Activity
From 09/01/2014 to 09/30/2014
09/30/2014
- 11:57 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- Working in the container...
- 11:54 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- To make sure this is not environmental problem I clone a clean copy from your branch and removed .ccache entirely.
- 11:19 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- Running the test in the container still fails. ...
- 11:06 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- I reproduced the above valgrind output a few minutes ago on my development laptop. After upgrading from...
- 10:55 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- Using the same source tree with the same kernel but inside an ubuntu 14.04 docker container, I was not able to reprod...
- 11:12 PM Revision 824031b2 (ceph): upgrade/dumpling-giant-x: remove
- Signed-off-by: Sage Weil <sage@redhat.com>
- 11:09 PM Revision 28449710 (ceph): upgrade/cuttlefish, emperor, old: remove
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:45 PM Revision 544d8255 (ceph): Merge pull request #2613 from ceph/wip-8822
- osd: swap state spinlock for atomic_t
Reviewed-by: Samuel Just <sam.just@inktank.com> - 09:15 PM Revision 624aaf2a (ceph): PG: release backfill reservations if a backfill peer rejects
- Also, the full peer will wait until the rejection from the primary
to do a state transition.
Fixes: #9626
Backport: ... - 06:56 PM Revision 62e2bca8 (ceph): osd: swap state spinlock for atomic_t
- We are hitting a strange issue with valgrind and pthread spinlocks. Avoid
the issue by using an atomic_t here (which ... - 05:52 PM Revision 79825e7b (ceph): Merge pull request #2611 from dachary/wip-9570-buffer-alignment-dumpling
- common: buffer alignment (dumpling)
Reviewed-by: Sage Weil <sage@redhat.com> - 05:49 PM Revision e30969b1 (ceph): Merge pull request #2612 from ceph/wip-9562-backport
- osdc/Filer: drop probe/purge locks before calling objecter
Reviewed-by: Greg Farnum <greg@inktank.com> - 05:46 PM Revision a8ac4b62 (ceph): osdc/Filer: drop probe/purge locks before calling objecter
- Fixes: #9562
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 8dc94a2d8ce3364c0d8d52f634... - 05:24 PM Revision 0ea20a66 (ceph): Locker: accept ctime updates from clients without dirty write caps
- The ctime changes any time the inode does. That can happen even without
the file itself having changed, so we'd bette... - 05:24 PM rgw Bug #8587 (Resolved): rgw: subuser object not created correctly
- commit:1441ffe8103f03c6b2f625f37adbb2e1cfec66bb
- 05:19 PM Bug #9635: mon/Paxos.cc: 1033: FAILED assert(mon->is_leader())
- 05:19 PM Bug #9635 (Fix Under Review): mon/Paxos.cc: 1033: FAILED assert(mon->is_leader())
- from teh log it looks like this happened during shutdown. see wip-9635
- 04:54 PM Bug #9635 (Resolved): mon/Paxos.cc: 1033: FAILED assert(mon->is_leader())
- ...
- 05:14 PM Revision 70406edc (ceph): Merge pull request #2605 from ceph/wip-osd-objecter
- osd: do not bind ms_objecter messenger
Reviewed-by: Greg Farnum <greg@inktank.com> - 04:58 PM rgw Bug #9575: s3tests.functional.test_s3.test_region_copy_object fails (races with radosgw-agent?)
- hmm, these seem to always happen with valgrind!
- 04:52 PM rgw Bug #9575: s3tests.functional.test_s3.test_region_copy_object fails (races with radosgw-agent?)
- ubuntu@teuthology:/a/teuthology-2014-09-29_23:02:01-rgw-giant-testing-basic-multi/519792
- 04:43 PM Revision d6a53ea0 (ceph): common/buffer.cc: fix rebuild_page_aligned typo
- Introduced: 66a9fbe2c7ba59b7cd034c17865adce3432cd2cb
Fixes: #6003
Signed-off-by: Samuel Just <sam.just@inktank.com>
(... - 04:43 PM Revision 1d7a228d (ceph): common: rebuild_page_aligned sometimes rebuilds unaligned
- rebuild_page_aligned relies on rebuild to create memory that is aligned
according to list::is_page_aligned(). However... - 03:32 PM Bug #9459 (Need More Info): osd: blocked request
- 03:31 PM Bug #9288 (Duplicate): "Assertion `nlock == 0' failed" in upgrade:firefly-firefly-testing-basic-v...
- see #9040
- 03:09 PM Bug #8997 (Can't reproduce): ceph_test_rados_watch_notify hangs
- I suspect the watch resend fix (commit:1349383ac416673cb6df2438729fd2182876a7d1 for #9220) fixed some of these. (It ...
- 03:06 PM Bug #8595: osd: client op blocks until backfill starts (dumpling)
- The simple fixes here seem insufficient (fail in qa). Haven't seen anybody else hitting this, which surprises me a b...
