would expect to be able to change these values and affect the balance of data in the cache/base tiers. The funct...- 04:52 PM Bug #8175 (Resolved): Some values of target_max_objects for tiering will crash OSDs
ceph osd pool set cache target_max_objects 10
It looks like a value below 1024 will cause x/1024 == 0 which will...- 03:48 PM CephFS Bug #8172 (Resolved): ceph_get_cap+0x2b/0x120
- commit b9baf44e(ceph: pre-allocate ceph_cap struct for ceph_add_cap())
- 02:44 PM CephFS Bug #8172 (Resolved): ceph_get_cap+0x2b/0x120
- ...
- 03:11 PM Revision 1448cdf5 (ceph): Work around #8166
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 02:54 PM Bug #8174 (Resolved): rados put of a long object name crashes the OSD process
rados -p testpool put
foo.0000000000000000.0000000000000000.0000000000000000.0000000000000000.0000000000000000.00...- 02:27 PM Bug #8124 (Resolved): too many hitset objects preventing full state from ending
- 02:10 PM Bug #8171 (Resolved): crypto: cryptopp has precendence over libnss
- We need libnss to have precedence over licryptopp, however, at the moment it's the other way around. If both are inst...
- 01:56 PM Revision b7394efe (ceph): multimds: bump up timeout for misc.yaml
- This keeps timing out after 3h.
Signed-off-by: Sage Weil <sage@inktank.com> - 12:41 PM rgw Bug #8170 (Resolved): rgw: missing manifest response header when reading swift user manifest object
- 12:41 PM rgw Bug #8169 (Resolved): rgw: swift user manifest does not compute etag
- etag for swift user manifest objects should contain the has of the concatenated etags for all the parts. Currently it...
- 11:02 AM Feature #7514 (Resolved): qa: add ceph_test_objectstore to rados test suite
- added by Josh Durgin on 7afc277736612eb624449f10743958da37f62f9a in the qa-suite repo
- 10:57 AM devops Bug #6726: Official packages do not appear to be available for Saucy
- Sandon Van Ness wrote:
> There was a problem with our repo generator script for release builds which was causing eve... - 10:53 AM Bug #8165 (Duplicate): mon: subscribe doesn't wait for PaxosService readable
- dup of #7997
- 09:38 AM Bug #8165 (In Progress): mon: subscribe doesn't wait for PaxosService readable
- 07:08 AM Bug #8165 (Duplicate): mon: subscribe doesn't wait for PaxosService readable
- ubuntu@teuthology:/a/teuthology-2014-04-20_19:33:18-upgrade:dumpling-x:parallel-firefly---basic-plana/205300...
- 03:30 AM Revision 476b929e (ceph): Update mkcephfs.rst
- There should be no blank between mount options.
- 01:59 AM Revision 95d0278d (ceph): ReplicatedPG::mark_all_unfound_lost: delete local copy if necessary
- There might be a local copy for an EC pool in the DELETE case. The replica
copies should be already handled by merge...
04/20/2014
- 03:06 PM Bug #8164 (Duplicate): "[ERR] 4.15 0 tried to pull" in upgrade:dumpling-x:stress-split-firefly---...
- I could not reproduced it manually.
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-19_19:55:01-u... - 02:42 PM Bug #8163 (Resolved): stuck recovering due to a 50 min delay in processing Push op
- Based on master, 7a61cdbfd533c1092fc61acb7042053251c03f7f (actual branch wip-sam-testing-safe sha1:
ef0fb611696929c... - 02:32 AM CephFS Bug #8025 (Resolved): nfs-on-kclient: rm -r failed
04/19/2014
- 10:23 PM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- Already using. :) Thanks for useful advise. Very helpful.
- 08:43 PM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- 3.15. For now, please use readdir_max_entries mount option
- 06:23 PM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- Makes sense, thank you for explaining.
- 09:34 PM Bug #8162 (Pending Backport): osd: dumpling advances last_backfill prematurely
- this is the bug where a dumpling osd advances last_backfill prematurely
- 09:03 PM Bug #8162 (In Progress): osd: dumpling advances last_backfill prematurely
- 11:55 AM Bug #8162 (Resolved): osd: dumpling advances last_backfill prematurely
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-18_20:35:03-upgrade:dumpling-x:stress-split-firefl...
- 06:50 PM Bug #8048: osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is_missing(soid))
- Please have a look at the comments of bug #8008 -- there may be some additional information related to this issue. I ...
- 06:46 PM Bug #8008: osd/ReplicatedPG.cc: 258: FAILED assert(missing_loc.needs_recovery(hoid)) during pg re...
- This may be similar to #8048 so I applied corresponding fix commit:3d0e80ac as well as number of other post-0.79 PG-r...
- 01:53 PM Bug #8008: osd/ReplicatedPG.cc: 258: FAILED assert(missing_loc.needs_recovery(hoid)) during pg re...
- Unfortunately revision 6ff645f5 applied on top of 0.79 did not fixed the issue.
I can't repair inconsistent PG.
<... - 12:58 PM Revision 61b6564b (ceph): Simple mechanical cleanups
- * Removed trailing and useless whitespaces
* Removed useless imports.
Signed-off-by: Christopher Glass <christopher.... - 09:03 AM Bug #8161 (Resolved): osd/ECBackend.cc: 475: FAILED assert(r == 0)
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-18_21:29:10-rados:thrash-testing-testing-basic-plana/2022...
- 08:15 AM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-18_21:29:10-rados:thrash-testing-testing-basic-plana/202157
- 03:04 AM devops Bug #8160 (Duplicate): multipath-tools does not co-exist with ceph
- If *multipath-tools 0.4.9-3ubuntu5* is installed on a...
- 12:33 AM Revision 7a61cdbf (ceph): buffer: adjust #include order
- The pthread.h include is somehow clobbering things, although it is not
clear how. :(
Signed-off-by: Sage Weil <sage...
04/18/2014
- 10:12 PM Revision 74f4d573 (ceph): Merge pull request #1696 from ceph/wip-8097
- buffer: use Mutex instead of Spinlock for raw crcs
Reviewed-by: Samuel Just <sam.just@inktank.com> - 09:32 PM Revision fbb90f6e (ceph): Merge pull request #26 from ceph/wip-rbd-cache
- test rbd with cache pool
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 09:24 PM Revision 0234bcfc (ceph): Merge pull request #1697 from ceph/wip-num_objects_omap
- osd_types::object_stat_sum_t: fix add/sub for num_objects_omap
Reviewed-by: Sage Weil <sage@inktank.com> - 09:09 PM Revision e087eae8 (ceph): Merge pull request #1695 from ceph/wip-8153
- Revert "ReplicatedPG::get_snapset_context: assert snap obj is not missin...
Reviewed-by: Sage Weil <sage@inktank.com> - 08:59 PM Revision f9e9365f (ceph): Revert "ReplicatedPG::get_snapset_context: assert snap obj is not missing"
- This breaks mark_lost_unfound_revert.
This reverts commit 0d2177a18071ad9c9581826a43751c36bab5b2db. - 08:57 PM Bug #7987: osd: backfill/recovery makes no progress
- /a/teuthology-2014-04-18_02:30:16-rados-master-testing-basic-plana/200757
- 08:55 PM Bug #7986: 3.1s0 scrub stat mismatch, got 2041/2044 objects, 0/0 clones, 2041/2044 dirty, 0/0
- ubuntu@teuthology:/a/teuthology-2014-04-18_02:30:16-rados-master-testing-basic-plana/200799
failure_reason: '"2014... - 08:54 PM Revision dec77c34 (ceph): Merge pull request #1693 from ceph/wip-7997
- mon: fix get_version race (more)
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com> - 08:50 PM Revision 4413670d (ceph): osd: throttle snap trimmming with simple delay
- This is not particularly smart, but it is *a* knob that lets you make
the snap trimmer slow down. It's a flow and a ... - 08:41 PM Revision 82edda23 (ceph): test: handle the create-pg delay when testing cache split syntax
- Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com> - 06:47 PM Revision d07ce841 (ceph): Merge pull request #1692 from ceph/wip-7784
- mon: OSDMonitor: HEALTH_WARN on 'mon osd down out interval == 0'
Reviewed-by: Sage Weil <sage@inktank.com> - 06:15 PM Revision b2112d50 (ceph): mon: OSDMonitor: HEALTH_WARN on 'mon osd down out interval == 0'
- A 'status' or 'health' request will return a HEALTH_WARN whenever the
monitor handling the request has the option set... - 06:12 PM Revision 09985d25 (ceph): mon: wait for PaxosService readable in handle_get_version
- We were waiting for the election to finish, but we need to *also* wait for
paxos to recover. Being a peon or leader ... - 06:10 PM Bug #8113 (In Progress): agent_work can be continuously rescheduled during recovery while most ob...
- 04:53 PM Revision d7967b42 (ceph): rbd/thrash: factor out install + ceph
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:51 PM Revision e97b8650 (ceph): rbd: do most tests with a (small) cache pool in front
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:51 PM Revision 03a84442 (ceph): rbd/basic: factor out install + ceph
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:49 PM Revision 007d9752 (ceph): Require requests >= 1.0
- 03:45 PM Bug #8156 (Rejected): Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" in upgr...
- this appears to be a simple out of memory (it's a few lines further up in the teuthology log). we need more memory o...
- 02:53 PM Bug #8156 (Rejected): Crash in Thread.cc "common/Thread.cc: 110: FAILED assert(ret == 0)" in upgr...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-17_20:35:01-upgrade:dumpling-x:stress-split-firefl...
- 03:11 PM Messengers Bug #8097 (Resolved): msgr: mon stuck in set_crc()
- 02:43 PM Fix #5844: osd: snaptrimmer should throttle itself
- This has been seen starving client I/O.
- 02:38 PM Feature #8155 (Resolved): Disallow changing cache_mode in nonsensical ways
- We currently have no limitations on when users can change cache_mode. Anybody who changes it from "writeback" to "non...
- 02:08 PM Bug #8153 (Resolved): osd/ReplicatedPG.cc: 7221: FAILED assert(attrs || (!pg_log.get_missing().is...
- 01:56 PM Bug #8153 (Resolved): osd/ReplicatedPG.cc: 7221: FAILED assert(attrs || (!pg_log.get_missing().is...
ceph version 0.79-247-gd07ce84 (d07ce84148edf0ee4a7271b9ee691815be91520e)
1: (ReplicatedPG::get_snapset_context(...- 01:59 PM Revision c623b3dd (ceph): rados/thrash: whitelist 'must scrub before tier agent can activate'
- Signed-off-by: Sage Weil <sage@inktank.com>
- 01:53 PM Bug #7997 (Resolved): handle_get_version returns old map epochs
- 09:33 AM Bug #7997 (Fix Under Review): handle_get_version returns old map epochs
- 11:47 AM Feature #7784 (Pending Backport): mon osd down out interval = 0 should prevent ceph health from r...
- 09:43 AM Feature #7784 (Fix Under Review): mon osd down out interval = 0 should prevent ceph health from r...
- Went with the simplest approach: have the leader spit out the warning if it has the option set to zero. All other mo...
- 11:30 AM devops Bug #8151 (Rejected): Perms on /etc/ceph/ceph.client.admin.keyring wrong on some nodes after install
- ...
- 09:40 AM Feature #8150 (Resolved): mon: disseminate config options throughout the mon cluster
- There are some options that will affect cluster-wide behavior.
Having all the monitors in the quorum using the lea... - 09:32 AM Bug #8133 (Duplicate): "Segmentation fault" in upgrade:dumpling-x:parallel-firefly---basic-plana ...
- dup #7997
- 06:59 AM Feature #8147 (Resolved): osd: make split automatically trigger scrub
- i think this is probably a good idea anyway, but it's more important given the cache tiering agent.
- 06:36 AM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- When -ENOMEM happens, the kclient does not properly release (cache coherence related) resources. that's why ceph-fuse...
- 05:04 AM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- I found commit in "ceph-client: https://github.com/ceph/ceph-client/commit/54008399dc0ce511a07b87f1af3d1f5c791982a4
... - 05:51 AM devops Bug #7889 (Resolved): IPv6 support with ceph-deploy
- Merged into ceph-deploy master branch with hash a3a61b7
- 04:21 AM Revision 7251983d (ceph): Merge pull request #1676 from ceph/wip-8092
- Wip 8092
Reviewed-by: Sage Weil <sage@inktank.com> - 04:19 AM Revision 375e4ee8 (ceph): Merge pull request #1678 from ceph/wip-8108
- osd: OSDMap: have osdmap json dump print valid boolean instead of string
Reviewed-by: Sage Weil <sage@inktank.com> - 04:17 AM Revision 8fb2388d (ceph): osd_types: pg_t: add get_ancestor() method
- Give us the ancestor for when the pool had a past value for pg_num.
Signed-off-by: Sage Weil <sage@inktank.com> - 01:58 AM Revision 7afc2777 (ceph): rados: include objectstore tests
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 12:53 AM Revision 2dd2b11f (ceph): Merge pull request #1683 from ceph/wip-mds-op-prio
- mds: dynamically adjust priority of committing dirfrags
Reviewed-by: Greg Farnum <greg@inktank.com>
04/17/2014
- 11:28 PM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- Sorry, that can't be right. First of all I can't find this commit. Could you please use correct commit ID? I'd like t...
- 11:13 PM CephFS Bug #8140: 0.79: MDS / CephFS: unable to read directory
- Which kernel version contains this fix?
- 05:14 PM CephFS Bug #8140 (Resolved): 0.79: MDS / CephFS: unable to read directory
- This issue should be fixed by commit 54008399 (ceph: preallocate buffer for readdir reply). For old kernel, you can a...
- 02:11 PM CephFS Bug #8140 (Resolved): 0.79: MDS / CephFS: unable to read directory
- With kernel client I got the following error when I attempted to list files in directory containing 1021 files:
<p... - 09:51 PM Revision dea70112 (ceph): Merge pull request #1689 from ceph/wip-8091
- Wip 8091
Reviewed-by: Sage Weil <sage@inktank.com> - 09:46 PM Revision 7e697b1b (ceph): ReplicatedPG::recover_replicas: do not recover clones while snap obj is...
- Otherwise, we cannot safely read the snapset for the clone.
Fixes: #8091
Signed-off-by: Samuel Just <sam.just@inktan... - 09:46 PM Bug #8143 (Resolved): BuildRoot is now silently ignored in .spec files
- http://fedoraproject.org/wiki/Packaging:Guidelines#BuildRoot_tag
I chased this for a long time until I found that ... - 09:21 PM CephFS Bug #8092 (Resolved): multimds ceph-fuse hang on write waiting for max size
- 09:18 PM Bug #8108 (Resolved): OSD json output uses strings for booleans
- 09:06 PM Revision 0e90c69f (ceph): watch_tube() belongs to the beanstalk module
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:33 PM Revision e9a1c778 (ceph): Update requests version
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:31 PM Revision 3ad51c8e (ceph): osd_types::object_stat_sum_t: fix add/sub for num_objects_omap
- Introduced in a130a4452e4fb159dc62fb417077d98dc9ebd621
Signed-off-by: Samuel Just <sam.just@inktank.com> - 08:18 PM Revision 79e7db75 (ceph): Merge pull request #1688 from ceph/wip-8048
- osd/ReplicatedPG: check clones for degraded
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:18 PM Revision ac014510 (ceph): Merge pull request #1685 from ceph/wip-8132
- mon: set leader commands prior to first election
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com> - 08:11 PM Revision 3d0e80ac (ceph): osd/ReplicatedPG: check clones for degraded
- We check whether the head is degraded, and we check whether a clone is
unreadable, but in the case where we have a ca... - 08:03 PM Revision 5dbc6426 (ceph): s/wait-for-package/wait_for_package/
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:49 PM Revision 224a0f57 (ceph): Merge pull request #1674 from ceph/wip-8086
- ReplicatedPG::agent_work: skip hitset objects before getting object cont...
Reviewed-by: Sage Weil <sage@inktank.com> - 07:40 PM Revision 5580ffb8 (ceph): Merge pull request #242 from ceph/wip-7773
- Mirror beanstalkd queue in paddles
- 05:50 PM Revision 26f4d5b0 (ceph): Merge pull request #1687 from ceph/wip-8130
- osdc/Objecter: fix osd target for newly-homeless op
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com> - 05:48 PM Revision 93c0515f (ceph): osdc/Objecter: fix osd target for newly-homeless op
- If we recalculate the mapping and find that there is no primary, we need
to set the 'osd' field to -1. Otherwise, th... - 05:35 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- 02:00 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- and to check that agent_work also does the right thing
- 02:00 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- ReplicatedPG::do_op already does the right thing as far as blocking ops which may flush. What remains is to avoid fl...
- 01:56 PM Bug #8011 (In Progress): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >...
- 05:27 PM Revision 03b8cdac (ceph): Refactor try_delete_jobs()
- Also tweak its error message
- 05:27 PM Revision 66a27422 (ceph): Add methods for querying and deleting jobs
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision 165f5d53 (ceph): When killing a run, delete paddles jobs
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision 1449e753 (ceph): Use shared methods to connect to beanstalkd
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision 8a4de411 (ceph): Rename teuthology.queue to teuthology.worker
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision 8fdea4d1 (ceph): Submit queued jobs to paddles
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision ee33192f (ceph): When deleting jobs, also delete them from paddles
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision 741c773b (ceph): Look for archive_base in config
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:27 PM Revision d12e6f4e (ceph): Be slightly less verbose about logging
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 05:22 PM Bug #8044 (Duplicate): osd/ReplicatedPG.cc: 2276: FAILED assert(p != snapset.clones.end())
- this looks like a dup of #8091
- 05:07 PM Revision fe71a12d (ceph): Merge pull request #1684 from onlyjob/debian
- spelling corrections
Reviewed-by: Sage Weil <sage@inktank.com> - 05:05 PM Revision b0338ca3 (ceph): Merge pull request #1671 from ceph/wip-7699
- mds: Fix respawn (add path resolution)
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage@inkt... - 05:03 PM Revision 3a794d5f (ceph): Merge pull request #1677 from ceph/wip-poolset-noblock
- mon: Don't block on EAGAIN from `osd pool set`
Reviewed-by: Sage Weil <sage@inktank.com> - 04:33 PM Revision 881680ee (ceph): mon: set leader commands prior to first election
- If we have just started and receive a command, we currently will reply with
EINVAL because the leader commands are em... - 04:08 PM Revision fc948794 (ceph): safe_while: Don't sleep() on the first attempt
- This was causing unnecessary delays in several places
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> - 03:55 PM RADOS Feature #8141 (New): Nice if we had a state for when a pg can't recover because all missing objec...
I put a pg into the following state by taking down 2 OSDs at just the right time after peering but before recovery ...- 02:50 PM Bug #8091 (Resolved): osd/SnapMapper.cc: 217: FAILED assert(r == -2)
- 02:43 PM Revision e3233927 (ceph): Pass -D flag to teuthology report
- Fixes an issue where tests run on old teuthology branches that died for
uncommon reasons were not being marked as dea... - 02:28 PM Revision 40e8dbbb (ceph): mon: EBUSY instead of EAGAIN when pgs creating
- In 69321bf, EAGAIN changed behaviour to block indefinitely
rather than returning to user. Change the return for
`osd... - 02:23 PM Bug #8139 (Fix Under Review): osd/osd_types.cc: 398: FAILED assert(m_seed < old_pg_num)
- 01:20 PM Bug #8139 (Resolved): osd/osd_types.cc: 398: FAILED assert(m_seed < old_pg_num)
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-15_02:30:04-rados-firefly-distro-basic-plana/193012...
- 01:54 PM Bug #8124: too many hitset objects preventing full state from ending
- 01:17 PM Bug #8048 (Resolved): osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is_missing...
- 01:11 PM Bug #8048 (Fix Under Review): osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is...
- 01:16 PM Bug #8132 (Resolved): mon: no leader commands before first election
- 09:06 AM Bug #8132: mon: no leader commands before first election
- ...
- 09:06 AM Bug #8132 (Resolved): mon: no leader commands before first election
- got EINVAL on pool create from leader mon who had just started and was starting its election:
ubuntu@teuthology:/a... - 01:10 PM Bug #8099: LibRBD.DiffIterateStress failure - extra extent in diff
- Possibly related to #8091
- 12:48 PM Bug #8086 (Resolved): FDCache::clear failed assert
- 12:13 PM Bug #8086 (Fix Under Review): FDCache::clear failed assert
- 12:12 PM Bug #8086: FDCache::clear failed assert
- 12:40 PM Tasks #7864: please clarify copyright and the license
- Also please clarify version of CC-BY-SA license used for files in /doc and /man.
Which particular version of the l... - 11:55 AM Feature #7873 (Fix Under Review): pg query: dump peer_info, peer_missing in all states
- 11:53 AM Bug #8138 (Won't Fix): Make PG repair safe by requiring force flag to repair an ambiguous situation
be_select_auth_object() should have a force flag and not arbitrarily use the first shard as the authoritative objec...- 11:19 AM Bug #8103: pool has too few PGs warning misleading when using cache pools
- Mark Nelson wrote:
> It seems like there may be other situations where this is misleading too. Say if you have many ... - 10:53 AM Bug #8130 (Resolved): Objecter: resending Ops to wrong target
- 10:47 AM Bug #8130 (Fix Under Review): Objecter: resending Ops to wrong target
- 09:37 AM Bug #8130 (In Progress): Objecter: resending Ops to wrong target
- this is affecting master now too:
teuthology-2014-04-16_02:30:03-rados-master-testing-basic-plana has many failures - 10:12 AM Bug #8133: "Segmentation fault" in upgrade:dumpling-x:parallel-firefly---basic-plana suite
- FYI - Manual re-run did not produce errors.
- 09:27 AM Bug #8133 (Duplicate): "Segmentation fault" in upgrade:dumpling-x:parallel-firefly---basic-plana ...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-16_19:33:25-upgrade:dumpling-x:parallel-firefly---...
- 10:05 AM CephFS Bug #7966 (Resolved): ceph-mds respawn doesn't always work
- 10:00 AM Bug #8043 (Resolved): until we fix it more better, we should disallow split on cache pools
- 09:04 AM Feature #7784 (In Progress): mon osd down out interval = 0 should prevent ceph health from report...
- 08:40 AM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- ubuntu@teuthology:/a/teuthology-2014-04-15_22:35:26-upgrade:dumpling-x:stress-split-firefly-distro-basic-vps/196331
- 04:57 AM Revision 2e375b6f (ceph): Merge pull request #1675 from guangyy/wip-bench
- Make rados/rest bench work for multiple write instances without metadata conflict.