- 02:59 PM Revision 58961d97 (ceph): Merge pull request #2597 from cchengleo/wip-test-shared-cache
- test shared cache
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 02:54 PM Revision 2e9c7c9a (ceph): test-shared-cache:
- Add test for "lower_bound" method.
Add multithread tests for "lookup" and "lower_bound" methods.
Signed-off-by: Ch... - 02:53 PM Revision d34c21ce (ceph): test-shared-cache:
- Initial draft for the unit test of "common/shared_cache.hpp".
Signed-off-by: Cheng Cheng <ccheng.leo@gmail.com>
... - 01:22 PM Feature #9198 (In Progress): librados: notify callback includes gid of notifier
- 01:22 PM Feature #9197 (In Progress): librados/osd: notify reply payload
- 01:13 PM Feature #8899 (Fix Under Review): Kerberos/LDAP Support:: mon: define mon role capabilities
- 01:03 PM RADOS Feature #9632 (New): testing: test CrushWrapper::get_full_location_ordered()
- A recent backport of changes to get_full_location_ordered() passed all the make check and RADOS suite tests, but caus...
- 01:03 PM Revision 172db92c (ceph): Merge pull request #2609 from dachary/wip-doc-context
- common: document C_GatherBuilder and C_Gather
Reviewed-by: John Spray <john.spray@redhat.com> - 01:01 PM Revision de87d545 (ceph): common: document C_GatherBuilder and C_Gather
- * Introduction explaining what they are used for
* Commented example to illustrate the abstract description
Signed-o... - 12:58 PM Revision a67c2f97 (ceph): mds: Add session to sessionmap when its state becomes opening
- following sequence events may happen.
- client sends reconnect message to a recovering MDS.
- MDS::ms_verify_authoriz... - 12:56 PM Feature #9031: List RADOS namespaces and list all objects in all namespaces
- 11:52 AM Bug #8822 (Need More Info): osd: hang on shutdown, spinlocks
- 11:51 AM Bug #8822: osd: hang on shutdown, spinlocks
- valgrind is 1:3.10~20140411-0ubuntu1
3.10.0 release notes claim to have fixed
336435 Valgrind hangs in pthread... - 11:39 AM Bug #8822: osd: hang on shutdown, spinlocks
- http://stackoverflow.com/questions/24558914/valgrind-hangs-in-pthread-spin-lock-consuming-100-cpu
valgrind bug? - 11:38 AM Bug #8822: osd: hang on shutdown, spinlocks
- happened again:...
- 11:26 AM Bug #9617: objecter shutdown races with msg dispatch
- wip-objecter-shutdown
- 11:17 AM rgw Feature #8911 (In Progress): RGW doesn't return 'x-timestamp' in header which is used by 'View De...
- 10:29 AM CephFS Bug #9562 (Pending Backport): Lockdep assertion in Filer purge
- This is popping up in Giant as well, which I believe has the new code that was the proximate cause. :)
- 10:27 AM CephFS Bug #9514 (Pending Backport): ceph-fuse pjd test is failing in giant nightlies
- In giant as commit:0ea20a668cf859881c49b33d1b6db4e636eda18a.
Needs to go to firefly as well. - 09:58 AM devops Tasks #8366 (In Progress): Update ceph.com/docs to default to the latest major release (0.80)
- 09:47 AM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- https://github.com/ceph/ceph/pull/2611 seems like a good candidate for backport.
- 09:40 AM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- https://github.com/ceph/ceph/commit/66a9fbe2c7ba59b7cd034c17865adce3432cd2cb and https://github.com/ceph/ceph/commit/...
- 08:41 AM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- None of the commits in FileJournal.cc from dumpling to master fix something that could cause a problem of that nature.
- 09:40 AM Bug #9630 (Resolved): osd: leaked pg refs on shutdown (dumpling)
- ...
- 08:40 AM Bug #9610: Crash "RadosModel.h: In function 'virtual void WriteOp::_finish(TestOp::CallbackInfo*)...
- 9/30/14 update - Still waiting in queue http://pulpito.front.sepia.ceph.com/teuthology-2014-09-29_23:20:02-multi-vers...
- 08:36 AM rgw Bug #9612 (Resolved): "ERROR: test suite for <module 's3tests.functional'" in multi-version-giant...
- PR https://github.com/ceph/ceph-qa-suite/pull/154
- 01:12 AM Revision 1feba200 (ceph): Objecter: init with a constant of the correct type
- Just a tiny cleanup.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 12:08 AM CephFS Bug #9628: mds: race between ms_handle_accept() and ms_handle_reset()
- https://github.com/ceph/ceph/pull/2596
- 12:08 AM CephFS Bug #9628 (Resolved): mds: race between ms_handle_accept() and ms_handle_reset()
- ceph version 0.85-1003-g3ae673c (3ae673c764a4fac6e554e05722f0179566ed3fb3)
1: (ceph::BackTrace::BackTrace(int)+0x2... - 12:06 AM Revision e94d489c (ceph): Merge pull request #156 from ceph/wip-9609-wusui
- Fix workunit branches.