Reviewed-by: Greg Farnum <greg@in... - 02:43 AM Revision f22e2e9a (ceph): spelling corrections
- 01:16 AM Revision 75a5bd5d (ceph): Merge pull request #1681 from ceph/wip-8043
- mon/OSDMonitor: require force argument to split a cache pool
Reviewed-by: Samuel Just <sam.just@inktank.com> - 01:13 AM Revision 6d58e3c9 (ceph): Merge pull request #1682 from ceph/wip-8020
- OSD: split pg stats during pg split
Reviewed-by: Sage Weil <sage@inktank.com> - 01:10 AM Revision 18caa1cd (ceph): OSD: split pg stats during pg split
- Fixes: #8020
Signed-off-by: Samuel Just <sam.just@inktank.com> - 01:08 AM Revision 5e4a5dc6 (ceph): osd_types::osd_stat_sum_t: fix floor for num_objects_omap
- Introduced in a130a4452e4fb159dc62fb417077d98dc9ebd621
Signed-off-by: Samuel Just <sam.just@inktank.com>
04/16/2014
- 10:09 PM Revision a3d759eb (ceph): Merge branch 'wip-8100'
- Reviewed-by: Mark Nelson <mark.nelson@inktank.com>
- 10:06 PM Revision a3d452ac (ceph): common/obj_bencher: Fix error return check from read that is negative o...
- Fixed read return value in d99f1d9f68db41231e0ffff4082b05d6d095c231
Fixes: #8100
Signed-off-by: David Zafman <david... - 09:45 PM Bug #8130 (Resolved): Objecter: resending Ops to wrong target
- From teuthology:/a/gregf-2014-04-16_12:06:55-rados:thrash-wip-fast-dispatch-testing-basic-plana
Note how it marks_... - 09:35 PM Bug #8048 (In Progress): osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is_miss...
- 08:21 PM Revision 4b9202bc (ceph): Update to use psutil 2.x API
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:10 PM Feature #7873 (In Progress): pg query: dump peer_info, peer_missing in all states
- peer_info is already output in the "query" in ReplicatedPG::do_command()
I've added the peer_missing output right af... - 06:49 PM Revision 24da7d0c (ceph): Merge pull request #1680 from ceph/wip-7786
- civetweb: update subproject
- 06:29 PM Bug #8113: agent_work can be continuously rescheduled during recovery while most objects are missing
- this is probably just a matter of subtracting num_missing from num_flushable?
- 06:12 PM Bug #8020 (Resolved): evenly split stats on split
- 06:08 PM Revision 4db1984c (ceph): osd/ReplicatedPG: add missing whitespace in debug output
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 05:25 PM Bug #7891: osd: leaked pg refs on shutdown
- ubuntu@teuthology:/a/teuthology-2014-04-15_02:30:04-rados-firefly-distro-basic-plana/193023 but not debug patch applied
- 03:10 PM Bug #8100 (Resolved): Rados Bench seq read errors on tiered configuration
- a3d452acdf2fcf9ad10002c5f24c2548d12952bd
- 02:14 PM Bug #8100: Rados Bench seq read errors on tiered configuration
- 01:52 PM Bug #8100: Rados Bench seq read errors on tiered configuration
- Through some bisecting and a well-informed guess by Yehuda, it appears that this is being caused by d99f1d9f.
- 02:45 PM Revision 089dda15 (ceph): Optionally use civetweb instead of apache
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 02:45 PM Revision 761d7693 (ceph): Don't run apache functions if not using apache
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 02:45 PM Revision 8b93c03f (ceph): Generate subtasks instead of copy/pasting them
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 01:06 PM Bug #8124 (Resolved): too many hitset objects preventing full state from ending
- Unclear how, but the pg appears to have around 17 hitset objects which ends up causing the pg to be unable to evict e...
- 12:32 PM Bug #8123 (Can't reproduce): OSD: received operation against clone which was not backfilled (but ...
- Let's start by saying I saw this on my wip-fast-dispatch branch, so it *could* be a bug there rather than in master. ...
- 11:49 AM rgw Bug #7786 (Resolved): civetweb segfaults with file uploads larger than 2GB
- 11:17 AM Linux kernel client Bug #8122: bogus mount error "mount: error writing /etc/mtab: Invalid argumentnothing was mounted"
- This is on 0.79.
- 11:16 AM Linux kernel client Bug #8122 (Duplicate): bogus mount error "mount: error writing /etc/mtab: Invalid argumentnothing...
- mount.ceph print the following error every time I mount Ceph file system:...
- 10:48 AM Bug #8121 (Resolved): ReplicatedBackend::build_push_op() should handle a short read or assert
I noticed that the existing code in build_push_op() may not handle all scenarios of fixing data_included interval w...- 10:36 AM Bug #7892 (Duplicate): osd/ReplicatedPG.cc: 7881: FAILED assert((data_included.empty() && data.le...
There were 2 identical crashes. This is the trace of one of them:
object: cecc4d22/plana9117053-25/8d//3
pg: 3...- 10:15 AM Bug #8103: pool has too few PGs warning misleading when using cache pools
- It seems like there may be other situations where this is misleading too. Say if you have many mostly empty pools and...
- 10:11 AM Bug #8036: levedb: throws std::bad_allow on 14.04
- Have been spending a fair amount of time trying to figure out what may have gone wrong with this (and #8067, which ap...
- 09:45 AM devops Feature #8120 (Resolved): RHEL7 GA kernel build
- The new RC is out so we need to rebuild the kernel packages again...
- 09:43 AM Feature #7784: mon osd down out interval = 0 should prevent ceph health from reporting ok
- Mapping a config option to a map flag is not an intuitive thing to do or to expect. What if the user injects a diffe...
- 07:11 AM Revision 924064f8 (ceph): mds: dynamically adjust priority of committing dirfrags
- Adjust priority of committing dirfrags according to number of
expiring log segments. The more expiring log segments, ... - 05:51 AM Revision 0640a085 (ceph): mds: fix cap revoke confirmation
- when the _revokes list is emptied, it doesn't mean that client has
released the revoking caps. It's possible that cli... - 01:28 AM Revision 8c7a5ab8 (ceph): Use string instead of char* when saving arguments for rest-bench
- 01:21 AM CephFS Bug #8118 (Closed): MDS crashes
- Active MDS crashes (v0.79).
log file attached.
Host did not ran out of memory, Standby MDS took over successfully....
04/15/2014
- 10:37 PM Revision 0d2177a1 (ceph): ReplicatedPG::get_snapset_context: assert snap obj is not missing
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 08:57 PM Revision 015df934 (ceph): mon/OSDMonitor: require force argument to split a cache pool
- There are several perils when splitting a cache pool:
- split invalidstes pg stats, which disables the agent
- a s... - 07:28 PM Revision 823219bb (ceph): Don't pass apache's config to radosgw
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 06:53 PM devops Feature #7716 (In Progress): Build debug packages for EL6
- I believe this was caused by redhat-rpm-config not being installed on the centos gitbuilder used for release builds (...
- 06:12 PM Revision 12af2abb (ceph): Rename some functions and variables
- This is to make the refactoring a little smoother and easier to read.
Signed-off-by: Zack Cerza <zack.cerza@inktank.... - 05:32 PM Revision c2523458 (ceph): osd: OSDMap: have osdmap json dump print valid boolean instead of string
- Fixes: 8108
Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com> - 03:58 PM Revision f82f6637 (ceph): Fix all but one of the PEP-8 issues
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 02:58 PM Bug #8114 (Can't reproduce): "osd/RadosModel.h: 1055: FAILED assert" in upgrade:dumpling-x:stress...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-13_22:35:20-upgrade:dumpling-x:stress-split-firefl...
- 02:52 PM Bug #8103: pool has too few PGs warning misleading when using cache pools
- Given that this is a transient issue for a new, empty cluster, I'm not sure if it is worth making an exception for th...
- 02:20 PM Bug #8091: osd/SnapMapper.cc: 217: FAILED assert(r == -2)
- recover_replicas can cause us to read the snapset from an obsolete snapdir or head object. recover_replicas should n...
- 02:08 PM devops Tasks #7678 (Resolved): f20 Jenkins slave
- 02:07 PM devops Feature #6020 (Fix Under Review): radosgw-apache opinionated package
- 01:56 PM Bug #8080 (Resolved): objecter: linger ops don't pay attention to cache overlay
- 01:56 PM Bug #8043 (Fix Under Review): until we fix it more better, we should disallow split on cache pools
- 01:46 PM Bug #8098 (Can't reproduce): ceph v0.79-125 : Random osd's are flapping too frequently : OSD wron...
- Thanks for the follow-up. Please let us know if you can figure out how to reproduce the problem, or can gather more ...
- 02:24 AM Bug #8098: ceph v0.79-125 : Random osd's are flapping too frequently : OSD wrongly marked me down
- Thanks for your interest Sage , today morning the cluster looks normal and healthy. Unfortunately i do not have syste...
- 01:10 PM devops Fix #8109: OSD-disk fails to activate when final mount dir is not empty and shows no proper error...
- the other way to solve that is just to change
elif os.listdir('/var/lib/ceph/osd/{cluster}-{osd_id}'.format(
clu... - 06:59 AM devops Fix #8109 (Closed): OSD-disk fails to activate when final mount dir is not empty and shows no pro...
- OSD-disk fails to activate when STATEDIR + '/osd/{cluster}-{osd_id}'.format(cluster=cluster,osd_id=osd_id) is not emp...
- 01:03 PM Feature #7547: Basic docs for Cache Tiering functionality
- 01:01 PM Feature #7940 (Resolved): add pool snaps to ceph_test_rados
- 01:00 PM Feature #7831 (Resolved): OSD: track objects with omap entries and don't count toward caps
- 12:58 PM Feature #8041 (Resolved): ceph uses GCC-specific strerror_r; easy to make more portable
- d0a7632a31258d0963dc5d4cf7502905cc8abfe7, merged into master at ab4a35f75eac161d2509e9e34853942fcc4ed6bb
- 12:58 PM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- 12:52 PM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- It's because we reset the tid on redirected ops and the op_cancel in ReplicatedPG therefore fails to work.
- 11:43 AM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- ...I just hit this on 3 osds at the same time
- 10:56 AM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- 10:50 AM Bug #7588: OSD Seg fault in string assign ObjectOperation::C_ObjectOperation_copyget::finish()
- Saw this pop up once on my fast dispatch branch:
/a/gregf-2014-04-14_16:40:42-rados:thrash-wip-fast-dispatch-testing... - 12:44 PM Bug #8100: Rados Bench seq read errors on tiered configuration
- It's all automated, though I did try manually testing reads from the command line as well. FWIW, with debugging enab...
- 10:10 AM Bug #8100: Rados Bench seq read errors on tiered configuration
- Did you check for typos? :p Right pool name? That "-3" looks easy to get wrong.
- 10:08 AM Bug #8100: Rados Bench seq read errors on tiered configuration
- This appears to be happening on non-tiered pools as well, regardless if erasure coding or replication is used.
- 11:33 AM Bug #8113 (Resolved): agent_work can be continuously rescheduled during recovery while most objec...
- We probably need to detect when we've gone through the entire hash space without starting anything.
- 11:28 AM Revision aa6df59e (ceph): mds: Fix respawn (add path resolution)
- Previously assumed that ceph-mds executable was in
PWD - now use /proc/self/exe to find the
executable whereever it m... - 11:19 AM rgw Documentation #8112 (Resolved): radosgw usage and manpage need updating
- While working on #7933, I noticed a large gap in what flags our tests are passing to radosgw, and what flags are docu...
- 10:54 AM devops Bug #5193 (Resolved): RHEL6 does not ship with xfsprogs
- 10:54 AM Bug #8077 (Resolved): osd/ReplicatedPG.cc: 10862: FAILED assert(r >= 0) (agent_load_hit_sets)
- 10:54 AM Bug #8063 (Resolved): LibRadosTwoPoolsECPP.PromoteSnap got EAGAIN
- 10:54 AM Bug #8081 (Resolved): hitset-get on missing object fails
- 10:53 AM Bug #8085 (Resolved): osd/PG.cc: 2218: FAILED assert(!actingbackfill.empty()) from do_command
- 10:53 AM Bug #7997 (Resolved): handle_get_version returns old map epochs
- 10:52 AM Bug #8008 (Resolved): osd/ReplicatedPG.cc: 258: FAILED assert(missing_loc.needs_recovery(hoid)) d...
- 10:51 AM Bug #8089 (Resolved): osd: ENOENT on cache-evict
- 10:51 AM Bug #8086 (Resolved): FDCache::clear failed assert
- 10:34 AM Bug #8108: OSD json output uses strings for booleans
- ...
- 10:32 AM Bug #8108: OSD json output uses strings for booleans
- pull request: https://github.com/ceph/ceph/pull/1678
- 09:54 AM Bug #8108 (Fix Under Review): OSD json output uses strings for booleans
- wip-8108
- 06:56 AM Bug #8108 (Resolved): OSD json output uses strings for booleans
- The output for osd stat using JSON format uses strings for booleans:...
- 09:30 AM rgw Bug #8111 (Resolved): /etc/init.d/ceph-radosgw for RHEL needs QA
- During testing of the insallation procedure for RHEL, I found it possible to start ceph-radosgw in user space, but I ...
- 09:11 AM rgw Tasks #8110 (Resolved): rgw: diagram for rgw notifications (zone object sync)
- 09:08 AM rgw Documentation #7434 (Fix Under Review): rgw: doc user/group quota
- 08:30 AM Revision f6db1bc2 (ceph): mds: share max size to client who is allowed for WR cap
- WR cap is allowed for the loner client when filelock is in excl->mix
state. MDS should share max size with the loner ... - 08:02 AM Revision 358bde5d (ceph): Add clone test on store_test
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 07:48 AM Revision 308758b7 (ceph): Make rados/rest bench work for multiple write instances without metadat...
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 05:50 AM Bug #7159: ceph status --format=json num_in_osds and num_up_osds formatting not consistent
- fixed by 790dda9c
do we need to backport it? - 01:43 AM Bug #7710: Multiple rados bench instance will overwrite the metadata object
- A new patch was submitted, please help to review - https://github.com/ceph/ceph/pull/1675
- 01:30 AM CephFS Bug #8092: multimds ceph-fuse hang on write waiting for max size
- 12:21 AM Revision 4c015136 (ceph): Improve unlock error messages.
- Added messages if the hostname is invalid, and if
the user is not the owner of the lock.
Fixes: 6295
Signed-off-by: ... - 12:13 AM Revision 908fa5ed (ceph): Merge pull request #1666 from ceph/wip-mds
- Wip mds
04/14/2014
- 11:55 PM Revision edda6f74 (ceph): ReplicatedPG::_rollback_to: fix comment, clone certainly could be missing
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 11:55 PM Revision d39e0031 (ceph): ReplicatedPG::cancel_flush: requeue dup_ops even if !op
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 11:55 PM Revision 3f7861e2 (ceph): ReplicatedPG::agent_work: skip if head is missing
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 11:55 PM Revision 43b7c3a2 (ceph): ReplicatedPG::finish_promote: soid.clone may have been trimed, fix assert
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 11:55 PM Revision 9b7fa388 (ceph): ReplicatedPG::process_copy_chunk: don't check snaps if we got head
- Even if we are promoting a clone, we may be reading from head.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 11:12 PM Revision 9f6f7d35 (ceph): Merge pull request #1673 from ceph/wip-stress-watch
- ceph_test_stress_watch: test over cache pool
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Josh Durgi... - 11:11 PM Revision 895b6d4d (ceph): Merge pull request #1667 from ceph/wip-8089
- osd: fix dup request ahndling for ENOENT and cache ops
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:10 PM Revision 898ee489 (ceph): Merge pull request #1654 from ceph/wip-7940
- Wip 7940
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:09 PM Revision cab29ac1 (ceph): Merge pull request #1664 from ceph/wip-8085
- osd: handle misdirected pg command
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:07 PM Revision a2323a61 (ceph): Merge pull request #1660 from ceph/wip-hitset-missing
- osd: handle hitset-get on a missing hit_set object
Reviewed-by: Samuel Just <sam.just@inktank.com> - 11:07 PM Revision eef2bf67 (ceph): Fix kvm issues for Trusty
- Change kvm reference to qemu-system-x86_64 and use raw format
in kvm/qemu command. Tested on both Trusty and Precise... - 10:57 PM Revision 37ed4b60 (ceph): ceph_test_stress_watch: test over cache pool
- Signed-off-by: Sage Weil <sage@inktank.com>
- 10:36 PM Revision 4388d876 (ceph): Merge pull request #1661 from ceph/wip-objecter
- objecter: make linger watch the correct pool/object
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
Reviewed-by: Josh... - 08:57 PM Revision ab4a35f7 (ceph): Merge pull request #1672 from ceph/wip-strerror
- Use cpp_strerror() wherever possible, and use autoconf for portability
Reviewed-by: Josh Durgin <josh.durgin@inktank... - 08:07 PM Revision d0a7632a (ceph): Use cpp_strerror() wherever possible, and use autoconf for portability
- strerror_r is not portable; on Gnu libc it returns char * and sometimes
does not fill in the supplied buffer. Use au... - 08:04 PM devops Bug #8095: [chef] No version specified, and no candidate version available for libmpich2-3
- Sandon found the reason this afternoon:
(05:05:46 PM) houkouonchi-home: dmick: https://launchpad.net/ubuntu/+sour... - 07:06 AM devops Bug #8095 (Resolved): [chef] No version specified, and no candidate version available for libmpic...
- Some issues installing this package are making ceph-deploy tests fail....
- 06:44 PM Revision 29d83fef (ceph): Merge pull request #1668 from ceph/wip-librados-tests
- ceph_test_rados_api_*: fix build warnings and memset ranges
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 06:33 PM CephFS Bug #8092 (Need More Info): multimds ceph-fuse hang on write waiting for max size
- 06:08 PM Revision 502cc614 (ceph): ReplicatedPG::agent_work: skip hitset objects before getting object con...
- Otherwise, we might read the attr on a hitset object we are in the
process of deleting.
Fixes: #8086
Signed-off-by: ... - 05:55 PM Revision ba74f907 (ceph): Merge pull request #25 from ceph/wip-fastdumpemp-wusui
- Reduce runs in the dumpling-emperor upgrade suite.
- 05:07 PM Revision f6d1d922 (ceph): pipe: only read AuthSessionHandler under pipe_lock
- session_security, the AuthSessionHandler for a Pipe, is deleted and
recreated while the pipe_lock is held. read_messa... - 05:07 PM Revision 27d6032e (ceph): auth: add rwlock to AuthClientHandler to prevent races
- For cephx, build_authorizer reads a bunch of state (especially the
current session_key) which can be updated by the M... - 05:06 PM Bug #8104 (Resolved): OSD: changing min_size to larger than the acting set does not make the PG g...
- It looks like maybe we can only change into the Inactive state when going through peering, and changing the min_size ...
- 05:06 PM Bug #8103 (Won't Fix): pool has too few PGs warning misleading when using cache pools
- When using cache pools on a fresh filesystem, quickly the number of objects in the cache pool can greatly exceed the ...
- 05:04 PM Revision 25c32482 (ceph): osdc/ObjectCacher: back off less during flush
- In cce990efc8f2a58c8d0fa11c234ddf2242b1b856 we added a limit to avoid
holding the lock for too long. However, if we ... - 04:54 PM Revision 2b4b00b7 (ceph): pipe: only read AuthSessionHandler under pipe_lock
- session_security, the AuthSessionHandler for a Pipe, is deleted and
recreated while the pipe_lock is held. read_messa... - 04:54 PM Revision c049967a (ceph): auth: add rwlock to AuthClientHandler to prevent races
- For cephx, build_authorizer reads a bunch of state (especially the
current session_key) which can be updated by the M... - 04:27 PM Revision 0d4aed88 (ceph): mon: fix Session ref leak in handle_get_version
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:20 PM CephFS Bug #8025: nfs-on-kclient: rm -r failed
- yes, commit 85f6def97b75b840d6be97671cb8bacd2a074a24 should fix this
- 01:37 PM CephFS Bug #8025: nfs-on-kclient: rm -r failed
- This turned up again on yesterday's run: http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-13_23:01:11-knfs-mast...
- 04:19 PM Revision d5634349 (ceph): doc: Distro specific rgw.conf example.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:19 PM Revision bd22cac9 (ceph): doc: Distro specific rgw.conf example.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:18 PM Revision 75606e98 (ceph): doc: Created standalone doc for Keystone Integration.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:18 PM Revision 6650c0e8 (ceph): doc: Added new docs to index.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:18 PM Revision 1310af23 (ceph): doc: Reworked the simple configuration guide to be more generic.
- Changes include removing keystone and putting it into a separate document,
removing user config and putting it into a... - 04:16 PM Revision 6853d21a (ceph): doc: New admin guide for Ceph Object Gateway. Needs some clarification ...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:15 PM Revision e02b8458 (ceph): doc: Admin API usage for quotas. Needs additional clarification on syntax.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 03:56 PM Revision d0d162a5 (ceph): buffer: use Mutex instead of Spinlock for raw crcs
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:48 PM devops Bug #8017 (Resolved): Redhat Dependencies Unmet
- All RHEL dependencies in jenkins and built.
- 03:47 PM devops Bug #5193: RHEL6 does not ship with xfsprogs
- And this can now be built in jenkins. Closing it out.
- 03:45 PM Bug #8098: ceph v0.79-125 : Random osd's are flapping too frequently : OSD wrongly marked me down
- This is usually caused by low memory leading to swapping. Can you verify the CPU and memory are not oversubscribed?
- 09:23 AM Bug #8098: ceph v0.79-125 : Random osd's are flapping too frequently : OSD wrongly marked me down
- I have been searching on internet and found there has been similar kind of bugs came up in past as well and fixes wer...
- 09:19 AM Bug #8098 (Can't reproduce): ceph v0.79-125 : Random osd's are flapping too frequently : OSD wron...
- Hello Developers
I have been facing weird problem with my ceph cluster
Problem : Randomly OSDs are flapping , ... - 02:48 PM Bug #8086: FDCache::clear failed assert
- wip-8086 appears to have solved this. Thanks Sam!
- 11:08 AM Bug #8086: FDCache::clear failed assert
- 09:27 AM Bug #8086: FDCache::clear failed assert
- On further review, this seems to be happening when an erasure coded base pool (that has an associated writeback cache...
- 02:17 PM Bug #8102 (Duplicate): OSD json output mixes ints with strings
- Oh, this was already input (and fixed) as #7159
- 01:56 PM Bug #8102 (Duplicate): OSD json output mixes ints with strings
- This is unexpected output from checking the OSD status in JSON format:...
- 02:14 PM Linux kernel client Feature #190 (Fix Under Review): krbd: DISCARD support
- 01:50 PM RADOS Feature #8101 (New): Quickly delete base pool and cache pool without removing overlay
- Right now it's rather cumbersome to delete pools that have a cache tier. Simply removing them doesn't work (as we in...
- 01:41 PM Bug #8100 (Resolved): Rados Bench seq read errors on tiered configuration
- Attempting to perform the following rados bench seq test on a tiered pool setup that had data written to it via rados...
- 11:39 AM rgw Bug #7450 (Resolved): "radosgw-admin key create" ignores specified access key when subuser specified
- commit:c09f58ef81db9f6dbd528b2aa4c2f135aa6d262e
- 11:28 AM rbd Bug #7165 (Resolved): high ping latency in vm using rbd writeback cache in qemu 1.7 with async flush
- From the logs it was caused by the ObjectCacher's flusher thread holding the lock too long. Backported the fixes for ...