09/29/2014
- 11:48 PM Revision 9c4616dc (ceph): Merge pull request #2556 from dachary/wip-9572-erasure-code-blaumroth
- erasure-code: restore jerasure BlaumRoth default w
Reviewed-by: Sage Weil <sage@redhat.com> - 11:43 PM Revision ccf3e06a (ceph): Merge pull request #2502 from yuyuyu101/kvstore-meta-cache
- KeyValueStore: performance improvement and docs
- 11:41 PM Revision 1daf0551 (ceph): Merge pull request #2523 from ceph/wip-6926
- rbd: Fix rbd diff for non-existent objects
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 11:40 PM Bug #9582: librados: segmentation fault on timeout
- Thanks for your investigations and the quick fix! We have not been able to test this fix yet, but I will report back ...
- 01:13 PM Bug #9582: librados: segmentation fault on timeout
- in giant, dumpling. still need to merge firefly backport.
- 01:10 PM Bug #9582 (Pending Backport): librados: segmentation fault on timeout
- 08:16 AM Bug #9582 (Fix Under Review): librados: segmentation fault on timeout
- 11:28 PM Revision 9a2974e5 (ceph): Fix workunit branches.
- FIxes: 9606
Signed-off-by: Warren Usui <warren.usui@inktank.com> - 11:24 PM Revision 5281ac7a (ceph): Merge pull request #2601 from ceph/wip-python-flask
- debian: python-flask is needed by ceph, not python-ceph
Reviewed-by: Loic Dachary <loic@dachary.org> - 11:17 PM Revision f36e8540 (ceph): Merge remote-tracking branch 'gh/giant'
- Conflicts:
qa/workunits/cephtool/test.sh
src/test/erasure-code/TestErasureCodeIsa.cc - 11:11 PM Revision 46d55186 (ceph): osd: do not bind ms_objecter messenger
- The objecter messenger is only used as a client to initiate client-side
connections to other OSDs. It doesn't need t... - 10:09 PM Bug #9459: osd: blocked request
- saw something similar on another cluster, ...
- 09:28 PM Revision fe3434f4 (ceph): debian: move ceph_rest_api.py into ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:24 PM Revision 8cda623e (ceph): ceph.spec.in: move ceph_rest_api.py into ceph
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:18 PM rbd Bug #9602: rbd export -> nc ->rbd import = memory leak
- As a suggestion, prohibit the use of the cache when RDB imports.
- 09:03 PM rbd Bug #9602: rbd export -> nc ->rbd import = memory leak
- Hi, Sage.
I'm sorry, was wrong to put up parameter: rbd_cache size
The problem is not confirmed.
- 09:14 PM Revision 0d1a4171 (ceph): Merge pull request #2602 from athanatos/wip-9574
- PG: check full ratio again post-reservation
Reviewed-by: Sage Weil <sage@redhat.com> - 09:02 PM Revision acd49892 (ceph): Merge pull request #2604 from athanatos/wip-9113
- ReplicatedPG: clean out completed trimmed objects as we go
Reviewed-by: Sage Weil <sage@redhat.com> - 08:55 PM Revision 2ecf362e (ceph): Merge pull request #2603 from dachary/wip-9620-test-mon-thrash-firefly
- qa/workunits/cephtool/test.sh: fix thrash (ultimate)
Reviewed-by: Sage Weil <sage@redhat.com> - 08:54 PM Revision 78fc7b81 (ceph): Merge pull request #2549 from ceph/wip-9545
- os/FileJournal: do not request sync while shutting down
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:52 PM Revision f91c571e (ceph): Merge pull request #2550 from ceph/wip-8629
- osd: fix cache_evict vs make_writeable/finish_ctx snapdir bug #8629
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:50 PM Revision 76341b0b (ceph): qa/workunits/cephtool/test.sh: fix thrash (ultimate)
- Keep the osd trash test to ensure it is a valid command but make it a
noop by giving it a zero argument (meaning thra... - 08:44 PM Revision ffda34c4 (ceph): Merge pull request #2510 from somnathr/wip-obj-delete-fix
- FileStore: Race condition during object delete is fixed
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:44 PM Revision b2416240 (ceph): ceph.spec: fix python-flask dependency
- This is needed by ceph-rest-api, which is in ceph.rpm; it's not related to
python-ceph (except that ceph-rest-api hap... - 08:40 PM Revision e42424e7 (ceph): debian: python-flask is needed by ceph, not python-ceph
- It's used by ceph-rest-api which is in the 'ceph' (server) package.