- 11:25 AM rbd Bug #6480 (Resolved): librbd crashed qemu-system-x86_64
- Thanks for the update Mike! Backported the fixes to dumpling (commit:2b4b00b76b245b1ac6f95e4537b1d1a4656715d5 commit:...
- 09:19 AM rbd Bug #6480: librbd crashed qemu-system-x86_64
- I've run wip-6480-0.67.7 across our infrastructure for 10 days now without any qemu crashes. Thanks Josh!
- 10:57 AM Bug #8099 (Duplicate): LibRBD.DiffIterateStress failure - extra extent in diff
- rbd's diff iterate stress test failed: http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-13_23:00:01-rbd-master-...
- 10:36 AM rbd Bug #7310 (Can't reproduce): rbd snap unprotect test failing
- This hasn't occurred again afaict.
- 09:52 AM Revision 64cd332e (ceph): Merge pull request #1622 from dachary/wip-mailmap
- mailmap updates
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail.com> - 09:39 AM CephFS Bug #8004 (Resolved): LibCephFS.HardlinkNoOriginal hang
- 09:31 AM Revision 7c17fc4a (ceph): mds: don't modify inode when calculating client ranges
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 09:27 AM Bug #7997 (Fix Under Review): handle_get_version returns old map epochs
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-14_02:30:16-rados-master-testing-basic-plana/190255
- 08:39 AM Revision 3dd88006 (ceph): Merge pull request #1669 from ceph/wip-client-debug
- client: print inode max_size
- 08:38 AM Revision 65ec24e3 (ceph): client: print inode max_size
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 08:22 AM Messengers Bug #8097 (Resolved): msgr: mon stuck in set_crc()
- ...
- 07:54 AM Bug #8069: ceph-deploy README suggests questionable SSH parameters
- Thank you!
- 07:10 AM Bug #8069 (Resolved): ceph-deploy README suggests questionable SSH parameters
- Updated and pushed to master ceph-deploy c5fc7ad438cf276f4fb566eb4572b70057e7bb52
- 07:18 AM CephFS Fix #8094: MDS: be accurate about stats in check_rstats
- 07:03 AM CephFS Fix #8094 (Resolved): MDS: be accurate about stats in check_rstats
- We see occasional crashes in CDir::check_rstats function, on...
- 07:04 AM CephFS Bug #8090: multimds: mds crash in check_rstats
- I made a fix ticket for that: http://tracker.ceph.com/issues/8094
- 06:54 AM CephFS Bug #8090: multimds: mds crash in check_rstats
- it's CDir::check_rstats() bug instead of rstats corruption
- 04:59 AM Revision d6c71b76 (ceph): osd/ReplicatedPG: add missing whitespace in debug output
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:37 AM Revision 171d5c50 (ceph): ceph_test_rados_api_*: fix build warnings, memset ranges
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:33 AM Revision 8905e3e2 (ceph): osd/ReplicatedPG: handle dup ops earlier in do_op
- Current the dup op checks happen in execute_ctx, long after we handle
cache ops or get the obc and (potentially) retu...
04/13/2014
- 10:12 PM CephFS Bug #8092 (Resolved): multimds ceph-fuse hang on write waiting for max size
- ...
- 10:10 PM Bug #8091 (Resolved): osd/SnapMapper.cc: 217: FAILED assert(r == -2)
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-13_09:43:35-rados:thrash-testing-testing-basic-plana/1892...
- 10:08 PM Bug #8089 (Fix Under Review): osd: ENOENT on cache-evict
- 03:32 PM Bug #8089 (Resolved): osd: ENOENT on cache-evict
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-13_09:43:35-rados:thrash-testing-testing-basic-plana/1891...
- 09:47 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-13_09:43:35-rados:thrash-testing-testing-basic-plana/189348
- 09:42 PM Bug #8011: osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || soid >= scrubber.end)
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-13_09:43:35-rados:thrash-testing-testing-basic-plana/189166
- 09:45 PM CephFS Bug #8090 (Duplicate): multimds: mds crash in check_rstats
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-11_23:01:29-multimds-master-testing-basic-plana/187...
- 08:11 PM Feature #8088 (Closed): Make FileStore::_do_copy_range use fiemap
- 02:20 AM Feature #8088: Make FileStore::_do_copy_range use fiemap
- I noticed discuss on http://www.spinics.net/lists/ceph-devel/msg06858.html and http://tracker.ceph.com/issues/2535.
... - 12:37 AM Feature #8088: Make FileStore::_do_copy_range use fiemap
- It can also reduce the consuming time of "_set_replay_guard", because "_set_replay_guard" has two fsync which are aff...
- 02:11 PM Bug #7892: osd/ReplicatedPG.cc: 7881: FAILED assert((data_included.empty() && data.length() == 0)...
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-13_09:43:35-rados:thrash-testing-testing-basic-plana/189080
- 12:38 PM Revision 26659a5a (ceph): mds: don't issue/revoke caps before client has caps
- If early reply is not allowed, MDS does not send reply to client immediately
after Locker::issue_new_caps adds new ca... - 12:33 PM Revision bd8aa6f4 (ceph): mds: do file recover after authpin inode
- MDCache::do_file_recover may call Locker::evel_gather, which may change
filelock to stable state. So we should authpi... - 05:23 AM Revision a72bcddf (ceph): osd/ReplicatedPG: handle misdirected do_command
- We can get a query on a pg we still have but are no longer primary for. If
that happens, do not reply. The client w...
04/12/2014
- 11:07 PM Feature #8088: Make FileStore::_do_copy_range use fiemap
- https://github.com/ceph/ceph/pull/1665
- 10:43 PM Feature #8088 (Closed): Make FileStore::_do_copy_range use fiemap
- In rbd usage, a large volume which has lots of objects will be in a low performance during a long time after creating...
- 10:29 PM Linux kernel client Bug #8087 (Resolved): libceph: null deref in osd_reset -> __reset_osd -> __remove_osd
- I thought there was an open bug for this but I couldn't find it....
- 01:40 PM rgw Bug #7799: Errors in upgrade:dumpling-x:stress-split-firefly---basic-plana suite
- I don't see those errors in latest runs.
- 12:21 PM Bug #8086: FDCache::clear failed assert
- you'll probably have to reproduce with debug filestore = 20 debug osd = 20
- 12:15 PM Bug #8086: FDCache::clear failed assert
- Note; of the 6 OSDs in the cache tier, 5 appeared to fail with a similar stacktrace.
- 12:10 PM Bug #8086 (Resolved): FDCache::clear failed assert
- Hit this today during tiering performance testing with EC backend. OSD is on an SSD that is part of the cache tier a...
- 10:51 AM Bug #8085 (Resolved): osd/PG.cc: 2218: FAILED assert(!actingbackfill.empty()) from do_command
- ...
- 06:14 AM Revision 4ccc845c (ceph): mds: fix typo in Server::do_rename_rollback()
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 05:33 AM Revision 93b933fb (ceph): Merge pull request #1659 from ceph/wip-8054
- mds: finish table servers recovery after creating newfs
Reviewed-by: Sage Weil <sage@inktank.com> - 02:06 AM Bug #8083 (Need More Info): erasure-code: fix static code analysis errors found in gf-complete
- 01:47 AM Bug #8083: erasure-code: fix static code analysis errors found in gf-complete
- "static overrun":https://github.com/ceph/gf-complete/blob/e1dcbfaa9bac94fd48e2c3f2fb8e133720e57359/src/gf_w8.c#L1267
... - 01:44 AM Bug #8083: erasure-code: fix static code analysis errors found in gf-complete
- The reason for "the deadcode":https://github.com/ceph/gf-complete/blob/e1dcbfaa9bac94fd48e2c3f2fb8e133720e57359/src/g...
- 01:20 AM Bug #8083 (Resolved): erasure-code: fix static code analysis errors found in gf-complete
- ...
- 01:15 AM Bug #7914 (Resolved): osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length f...
- The workload ran 30 more times without errors. The loop was interrupted and the plana machines released.
- 12:46 AM Revision 809d0fab (ceph): osd/ReplicatedPG: handle missing hit_set on HITSET_GET rados op
- Fixes: #8081
Signed-off-by: Sage Weil <sage@inktank.com> - 12:33 AM Revision d72602f6 (ceph): Merge pull request #1655 from ceph/wip-8077
- osd: handle missing hti_set objects in agent_load_hit_sets
Reviewed-by: Samuel Just <sam.just@inktank.com> - 12:11 AM Revision b31107b7 (ceph): test/librados/TestCase: add Param option that can set up a cache pool
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:11 AM Revision 53a4b719 (ceph): ceph_test_rados_api_watch_notify: test over cache pool
- Signed-off-by: Sage Weil <sage@inktank.com>
04/11/2014
- 11:22 PM Revision 6e0496b9 (ceph): librados: Rollback interface additions
- Add C interface rados_ioctx_snap_rollback() and indicate that rados_rollback()
is deprecated.
Add C++ interface IoCt... - 11:22 PM Revision b6b0c3bd (ceph): librados: Add ObjectWriteOperation::snap_rollback() for pool snapshots
- snap_rollback() is the same as selfmanaged_snap_rollback() but we want an
independent interface for pool snapshots. ... - 11:22 PM Revision 09b5538c (ceph): test: Fix ceph_test_rados to not core dump with invalid arguments
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 11:22 PM Revision 3ce40780 (ceph): test: Add --pool-snaps option to ceph_test_rados
- Fixes: #7940
Signed-off-by: David Zafman <david.zafman@inktank.com> - 10:50 PM Revision 3853cc90 (ceph): Merge pull request #1658 from ceph/wip-8008
- osd: fix repair_object
Reviewed-by: Samuel Just <sam.just@inktank.com> - 10:50 PM Revision 2e40afcd (ceph): Merge pull request #1657 from ceph/wip-8063
- ceph_test_rados_api_tier: fix scrub test
Reviewed-by: Samuel Just <sam.just@inktank.com> - 10:39 PM Revision 6ff645f5 (ceph): osd/PG: fix repair_object when missing on primary
- If the object is missing on the primary, we need to fully populate the
missing_loc.needs_recovery_map. This broke wi... - 10:32 PM CephFS Bug #8054 (Resolved): multimds hang on fsstress
- 10:34 AM CephFS Bug #8054 (In Progress): multimds hang on fsstress
- reverted the commit.. it breaks mds startup in the trivial single-mds case
- 06:55 AM CephFS Bug #8054 (Resolved): multimds hang on fsstress
- 10:12 PM Bug #8044: osd/ReplicatedPG.cc: 2276: FAILED assert(p != snapset.clones.end())
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-11_15:47:29-rados:thrash-testing-testing-basic-plana/185859
- 05:57 PM Bug #8044: osd/ReplicatedPG.cc: 2276: FAILED assert(p != snapset.clones.end())
- 09:48 PM Revision 19acfebc (ceph): ceph_test_librados_tier: tolerage EAGAIN from pg scrub command
- We may get EAGAIN if the osd happens to be down, for example due to
thrashing. Try a few times and then give up.
No... - 09:44 PM Revision b1db0751 (ceph): Merge pull request #1656 from ceph/wip-osd-boot
- mon: fix osd boot check
Reviewed-by: Greg Farnum <greg@inktank.com> - 09:32 PM Revision 39b9d9d8 (ceph): mon/OSDMonitor: fix osd epoch in boot check
- This was introduced in 4c99e978a77a242e540cb8ccacb967d24322416c and was
incorrect; boot_epoch is the previous epoch t... - 08:14 PM Revision 78df66f5 (ceph): osd/ReplicatedPG: skip missing hit_sets when loading into memory
- We weren't handling hit_sets that were missing.