Backport: firefly
Signed-off-by: Sage Weil <sage... - 08:18 PM CephFS Bug #9562 (Resolved): Lockdep assertion in Filer purge
- 08:12 PM Revision 3ebc3502 (ceph): osdc/Objecter: only post_rx_buffer if no op timeout
- If we post an rx buffer and there is a timeout, the revocation can happen
while the reader has consumed the buffers b... - 08:08 PM Revision 614157c2 (ceph): Merge pull request #2598 from ceph/wip-9582
- librados: fix other timeout segfault
Reviewed-by: Greg Farnum <greg@inktank.com> - 08:05 PM Revision 98322391 (ceph): Merge pull request #155 from ceph/wip-9341
- tasks/mds_client_recovery: client trim its cache on reconnect
Reviewed-by: John Spray <john.spray@redhat.com> - 08:02 PM Revision 9d2481c7 (ceph): Merge pull request #2458 from ceph/wip-9341
- client: trim unused inodes before reconnecting to recovering MDS
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewe... - 07:59 PM Revision ff03b465 (ceph): tasks/mds_client_recovery: client trim its cache on reconnect
- make sure CephFS client trim its cache before reconnect to the MDS.
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 06:25 PM Revision 2974c7b8 (ceph): Merge pull request #2555 from theanalyst/wip/mailmap
- mailmap updates
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 06:12 PM rgw Bug #9615 (Resolved): "ERROR: test suite for <module 's3tests.functional'" in upgrade:dumpling-du...
- Fixed typo https://github.com/ceph/ceph-qa-suite/pull/157
- 05:45 PM rgw Bug #9615: "ERROR: test suite for <module 's3tests.functional'" in upgrade:dumpling-dumpling-dist...
- interesting how is it possible if I only added one yaml file v0.67.11.yaml, will look
- 05:36 PM rgw Bug #9615: "ERROR: test suite for <module 's3tests.functional'" in upgrade:dumpling-dumpling-dist...
- this yaml has no 'rgw' task... that's why it gets connection refused.
- 06:11 PM rgw Bug #9616: upgrade test restarts rgw, test gets 500
- Even if we do somehow get it to retry (might require changes to the fastcgi module), we'll still get 500s from reques...
- 05:47 PM rgw Bug #9616: upgrade test restarts rgw, test gets 500
- Yehuda Sadeh wrote:
> Not sure what the test is doing exactly, but the 500 is because the rgw process was restarted ... - 05:36 PM rgw Bug #9616: upgrade test restarts rgw, test gets 500
- Not sure what the test is doing exactly, but the 500 is because the rgw process was restarted in the middle of the te...
- 05:52 PM rgw Bug #9612: "ERROR: test suite for <module 's3tests.functional'" in multi-version-giant-testing-ba...
- pls update with new test... this one was specifying firefly
- 05:45 PM rgw Bug #9169: 100-continue broken for centos/rhel
- maybe we are lacking the apache or mod_fastcgi packages here?
- 05:41 PM rgw Bug #9169: 100-continue broken for centos/rhel
- Yuri Weinstein wrote:
> Similar issue in suite:upgrade:firefly
>
> http://pulpito.front.sepia.ceph.com/teuthology... - 05:32 PM Bug #9617 (In Progress): objecter shutdown races with msg dispatch
- 04:21 PM Bug #9617: objecter shutdown races with msg dispatch
- ...
- 05:28 PM Feature #8960 (Resolved): filestore: store backend type persisently
- 05:24 PM Bug #9142 (Can't reproduce): [ RUN ] LibRadosTwoPoolsPP.PromoteSnapScrub hang
- 05:24 PM Bug #9141 (Can't reproduce): [ RUN ] LibRadosAio.IsCompletePP hang
- 04:49 PM Bug #6301: ceph-osd hung by XFS using linux 3.10
- fwiw, after upgrading the performance test nodes from Ubuntu 13.10 to Fedora Core 20, I appear to be hitting this und...
- 04:44 PM Feature #9580: ceph-disk, ceph-osd: make journal [partition] creation conditional based on osd_ob...
- Mark Kirkwood wrote:
> While we are thinking about this, note that some of the keyvalue backends have facility to ha... - 04:43 PM CephFS Bug #9341: MDS: very slow rejoin
- John Spray wrote:
> The userspace change and test for this are merged into master. Is the kernel side all done too?... - 01:07 PM CephFS Bug #9341: MDS: very slow rejoin
- The userspace change and test for this are merged into master. Is the kernel side all done too?
- 04:33 PM CephFS Bug #9514: ceph-fuse pjd test is failing in giant nightlies
- 03:49 PM CephFS Bug #9514: ceph-fuse pjd test is failing in giant nightlies
- So here's a question: why does the client (temporarily) remember its ctime as being 2014-09-26 19:22:06.889397, but n...
- 02:58 PM CephFS Bug #9514 (In Progress): ceph-fuse pjd test is failing in giant nightlies
- Hah, we got the failure with logs in /a/sage-2014-09-26_17:51:11-smoke-giant-distro-basic-multi/513914
All of the ... - 04:26 PM Bug #9614: PG stuck with remapped
- Thanks Loic for the following up.
After talking to other engineers, the backfilling seems like due to he removed O... - 12:32 PM Bug #9614: PG stuck with remapped
- It looks like you are on the right track :-)
- 12:23 PM Bug #9614: PG stuck with remapped
- ...