Two changes here:
1- Load the hit_sets oldest to newest. That mean... - 07:23 PM Revision 7077438b (ceph): mds: finish table servers recovery after creating newfs
- Fixes: #8054
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 06:12 PM Linux kernel client Bug #6809: 3.11 kernel panic: Workqueue: ceph-msgr con_work
- Another instance of the crash in the last comment, on ubuntu 12.04 with xfs on top with a 3.13 kernel when osds were ...
- 06:02 PM Bug #8082 (Duplicate): hung recovery
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-11_15:47:29-rados:thrash-testing-testing-basic-plana/1856...
- 05:57 PM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- 05:56 PM Bug #8048: osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is_missing(soid))
- 05:54 PM Bug #8080 (Fix Under Review): objecter: linger ops don't pay attention to cache overlay
- 04:14 PM Bug #8080 (In Progress): objecter: linger ops don't pay attention to cache overlay
- 04:07 PM Bug #8080 (Resolved): objecter: linger ops don't pay attention to cache overlay
- see wip-objecter for an initial stab. need to build out the tests now.
- 05:44 PM Bug #8081 (Resolved): hitset-get on missing object fails
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-04-11_15:47:29-rados:thrash-testing-testing-basic-plana/1857...
- 05:33 PM Revision 052519ed (ceph): Revert "mds: finish table servers recovery after creating newfs"
- This reverts commit f6c20730c16a7632061639dd83be523fc6a9a44f.
This breaks single MDS startup.
Signed-off-by: Sage W... - 05:20 PM Revision f3745912 (ceph): Merge pull request #1650 from dachary/wip-erasure-code-doc
- erasure-code: document the ruleset-root profile parameter
Reviewed-by: Mark Nelson <mark.nelson@inktank.com> - 05:08 PM Revision f12bccc9 (ceph): radosgw-admin: check params for object rewrite
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
(cherry picked from commit 28c716bccd955eaf68e9c739139c901ca4f419f7) - 05:08 PM Revision 9130e7d6 (ceph): radosgw-admin: new 'bucket rewrite' command
- Iterates through objects and rewrites them.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
(cherry picked from com... - 05:08 PM Revision 6f2ee99a (ceph): radosgw-admin: add some conditions for bucket rewrite
- --min-rewrite-size, --max-rewrite-size to specify object size
conditions
--start-date, --end-date to specify objec... - 05:08 PM Revision 82d8397a (ceph): rgw: update bucket / object rewrite
- Get code up to date for firefly, need to pass bucket owner for quota
related functionality.
Signed-off-by: Yehuda Sa... - 05:04 PM Revision fcd94d6a (ceph): Merge pull request #1630 from ceph/wip-7450
- Wip 7450
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 03:54 PM Bug #8043: until we fix it more better, we should disallow split on cache pools
- Do you really expect cache pool targets to not be set based on the underlying storage capacity? (Hint: no.)
;) - 03:40 PM Bug #8043: until we fix it more better, we should disallow split on cache pools
- How about a flag in the pg info that prevents the agent from blocking on full until it sees the thresholds drop below...
- 03:04 PM Bug #8043: until we fix it more better, we should disallow split on cache pools
- Requiring a scrub first means that any cache pool which is actually being used and gets split will fill up and block ...
- 02:55 PM Bug #8043: until we fix it more better, we should disallow split on cache pools
- Is the concern the need for scrub, or are there other issues?
We could make the (shallow) scrub triggered automati... - 03:48 PM Bug #7987: osd: backfill/recovery makes no progress
- Those two pgs appeared to go clean from the osd log.
- 03:38 PM Bug #8008 (Fix Under Review): osd/ReplicatedPG.cc: 258: FAILED assert(missing_loc.needs_recovery(...
- 02:58 PM Bug #8008 (In Progress): osd/ReplicatedPG.cc: 258: FAILED assert(missing_loc.needs_recovery(hoid)...
- 03:33 PM Revision ec4a6ce0 (ceph): Merge pull request #1635 from ceph/wip-7437
- Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Josh Durgin <j... - 03:25 PM rbd Feature #6934: Consistency check when importing RBD diff
- related: http://wiki.ceph.com/Planning/Blueprints/Giant/Diff%3A_integrity___local_import
- 02:48 PM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
- we haven't seen this in several days; i'm ready to call it resolved!
- 05:08 AM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
- All changes have been accepted upstream ( jerasure / gf-complete ), backported and copied into http://github.com/ceph...
- 03:29 AM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
- The above workload ran 82 times without crashing. The loop stopped because a DNS resolution failure made it impossibl...
- 02:48 PM Bug #8063 (Fix Under Review): LibRadosTwoPoolsECPP.PromoteSnap got EAGAIN
- 01:22 PM Bug #8063 (In Progress): LibRadosTwoPoolsECPP.PromoteSnap got EAGAIN
- 01:59 PM Revision 1ad5bdcd (ceph): Merge pull request #1641 from ceph/wip-multimds
- mds: guarantee message ordering when importing non-auth caps
Reviewed-by: Sage Weil <sage@inktank.com> - 01:57 PM Revision 60cb2aec (ceph): Merge branch 'master' of https://github.com/enovance/ceph-brag into fir...
- 01:56 PM Revision 204b7a46 (ceph): Merge pull request #1645 from ceph/wip-8054
- mds: finish table servers recovery after creating newfs
Reviewed-by: Sage Weil <sage@inktank.com> - 01:51 PM devops Bug #7617: ceph-deploy uninstall should document why it doesn't remove all relevant packages
- Oh, I dunno. I was depressed enough about the "qemu requires librbd" that I kinda lost interest.
I'm good with cl... - 07:27 AM devops Bug #7617: ceph-deploy uninstall should document why it doesn't remove all relevant packages
- @dmick would an acceptable workaround to this problem in the meantime be to add an extra flag to handle this?
Some... - 01:50 PM RADOS Feature #8079 (New): Warn user when there are no OSDs in the acting set during pool creation
- Create a parallel crush hierarchy using the ceph cli with:
1) it's own ruleset
2) no OSDs in the acting set
Wh... - 01:28 PM Revision ad40356d (ceph): Included the total cluster size in components_count object
- Signed-off-by: Babu Shanmugam <anbu@enovance.com>
- 01:14 PM Bug #8077 (Fix Under Review): osd/ReplicatedPG.cc: 10862: FAILED assert(r >= 0) (agent_load_hit_s...
- 01:03 PM Bug #8077 (In Progress): osd/ReplicatedPG.cc: 10862: FAILED assert(r >= 0) (agent_load_hit_sets)
- It's just missing; we don't handle that yet.
- 12:53 PM Bug #8077 (Resolved): osd/ReplicatedPG.cc: 10862: FAILED assert(r >= 0) (agent_load_hit_sets)
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-11_10:40:02-rados:thrash-firefly-distro-basic-plana...
- 12:05 PM Revision 78fcb1a0 (ceph): Fetching the date from ceph osd dump as that is more reliable across ce...
- Signed-off-by: Babu Shanmugam <anbu@enovance.com>
- 11:59 AM Revision 99d74eef (ceph): doc: Add additional information over CloudStack and RBD
- 11:51 AM Revision db3e0b51 (ceph): erasure-code: document the ruleset-root profile parameter
- If unspecified it is ruleset-root=default and will translate into
take default
when a ruleset is created for an ... - 11:28 AM Revision 6d42bd9f (ceph): While generating crush_types, avoiding item parsing, and calculating ty...
- Signed-off-by: Babu Shanmugam <anbu@enovance.com>
- 11:20 AM Bug #7952 (Resolved): After aio_read() completes a call to return_value() doesn't return bytes read
- d99f1d9f68db41231e0ffff4082b05d6d095c231
I forgot about this bug report, so it wasn't specified in the commit comm... - 10:38 AM Feature #7940 (Fix Under Review): add pool snaps to ceph_test_rados
- 10:06 AM Bug #8049 (Resolved): ceph_test_rados can seg fault
- 3371a251159a6e4ecd28a29e6a53705ccb3083a3
- 09:27 AM rgw Documentation #8073: rgw: document placement_pools
- See here:
[[http://comments.gmane.org/gmane.comp.file-systems.ceph.user/4992]] - 09:27 AM rgw Documentation #8073 (New): rgw: document placement_pools
- Need to document how different users / buckets can be assigned to different placement targets.
- 09:18 AM devops Bug #7889: IPv6 support with ceph-deploy
- There is a PR open with failing tests: https://github.com/ceph/ceph-deploy/pull/172
- 08:45 AM Feature #7437 (Resolved): EC: add adapt unittest teuthology task and add to nightly
- ec4a6ce0c2d175a3bf82059ff992f20ce18bb80e
The merged branch also includes improvements to the librados interface. F... - 08:42 AM Revision 19878325 (ceph): Bug fix in the way crush_type is extracted from osd crush dump
- Signed-off-by: Babu Shanmugam <anbu@enovance.com>
- 07:11 AM Revision e46af060 (ceph): Merge pull request #1647 from ceph/wip-lockdep
- a couple of lockdep fixes
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 05:11 AM Feature #8071 (Resolved): erasure-code: simplify jerasure initialization
- Take advantage of the newly exposed "galois_init_default_field":https://bitbucket.org/jimplank/jerasure/commits/af8d2...
- 04:36 AM Revision 072d3711 (ceph): RWLock: make lockdep id mutable
- This allows us to keep the lock/unlock methods const, as per commit
970d53fc0fefc89ffe7550880a4aaa36bd534955.
Signed... - 04:34 AM Revision da0d3820 (ceph): Revert "RWLock: don't assign the lockdep id more than once"
- This reverts commit 957ac3cbe394473f225ffd2b632461fcdaca99e6.
It's important to assign these for all operations for ... - 04:34 AM Revision 632098f2 (ceph): common_init: remove dup lockdep message
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:27 AM Bug #8069 (Resolved): ceph-deploy README suggests questionable SSH parameters
- In README.rst in ceph-deploy, the following sshd options are suggested for any host to which ceph-deploy should conne...
- 03:02 AM Revision 8c38ec7a (ceph): Merge pull request #1646 from dmick/wip-erasure-doc
- doc: Wordsmith the erasure-code doc a bit.
- 02:55 AM Revision 3c54a49e (ceph): Wordsmith the erasure-code doc a bit
- Signed-off-by: Dan Mick <dan.mick@inktank.com>
- 01:57 AM Revision f6c20730 (ceph): mds: finish table servers recovery after creating newfs
- Fixes: #8054
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> - 01:30 AM Revision 9fa5c5fa (ceph): big: test xfs + btrfs
- Note that previously we were specifying neither, so the OSDs were running
off of /.
Signed-off-by: Sage Weil <sage@i... - 01:25 AM Revision 756e3626 (ceph): Merge pull request #1643 from ceph/wip-8062
- mon/OSDMonitor: ignore boot message from before last up_from
Reviewed-by: Greg Farnum <greg@inktank.com> - 12:39 AM Revision 3db74861 (ceph): mds: issue new caps before starting log entry
- Locker::issue_new_caps() calls Locker::eval(), which may dispatch
other requests.
Signed-off-by: Yan, Zheng <zheng.z... - 12:22 AM Revision 98127202 (ceph): librados, test: Have write, append and write_full return 0 on success
- Fix consistency of write, append, write_full, all return 0 on success
Include C (rados_*) variants, C++ ctx variants
... - 12:22 AM Revision d2113814 (ceph): pybind: Check that "key" is a string
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:22 AM Revision 39bf68c3 (ceph): pybind, test: Add python binding for append and add to test
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:22 AM Revision 3d290c2f (ceph): test: Add EC unaligned append write test to ceph_test_rados_api_io
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:22 AM Revision d99f1d9f (ceph): test, librados: aio read *return_value consistency, fix ceph_test_rados...