- 12:13 PM Bug #9614: PG stuck with remapped
- could you attach the full output of pg query 3.1ee7 please ? And also the ceph osd tree would help to get an idea why...
- 02:21 AM Bug #9614: PG stuck with remapped
- There are still two issues:
# Some PGs are stuck with active+remapped forever (for both replicated pool and EC pool)... - 02:07 AM Bug #9614: PG stuck with remapped
- Guang Yang wrote:
> Another observation is that even the pg dump result for such PG:
> [...]
>
> Even there is a... - 01:53 AM Bug #9614: PG stuck with remapped
- Attaching CRUSH / EC profile / OSD dump.
- 01:28 AM Bug #9614: PG stuck with remapped
- Loic Dachary wrote:
> [...]
> The *2147483647* here shows mapping failed. Is this something you expect ?
As there ... - 01:22 AM Bug #9614: PG stuck with remapped
- ...
- 04:24 PM Bug #9113: osd: snap trimming eats memory, linearly
- There's another piece. The trimmer is constantly requeueing.
- 02:02 PM Bug #9113 (Pending Backport): osd: snap trimming eats memory, linearly
- 01:59 PM Bug #9113 (Fix Under Review): osd: snap trimming eats memory, linearly
- 04:15 PM Bug #9620: tests: qa/workunits/cephtool/test.sh race condition
- I will verify the result when they are ready but I'm not too concerned ;-)
- 04:15 PM Bug #9620: tests: qa/workunits/cephtool/test.sh race condition
- 02:42 PM Bug #9620 (Resolved): tests: qa/workunits/cephtool/test.sh race condition
- i jumped the gun and merged, oops!
- 01:52 PM Bug #9620: tests: qa/workunits/cephtool/test.sh race condition
- gitbuilder running
- 01:51 PM Bug #9620 (Fix Under Review): tests: qa/workunits/cephtool/test.sh race condition
- https://github.com/ceph/ceph/pull/2603
- 08:18 AM Bug #9620 (Pending Backport): tests: qa/workunits/cephtool/test.sh race condition
- 04:53 AM Bug #9620 (Fix Under Review): tests: qa/workunits/cephtool/test.sh race condition
- https://github.com/ceph/ceph/pull/2594
- 04:36 AM Bug #9620: tests: qa/workunits/cephtool/test.sh race condition
- The *ceph osd thrash* command will randomly "mark osds down and up":https://github.com/ceph/ceph/blob/firefly/src/mon...
- 03:29 AM Bug #9620: tests: qa/workunits/cephtool/test.sh race condition
- The following sequence happens:
* ceph osd dump finds 3 osd "down"
* ceph osd dump finds no osd "down"
* ceph os... - 03:24 AM Bug #9620 (Resolved): tests: qa/workunits/cephtool/test.sh race condition
- "osd are marked down":https://github.com/ceph/ceph/blob/master/qa/workunits/cephtool/test.sh#L604 and a loop checking...
- 04:02 PM Revision f0bbbadd (ceph): mailmap: Janne Grunau affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 1a963484 (ceph): mailmap: Johnu George affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Johnu George <johnugeo@cisco.com> - 04:02 PM Revision dc601c79 (ceph): mailmap: Laurent Guerby affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision d0a22025 (ceph): mailmap: Volker Voigt affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 214e6cda (ceph): mailmap: Swami Reddy name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Swami Reddy <swamireddy@gmail.com> - 04:02 PM Revision 5d3c0255 (ceph): mailmap: Pavan Rallabhandi name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 75716656 (ceph): mailmap: Samuel Just name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 242f3344 (ceph): mailmap: John Spray name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 91af25f6 (ceph): mailmap: John Wilkins name normalization
- Also adding John Wilkins to .peoplemap to track org change
Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@g... - 04:02 PM Revision eedcd9cc (ceph): mailmap: Gerben Meijer affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Gerben Meijer <gerben@daybyday.nl> - 04:02 PM Revision 23c8fe52 (ceph): mailmap: Scott A. Brandt affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 26941eac (ceph): mailmap: Feng Wang affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 8d9fcbca (ceph): mailmap: Casey Marshall affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 984a1476 (ceph): mailmap: Riccardo Ferretti affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision e6bba0d5 (ceph): mailmap: Roman Haritonov affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision b386b598 (ceph): mailmap: JuanJose Galvez affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision 4b33b584 (ceph): mailmap: Jiang Heng affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision b167f704 (ceph): mailmap: Sahid Ferdjaoui affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Sahid Orentino Ferdjaoui <sahid.ferdj... - 04:02 PM Revision 29efada5 (ceph): mailmap: Cheng Cheng affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:02 PM Revision f8179353 (ceph): mailmap: Daniel Schepler affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:01 PM Revision e133a92e (ceph): mailmap: Yongyue Sun affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 03:46 PM RADOS Bug #9606: mon: ambiguous error_status returned to user when type is wrong in a command
- This may be easier if/when ceph_argparse gets made into a proper Python package; I hear there is renewed interest in ...