- test:
Add set_completion*PP() functions to cast arg to correct class
Add return_value checks
Add some reads wit... - 12:22 AM Revision 69afc59b (ceph): test: Add multiple write test cases to ceph_test_rados_api_aio
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:22 AM Revision 07e8ee20 (ceph): test: Add EC testing to ceph_test_rados_api_aio
- Fixes: #7437
Signed-off-by: David Zafman <david.zafman@inktank.com>
04/10/2014
- 11:48 PM Revision 008663aa (ceph): rgw, radosgw-admin: bucket link uses bucket instance id now
- Fixes: 7499
We need to link user to a specific bucket instance. This updates both
the radosgw-admin link command, and... - 10:54 PM Revision 6ce7116f (ceph): civetweb: update subproject
- Fixes: #7786
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 09:53 PM Revision 6dc29904 (ceph): Change status to _status everywhere (fix regression)
- 0550dd119150acf8f7b7da2f32b3938c910c3c99 didn't add '_' to 'status'
everywhere it was needed.
Signed-off-by: Dan Mic... - 08:40 PM Revision 43d837d3 (ceph): rgw: radosgw-admin object rewrite
- A radosgw-admin command that copies the object into itself while
preserving mtime and attributes so that data can be ... - 08:34 PM Revision 4c99e978 (ceph): mon/OSDMonitor: ignore boot message from before last up_from
- It is possible we will have a dup OSDBoot message queued up in the mon
and will process it again after that osd was m... - 08:24 PM Revision 364d0b40 (ceph): ceph.conf: longer client_moutn_timeout (which also applies to librados)
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:50 PM CephFS Bug #8054: multimds hang on fsstress
- 06:54 AM CephFS Bug #8054 (Resolved): multimds hang on fsstress
- ceph-fuse:
ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-09_23:01:37-multimds-master-testing-bas... - 06:35 PM Bug #8067: mon: enomem on vps, killed at ~800MB
- this is probably a dup of #8036
- 06:13 PM Bug #8067 (Duplicate): mon: enomem on vps, killed at ~800MB
- ubuntu@teuthology:/a/teuthology-2014-04-09_22:35:02-upgrade:dumpling-x:stress-split-firefly-distro-basic-vps/182651
... - 06:31 PM Bug #8036: levedb: throws std::bad_allow on 14.04
- It would be interesting to know why the monitor's virtual mem usage got to 700MB, although a portion of it should go ...
- 01:19 PM Bug #8036: levedb: throws std::bad_allow on 14.04
- It's a good practice to run tests on scaled down machines, the question then is - do we fix bugs related like this on...
- 11:54 AM Bug #8036: levedb: throws std::bad_allow on 14.04
- This was run on vms, so they have much less memory than the usual physical machines.
- 11:25 AM Bug #8036: levedb: throws std::bad_allow on 14.04
- core is corrupted:...
- 06:24 PM Bug #8062 (Resolved): [ERR] map e10 had wrong cluster addr (10.214.131.8:6815/1007096 != my 10....
- 01:34 PM Bug #8062 (Fix Under Review): [ERR] map e10 had wrong cluster addr (10.214.131.8:6815/1007096 !...
- 01:33 PM Bug #8062 (Resolved): [ERR] map e10 had wrong cluster addr (10.214.131.8:6815/1007096 != my 10....
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-10_02:30:24-rados-firefly-distro-basic-plana/183189...
- 06:18 PM Bug #8068 (Resolved): try_flush_mark_clean can end up using a snapset from the past corrupting th...
- 06:05 PM Bug #8066: osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + firefly)
- ubuntu@teuthology:/a/teuthology-2014-04-09_22:35:02-upgrade:dumpling-x:stress-split-firefly-distro-basic-vps/182696
- 06:02 PM Bug #8066 (Duplicate): osd/PG.cc: 2826: FAILED assert(r == 0) in update_snap_map (dumpling + fire...
- ubuntu@teuthology:/a/teuthology-2014-04-09_22:35:02-upgrade:dumpling-x:stress-split-firefly-distro-basic-vps/182710<p...
- 06:01 PM Revision 28371a24 (ceph): Merge pull request #1624 from ceph/wip-6789
- mon: Monitor: suicide on start if mon has been removed from monmap
Reviewed-by: Sage Weil <sage@inktank.com> - 05:55 PM Revision a8f09539 (ceph): osd/ReplicatedPG: adjust obc + snapset_obc locking strategy
- Prevoiusly we assumed that if we had snapset_obc set, !exists on the head
and if we got the snapdir lock we were good... - 05:53 PM Bug #7996: 0.78: OSD is not suspend-friendly (unresponsive cluster on OSD crash)
- When cluster is stable output of `ceph -s` is normal, showing health status HEALTH_OK, 5 monitors and 11 OSDs up and ...
- 05:26 PM Feature #7940: add pool snaps to ceph_test_rados
- 02:14 PM Revision 86b85947 (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... - 01:49 PM devops Bug #8034 (Resolved): ceph-deploy should run sudo yum clean all after installing ceph-release rpm
- Merged into ceph-deploy's master branch with hash: 5d82162
- 09:54 AM devops Bug #8034 (Fix Under Review): ceph-deploy should run sudo yum clean all after installing ceph-rel...
- Pull request opened https://github.com/ceph/ceph-deploy/pull/177
- 09:48 AM devops Bug #8034 (In Progress): ceph-deploy should run sudo yum clean all after installing ceph-release rpm
- 01:41 PM Bug #8063 (Resolved): LibRadosTwoPoolsECPP.PromoteSnap got EAGAIN
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-10_02:30:24-rados-firefly-distro-basic-plana/183560...
- 01:25 PM Bug #8061: LibRadosAio.FlushAsync failure
- teuthology commit 364d0b4091a9b87db9ca21238052b776d0dce27c
- 01:24 PM Bug #8061 (Resolved): LibRadosAio.FlushAsync failure
- mon thrashing. increased the timeout from 5m to 10m.
- 01:20 PM Bug #8061 (Resolved): LibRadosAio.FlushAsync failure
- ...
- 11:46 AM Revision 02048dcc (ceph): mds: guarantee message ordering when importing non-auth caps
- Current code allow importing non-auth caps when inode is being exported.
This can breaks message ordering because the... - 11:01 AM Bug #6789 (Pending Backport): cannot remove the leader when there only are two monitors
- 10:55 AM Bug #8046 (Resolved): osd/ReplicatedPG.h: 666: FAILED assert(got) in get_rw_locks()
- 10:45 AM rgw Bug #8016: "testPrefixAndLimit (test.functional.tests.TestContainerUTF8) ... ERROR" in upgrade:du...
- I see seemingly the same issue in today run.
Logs are here - http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-... - 10:37 AM RADOS Feature #8060 (New): Make it easier to map crush rules to pools
- Right now when you want to create a new crush rule via create-simple, you do something like:...
- 10:31 AM rgw Bug #8059 (Resolved): ACL updated in master zone not updating in slave
- In Summary:
While using a setup with a master and slave zone the ACL for an object is updated on the master zone, ... - 10:21 AM Linux kernel client Bug #8024 (Resolved): kclient: misdirected osd request
- 10:21 AM Linux kernel client Bug #7954 (Resolved): misdirected op
- 07:13 AM Linux kernel client Bug #7954 (Fix Under Review): misdirected op
- 07:05 AM Linux kernel client Bug #7954: misdirected op
- probably this: ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2014-04-09_23:01:12-kcephfs-master-testing-b...
- 10:17 AM Bug #8021: osd: ENOENT on clone on dumpling
- I think I see it on firefly if it's the same issue.
See logs here:
http://qa-proxy.ceph.com/teuthology/teuthology-2... - 09:51 AM rgw Bug #7786: civetweb segfaults with file uploads larger than 2GB
- Please pull down the newest civitweb that includes this fix.
- 09:50 AM rgw Bug #7450 (Fix Under Review): "radosgw-admin key create" ignores specified access key when subuse...
- Josh please review PR for wip-7450
- 09:30 AM Bug #8058 (Resolved): "LibRadosTierECPP.FlushWriteRaces" failed in upgrade:dumpling-x:parallel-fi...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-04-09_19:33:29-upgrade:dumpling-x:parallel-firefly---...
- 06:58 AM CephFS Bug #8055 (Can't reproduce): knfs: NFS: nfs4_discover_server_trunking unhandled error -5. Exiting...
- ...
- 04:19 AM Revision cf69bdbd (ceph): Merge pull request #1639 from ceph/wip-multimds
- Wip multimds
Reviewed-by: Sage Weil <sage@inktank.com> - 03:13 AM Revision ac51fcac (ceph): mds: include truncate_seq/truncate_size in filelock's state
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 03:13 AM Revision 808ba130 (ceph): mds: remove wrong assertion for remote frozen authpin
- For across authority rename, the MDS first freezes the source inode's
authpin. It happens while the source dentry isn... - 01:02 AM Revision 860d7277 (ceph): osdc/Objecter: move mapping into struct, helper
- Move the common bits of Op and LingerOp into op_target_t and separate the
actual mapping calculation into calc_target... - 12:21 AM Revision 5df98f47 (ceph): Merge pull request #1637 from ceph/wip-8042
- mon: fix election required_features checks
Reviewed-by: Greg Farnum <greg@inktank.com>
04/09/2014
- 11:51 PM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
- The above workload ran 51 times without crashing.
- 03:16 AM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
"init fix merged":https://bitbucket.org/jimplank/jerasure/pull-request/17/make-galois_init_default_field-int-w , "b...- 03:15 AM Bug #7914: osd: SEGV on ec write, ErasureCodeJerasure: encode adjusted buffer length from 4096 to...
- The above workload ran 26 times without crashing. rados suite did not run yet on the latest master.
- 11:25 PM Revision 71d97f99 (ceph): Merge pull request #1636 from ceph/wip-6480
- fix auth races that may have lead to qemu crashes
Reviewed-by: Sage Weil <sage@inktank.com> - 11:21 PM Revision 48895a46 (ceph): osdc/ObjectCacher: back off less during flush
- In cce990efc8f2a58c8d0fa11c234ddf2242b1b856 we added a limit to avoid
holding the lock for too long. However, if we ... - 11:21 PM Revision f3b5ba6f (ceph): osdc/ObjectCacher: limit writeback IOs generated while holding lock
- While analyzing a log from Mike Dawson I saw a long stall while librbd's
objectcacher was starting lots (many hundred... - 11:03 PM Revision 18642ed3 (ceph): mon: tell peers missing features during probe
- Use a new probe op to inform mons that they are missing features during
the earliest probe phase. This prevents them... - 11:01 PM Revision 39ca440b (ceph): mon: move required_features back into Monitor
- This is simpler and cleaner.
Signed-off-by: Sage Weil <sage@inktank.com> - 09:40 PM Revision c8039ab8 (ceph): mon: ignore sync clients without required_features
- If we let them sync data they don't understand they will get confused
and crash.
Signed-off-by: Sage Weil <sage@inkt... - 09:34 PM Bug #7987: osd: backfill/recovery makes no progress
- It looks like thrashosds gave up waiting as it usually does (it doesn't kill the child task so you just don't see any...
- 09:31 PM Revision 75948357 (ceph): utime: declare is_zero(), ceph_timespec(), and sleep() as const
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 09:31 PM Revision 9af10b2c (ceph): auth: CephxProtocol const cleanup
- need_key() and build_authorizer() can be const.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 09:31 PM Revision 3ccef662 (ceph): auth: AuthClientHandler const cleanup
- get_protocol(), build_request(), build_rotating_request(), and
build_authorizer() can all be declared const now.