- 03:18 PM Revision 9af9df42 (ceph): Merge pull request #2594 from dachary/wip-9620-test-mon-thrash
- qa/workunits/cephtool/test.sh: fix thrash (ultimate)
Reviewed-by: Sage Weil <sage@redhat.com> - 02:57 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- Exploring the idea that maybe the buffers pointed to by the iovec are overriden, mixed up
- 08:28 AM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- Reading the buffer.{h,cc} code it looks like the caller is protected from a situation where a bufferptr leftover can ...
- 02:53 PM Bug #9626 (Resolved): PG: cancel backfill reservations if we get a cancel during backfill
- 02:36 PM Fix #9566: osd: prioritize recovery of OSDs with most work to do
- Factor the number of backfill (or backfill_wait) pgs on the OSD into the recovery priority. Make sure this accounts ...
- 02:14 PM Bug #9574 (Pending Backport): Backfill: recheck full status once reservation is granted
- 01:51 PM Bug #9574 (Fix Under Review): Backfill: recheck full status once reservation is granted
- 02:04 PM Revision 70bcda54 (ceph): suites: enable mds_client_limits test
- Requires same cluster etc as 'recovery', though
we could now do with better name than that.
Signed-off-by: John Spra... - 02:04 PM Revision c2d298a4 (ceph): tasks: wait for mds active before mounting clients
- To make the logs clearer when trying to work out
if/when something went wrong, rather than always
having client logs ... - 02:04 PM Revision 0073e25d (ceph): tasks: rename FuseMount.get_client_id to get_global_id
- 'client_id' was ambiguous because in other places it
meant the '0' in client.0, whereas here it means
the runtime-gen... - 02:04 PM Revision b77b3bec (ceph): tasks: add mds_client_limits
- New CephFS tests for the behaviour of the system while
enforcing its resource limits.
Signed-off-by: John Spray <joh... - 02:04 PM Revision 1fa15011 (ceph): tasks: generalise CephFSTestCase
- Some of this stuff could be even more general for embedding
unittest-style suites, but for the moment let's keep the ... - 02:03 PM Revision b6ccf0d4 (ceph): tasks: generalize config writing for Filesystem
- Signed-off-by: John Spray <john.spray@redhat.com>
- 02:00 PM Bug #9388: osd/PG.cc: 2945: FAILED assert(r == 0) in update_snap_map
- This is the one with the import/export racing with split
- 01:59 PM Bug #9503 (Fix Under Review): Dumpling: removing many snapshots in a short time makes OSDs go ber...
- 01:54 PM Bug #9545 (Resolved): filestore stuck in journal->should_commit_now() loop on shutdown
- 01:52 PM Bug #8629 (Pending Backport): cache_evict needs to prevent make_writeable from creating a snapdir
- 01:45 PM Bug #9480 (Resolved): OSD is crashing while object deletion
- 01:30 PM Bug #9625: firefly: memory corruption
- /a/samuelj-2014-09-23_14:40:50-rados-firefly-wip-testing-old-vanilla-basic-multi/507058 another example
- 10:44 AM Bug #9625: firefly: memory corruption
- ubuntu@teuthology:/a/sage-2014-09-27_20:55:12-rados-firefly-distro-basic-multi/515818
ubuntu@teuthology:/a/sage-2014... - 10:43 AM Bug #9625 (Resolved): firefly: memory corruption
- I am guessing that these two coredumps are related.
#0 0x00007f1918142f07 in _dl_map_object_deps (map=map@entry=0... - 01:15 PM CephFS Bug #8576: teuthology: nfs tests failing on umount
- Trying the sync on Sage's go-ahead. :)
commit:56223ce98b659fe7b25b55161ef8163495f438fc in teuthology. - 10:45 AM CephFS Bug #8576: teuthology: nfs tests failing on umount
- Is there any chance that just running a sync on the node prior to trying to "exportfs -au" might prevent this? I'm he...
- 12:51 PM devops Fix #9017 (Fix Under Review): [paddles] implement validation across all controller methods
- Pull request opened https://github.com/ceph/paddles/pull/46
- 12:27 PM Revision 584a54a0 (ceph): Merge pull request #2582 from ceph/wip-crush-location
- crushtool: add --location <id> command
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 11:47 AM Revision beade63a (ceph): qa/workunits/cephtool/test.sh: fix thrash (ultimate)
- Keep the osd trash test to ensure it is a valid command but make it a
noop by giving it a zero argument (meaning thra... - 10:30 AM Bug #9623 (Won't Fix): On cluster with 3 mons, stopping 2 mons made cluster in-accessible, with I...
- This is expected and intended behavior. The monitors are a Paxos system and require a quorum of *more than* half to b...
- 07:30 AM Bug #9623: On cluster with 3 mons, stopping 2 mons made cluster in-accessible, with IO's hung/pause
- Removing myself as I may not have time to deal with this right now.
- 06:25 AM Bug #9623 (Won't Fix): On cluster with 3 mons, stopping 2 mons made cluster in-accessible, with I...
- Cluster with "n" number of monitor nodes, will be in-accessible if "n-1" number of monitors are down.
Its been obser... - 09:55 AM devops Bug #6461 (Rejected): ceph-deploy should at least issue a warning if there are parser errors read...
- `ConfigParser` will not have errors reading a config file that has duplicate sections.
In Python2.X a duplicate se... - 08:25 AM Bug #9613 (Duplicate): "Segmentation fault" in upgrade:dumpling-giant-x:parallel-giant-distro-bas...
- #9582
- 07:50 AM devops Bug #6489 (Can't reproduce): ceph-deploy: get_nonlocal_ip() should filter ipv6 addrs
- 07:44 AM devops Bug #7483 (Rejected): ceph-deploy should fetch keyrings always
- There isn't a reasonable way to implement this. The use case is deploying to a new node and having stale files in the...
- 06:38 AM Revision 9ced1378 (ceph): Merge pull request #2590 from dachary/wip-9592-librados-large-object
- librados large object early check
Reviewed-by: Sage Weil <sage@redhat.com> - 06:35 AM Revision 5d1d9dba (ceph): librados: cap the IoCtxImpl::{aio_}*{write,append} buffer length
- If the value of the len parameter is greater than UINT_MAX/2,
IoCtxImpl::aio_write, IoCtxImpl::aio_write_full, IoCtxI... - 06:00 AM Bug #9408: erasure-code: misalignment
- Running under the branch wip-9408-buffer-alignment in http://ceph.com/gitbuilder.cgi
- 05:58 AM Bug #9408: erasure-code: misalignment
- New pull request https://github.com/ceph/ceph/pull/2595
- 03:11 AM Revision 5f029ff4 (ceph): os/FileStore: using FIEMAP_FLAGS_SYNC instead of fsync() before call
- fiemap.
When call fiemap, it need sync the file. Now it used fsync() to achieve
this. But for fiemap, there is a fla... - 02:06 AM Bug #9572: erasure-code: BlaumRoth default encoding regression
- Brute force check of w=7 with all possible values for k prove it allows recovering all scenarios. ...
- 01:53 AM rbd Bug #9391: fio rbd driver rewrites same blocks
- Could you provide your fio job file / config to verify the issue?
- 12:46 AM Revision 4db51bb8 (ceph): Merge pull request #2400 from majianpeng/fix2
- osd: Make RPGTransaction::get_bytes_written return the correct size.
Reviewed-by: Sage Weil <sage@redhat.com>
09/28/2014
- 11:38 PM Bug #9592 (Resolved): librados: Not able to create Large Files with Librados
- 03:46 PM Bug #9592: librados: Not able to create Large Files with Librados
- Extend the checks to librados.hpp and aio_* https://github.com/ceph/ceph/pull/2590
- 11:30 PM Bug #9304 (Resolved): pool create with invalid crush rule name succeeds
- 11:02 PM Bug #6003: journal Unable to read past sequence 406 ...
- ...
- 09:26 PM Fix #9566: osd: prioritize recovery of OSDs with most work to do
- The recovery slows simply because there are fewer PGs left degraded and the per-pg (or per-osd) recovery rate is limi...
- 03:52 PM Revision 8023478e (ceph): Merge pull request #154 from ceph/wip_fix_multi_version
- Fixing branch assumptions
- 12:42 PM Bug #9619 (Can't reproduce): excessive mon memory usage when rbd rm 1PB
- Steps to reproduce:
* create a 1 peta byte rbd image
* remove the image
the mon memory usage will grow over 10GB - 12:37 PM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- Also in http://pulpito.front.sepia.ceph.com/teuthology-2014-09-28_08:42:11-upgrade:dumpling-firefly-giant:parallel-gi...
- 12:32 PM Bug #9618 (Won't Fix): kernel 3.14 in Debian Jessie : XFS bug
- For the record: the 3.14 kernel that was (until today) the default for Debian Jessie exhibited the following XFS bug ...
- 08:38 AM Revision 69fc1717 (ceph): Clock: Using clock_gettime instead of gettimeofday in linux os.
- The precision of utime_t is nsec. In func ceph_clock_now, it use
gettimeofday to set utime_t. But the precision of ge... - 08:37 AM Bug #9617 (Resolved): objecter shutdown races with msg dispatch
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-27_19:10:02-upgrade:firefly-giant-x:parallel-giant...
- 08:37 AM Revision becc1140 (ceph): librados: test s/E2BIG/TooBig/
- Because E2BIG does not allow selection
./ceph_test_rados_api_aio --gtest_filter=LibRadosAio.E2BIG
Running main() fro... - 08:26 AM Revision 32195f94 (ceph): librados: cap the rados_aio_*{write,append} buffer length
- If the value of the len parameter is greater than UINT_MAX/2,
rados_aio_write, rados_aio_write_full and rados_aio_app... - 08:12 AM Bug #9515 (New): "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:parall...