Sig... - 09:31 PM Revision b297689a (ceph): auth: make AuthClientHandler::validate_ticket() protected
- It's just used internally. Make it private in the subclasses since
there's just one level of inheritance.
Signed-off... - 09:31 PM Revision 50ed65fb (ceph): auth: remove unused get_global_id() method
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 09:31 PM Revision 4d3d89bf (ceph): auth: remove unused tick() method
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 09:31 PM Revision 957ac3cb (ceph): RWLock: don't assign the lockdep id more than once
- This never does anything since lockdep_register() assigns an id >= 0
in the RWLock constructor. This also prevents me... - 09:31 PM Revision 970d53fc (ceph): RWLock: make read locking methods const
- This allows methods using RWLock for reading to be declared const.
There might be cases where we'd want to take a wri... - 09:31 PM Revision 3119022d (ceph): auth: separate writes of build_request() into prepare_build_request()
- validate_tickets() updates internal state, as does
tickets.get_handler(). Move them into a new method called before
b... - 09:29 PM Revision 2cc76bcd (ceph): auth: add rwlock to AuthClientHandler to prevent races
- For cephx, build_authorizer reads a bunch of state (especially the
current session_key) which can be updated by the M... - 09:27 PM Bug #8042 (Resolved): mon: crash decoding incremental osdmap on split firefly/dumpling
- 04:14 PM Bug #8042 (Fix Under Review): mon: crash decoding incremental osdmap on split firefly/dumpling
- 11:23 AM Bug #8042: mon: crash decoding incremental osdmap on split firefly/dumpling
- 11:16 AM Bug #8042 (Fix Under Review): mon: crash decoding incremental osdmap on split firefly/dumpling
- 09:35 AM Bug #8042: mon: crash decoding incremental osdmap on split firefly/dumpling
- ...
- 09:20 PM Revision d4f2a8ce (ceph): Reduce runs in the dumpling-emperor upgrade suite.
- Add + to 2-workload, 4-workload and 6-final-workload
suites. Reduce 5-upgrade-sequence to only run two
tests and mak... - 08:59 PM Revision f1d63236 (ceph): Merge pull request #238 from ceph/wip-pep8
- Wip pep8
- 08:56 PM Revision 0550dd11 (ceph): more statuses redefinition fixes
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 08:36 PM Revision 30f3b01a (ceph): rename variable to avoid collision
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 08:27 PM Revision b476ec29 (ceph): remove unused import
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 08:26 PM Revision 15a3acb7 (ceph): fix redefinition of region
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 06:35 PM Revision 2e8035fa (ceph): osd: Fix appending write to return any error back to caller
- Also, correct double bumping of num_writes
Signed-off-by: David Zafman <david.zafman@inktank.com> - 06:35 PM Revision 3371a251 (ceph): test: Fix Seg fault in ceph_test_rados
- Fixes: #8049
Signed-off-by: David Zafman <david.zafman@inktank.com> - 06:35 PM Revision edd542e4 (ceph): tools: Improve ceph_scratchtoolpp
- Minor output improvements
Remove clone_range code that was asserting
Signed-off-by: David Zafman <david.zafman@inkta... - 06:13 PM Revision 34d69cdc (ceph): mon: refresh elector required_features when they change
- Currently we only refresh required_features on Elector::start(). This
does not prevent an old peer from calling an e... - 06:09 PM Revision b3b502f1 (ceph): mon/Elector: ignore ACK from peers without required features
- If an old peer gets a PROPOSE from us, we need to be sure to ignore their
ACK. Ignoring their PROPOSEs isn't suffici... - 05:37 PM Revision 5a567c47 (ceph): Merge pull request #1626 from ceph/wip-8031
- osd: improve misdirected op checks
Reviewed-by: Samuel Just <sam.just@inktank.com> - 05:34 PM Revision 5b16650b (ceph): Merge pull request #1627 from ceph/wip-8001
- osd/PG: set CREATING pg state bit until we peer for the first time
Reviewed-by: Samuel Just <sam.just@inktank.com> - 05:34 PM Revision 2a9f5fd5 (ceph): Merge pull request #1631 from ceph/wip-8045
- osd: fix check_osdmap_features deadlock
Reviewed-by: Samuel Just <sam.just@inktank.com> - 04:58 PM devops Bug #5193 (In Progress): RHEL6 does not ship with xfsprogs
- Not marking it as complete yet because the automation/insertion in jenkins is not complete but for now I built packaa...
- 04:24 PM rbd Bug #6480 (Pending Backport): librbd crashed qemu-system-x86_64
- 03:57 PM Bug #8020: evenly split stats on split
- 03:48 PM Revision 3382fe3c (ceph): os/FileStore: reset journal state on umount
- We observed a sequence like:
- replay journal
- sets JournalingObjectStore applied_op_seq
- umount
- mount
... - 03:47 PM Revision 06f27fc6 (ceph): os/FileStore: reset journal state on umount
- We observed a sequence like:
- replay journal
- sets JournalingObjectStore applied_op_seq
- umount
- mount
... - 03:26 PM rgw Feature #8052 (Duplicate): Support for Keystone Identity API v3
- With Havana, Identity V3.0 was implemented, with support for domains.
RGW seems to only support Identity API v2.0... - 03:14 PM Revision f1a89340 (ceph): Merge pull request #1632 from ceph/wip-5469
- librbd: fix zero length request handling
Reviewed-by: Sage Weil <sage@inktank.com> - 02:18 PM Bug #7976: 4.8 missing primary copy of ..., unfound (dumpling)
- 4139e75d63b0503dbb7fea8036044eda5e8b7cf1 and 7a06a71e0f2023f66d003dfb0168f4fe51eaa058 are part of the story. We prob...
- 02:17 PM Bug #7976 (Pending Backport): 4.8 missing primary copy of ..., unfound (dumpling)
- Actually, this is mildly tricky since the backfill logic changed a bit after dumpling to avoid some other related cat...
- 02:09 PM Bug #7976: 4.8 missing primary copy of ..., unfound (dumpling)
- mm, doesn't apply cleanly, working on it
- 01:48 PM Bug #7976 (Fix Under Review): 4.8 missing primary copy of ..., unfound (dumpling)
- 01:48 PM Bug #7976: 4.8 missing primary copy of ..., unfound (dumpling)
- wip-7676-dumpling I think
- 09:54 AM Bug #7976: 4.8 missing primary copy of ..., unfound (dumpling)
- This is a backfill bug in dumpling. The problem seems to be somewhere around here... the backfill for 57594e48/bench...
- 08:28 AM Bug #7976: 4.8 missing primary copy of ..., unfound (dumpling)
- 02:12 PM Revision dc4bbfa7 (ceph): Merge pull request #1634 from ceph/wip-8028
- rpm: add redhat-lsb dependency
Reviewed-by: Alfredo Deza <alfredo.deza@inktank.com> - 02:05 PM Revision f1c6b65b (ceph): ceph.spec.in: require redhat-lsb-core
- We need this for /lib/lsb/init-functions.
Fixes: #8028
Signed-off-by: Sage Weil <sage@inktank.com> - 01:55 PM Revision 1d0c62fa (ceph): Merge pull request #1606 from ceph/wip-shrink-icache
- client: try shrinking kernel inode cache when trimming session caps
Reviewed-by: Sage Weil <sage@inktank.com> - 11:15 AM Bug #8049 (Resolved): ceph_test_rados can seg fault
$ ./ceph_test_rados --pool ctrpool --op read 4 --op write 20
.
.
.
update_object_version oid 5 v 384 (ObjNum 32...- 10:37 AM Bug #8031 (Resolved): osd/ReplicatedPG.cc: 405: FAILED assert(needs_recovery)
- 10:35 AM Bug #8001 (Resolved): hung recovery; pg 3.f disappeared
- 10:35 AM Bug #8045 (Resolved): osd: deadlock from osdmap feature update
- 09:31 AM Bug #8047 (Closed): 0.79: new OSD crashed within minutes
- 12:12 AM Bug #8047: 0.79: new OSD crashed within minutes
- Never mind, root cause for this one is physical media errors in HDD. Apologies for noise -- please close this bug.
T... - 09:22 AM Feature #6707 (Rejected): Backport per pool iops data to 0.67
- 08:47 AM Bug #8019 (Resolved): os/JournalingObjectStore.cc: 121: FAILED assert(op > committed_seq) on wheezy
- 08:29 AM Bug #7891: osd: leaked pg refs on shutdown
- http://qa-proxy.ceph.com/teuthology/joshd-2014-04-08_17:21:20-rados-wip-6480-testing-basic-plana/179778/
- 08:23 AM Bug #8048 (Resolved): osd/ReplicatedPG: FAILED assert(!parent->get_log().get_missing().is_missing...
- From effectively the master branch: http://qa-proxy.ceph.com/teuthology/joshd-2014-04-08_17:21:20-rados-wip-6480-test...
- 08:14 AM rbd Bug #5469 (Resolved): qemu-io: segfault when tried IO with invalid arguments
- 07:19 AM Bug #7996: 0.78: OSD is not suspend-friendly (unresponsive cluster on OSD crash)
- What's the output of "ceph -s" while the cluster is stable, while it's frozen, and after it's recovered?
- 02:03 AM Bug #7996: 0.78: OSD is not suspend-friendly (unresponsive cluster on OSD crash)
- I reproduced this problem on 0.79. It looks like it is no different from 0.78 in regards to this problem.
I suspen... - 07:13 AM Bug #8028 (Resolved): /lib/lsb/init-functions does not exist in latest firefly rc
- 06:05 AM Bug #8028: /lib/lsb/init-functions does not exist in latest firefly rc
- It works great except that it pulls a lot of other things other than the init-functions that we need:...
- 06:07 AM Revision ae09361a (ceph): mailmap: Yan, Zheng name normalization
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 06:06 AM Revision 5ee0b7b2 (ceph): mailmap: Nikola Kotur affiliation
- Reviewed-by: Nikola Kotur <kotnick@gmail.com>
Signed-off-by: Loic Dachary <loic@dachary.org> - 03:48 AM Revision e5f3eb82 (ceph): Merge pull request #1633 from ceph/wip-8004
- client: wake up umount waiter if receiving session open message
Reviewed-by: Sage Weil <sage@inktank.com> - 03:23 AM Revision 02aedbc4 (ceph): client: wake up umount waiter if receiving session open message
- Wake up umount waiter if receiving session open message while
umounting. The umount waiter will re-close the session.... - 02:30 AM Revision 1848a237 (ceph): librados: Fix typo for read functions documentation
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 02:30 AM Revision 9c6733e6 (ceph): librados: Allow rados_pool_list() to get NULL buffer if len == 0
- Coverity wants a NULL check for buf, but caller should be able to
pass a NULL buf and 0 len in order to get byte coun... - 02:30 AM Revision 409999c3 (ceph): rbd: Prevent Seg fault by checking read result in snap_read_header()
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 01:25 AM Revision 0bed25c6 (ceph): Merge pull request #236 from ceph/wip-fix-mon-warn-corrected
- changed mon data avail warn to avoid monitor storage warnings
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 01:21 AM Revision b3218eeb (ceph): changed mon data avail warn to avoid monitor storage warnings
- 12:39 AM Revision a8330f5c (ceph): librbd: fix zero length request handling
- Zero-length writes would hang because the completion was never
called. Reads would hit an assert about zero length in... - 12:28 AM Revision 22a0c1fd (ceph): osd: do not block when updating osdmap superblock features
- We are holding osd_lock in check_osdmap_features, which means we cannot
block while waiting for filestore operations ...
Also available in: Atom