- Still see in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-27_18:40:01-upgrade:dumpling-giant-x:parallel-gia...
- 07:57 AM rgw Bug #9615: "ERROR: test suite for <module 's3tests.functional'" in upgrade:dumpling-dumpling-dist...
- Appears to be only on @1-dumpling-install/v0.67.11.yaml@
- 07:48 AM rgw Bug #9615 (Resolved): "ERROR: test suite for <module 's3tests.functional'" in upgrade:dumpling-du...
- In http://pulpito.front.sepia.ceph.com/teuthology-2014-09-27_18:45:01-upgrade:dumpling-dumpling-distro-basic-vps/ run...
- 07:52 AM rgw Bug #9616 (Resolved): upgrade test restarts rgw, test gets 500
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-27_18:45:01-upgrade:dumpling-dumpling-distro-basic...
- 07:01 AM Revision f777fc6e (ceph): osd: Make RPGTransaction::get_bytes_written return the correct size.
- It record size larger than clien wrote. It should like
ECTransaction::get_bytes_written only return the size which cl... - 04:29 AM Bug #9614: PG stuck with remapped
- Another observation is that even the pg dump result for such PG:...
- 03:45 AM Bug #9614 (Resolved): PG stuck with remapped
- In our pre-production cluster, we observed that the cluster starts backfilling even with OSD noout flag set when ther...
09/27/2014
- 10:48 PM rbd Bug #9595: librbd: internal methods can operate on extra objects when non-default striping is used
- https://github.com/ceph/ceph/pull/2588
- 10:48 PM rbd Bug #9595 (Fix Under Review): librbd: internal methods can operate on extra objects when non-defa...
- 10:03 PM Revision 62fa6b19 (ceph): Fixing branch assumptions
- Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
- 06:37 PM Revision 503f865d (ceph): osd/ReplicatedPG: respect RWORDERED rados flag
- If this flag is set, we need to order reads as writes. In particular, this
means that reads will wait for degraded o... - 04:42 PM Bug #9613: "Segmentation fault" in upgrade:dumpling-giant-x:parallel-giant-distro-basic-multi run
- Looks similar to #9508
- 04:40 PM Bug #9613 (Duplicate): "Segmentation fault" in upgrade:dumpling-giant-x:parallel-giant-distro-bas...
- Two failures in http://pulpito.front.sepia.ceph.com/teuthology-2014-09-26_18:44:02-upgrade:dumpling-giant-x:parallel-...
- 02:51 PM Revision 7849d792 (ceph): crushtool: add --show-location <id> command
- Include some tests.
Signed-off-by: Sage Weil <sage@redhat.com> - 01:17 PM Revision b68811ce (ceph): add autotools-generated files to .gitignore
- When running "make check", Automake generates test-suite.log, along with
various *.log and *.trs files in the tree. A... - 12:58 PM Revision 8badd5a4 (ceph): Merge pull request #2584 from dachary/wip-9592-librados-large-object
- librados: cap the rados*{write,append} buffer length
Reviewed-by: Sage Weil <sage@redhat.com> - 12:52 PM Revision c05bf10f (ceph): Merge pull request #2585 from dachary/wip-7648-get-full-location-emperor
- simpler get full location (emperor)
Reviewed-by: Sage Weil <sage@redhat.com> - 11:40 AM rgw Bug #9612: "ERROR: test suite for <module 's3tests.functional'" in multi-version-giant-testing-ba...
- i suspect the giant rgw won't work with firefly osds?
- 08:56 AM rgw Bug #9612 (Rejected): "ERROR: test suite for <module 's3tests.functional'" in multi-version-giant...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-26_23:20:01-multi-version-giant-testing-basic-mult...
- 11:39 AM Bug #9610 (Resolved): Crash "RadosModel.h: In function 'virtual void WriteOp::_finish(TestOp::Cal...
- pushed fix to dumpling branch, commit:503f865d6432bead72aac0ffba0539d807f078c4
- 08:33 AM Bug #9610: Crash "RadosModel.h: In function 'virtual void WriteOp::_finish(TestOp::CallbackInfo*)...
- Another similar crash in job http://pulpito.front.sepia.ceph.com/teuthology-2014-09-26_23:20:01-multi-version-giant-t...
- 08:29 AM Bug #9610 (Resolved): Crash "RadosModel.h: In function 'virtual void WriteOp::_finish(TestOp::Cal...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-26_23:20:01-multi-version-giant-testing-basic-mult...
- 11:36 AM devops Bug #9611 (Rejected): Missing packages in multi-version-giant-testing-basic-multi
- Doesn't look like a 'next' branch exists any longer so no way to fix this.
- 08:52 AM devops Bug #9611: Missing packages in multi-version-giant-testing-basic-multi
- In run http://pulpito.front.sepia.ceph.com/teuthology-2014-09-26_23:20:01-multi-version-giant-testing-basic-multi/
... - 08:50 AM devops Bug #9611 (Rejected): Missing packages in multi-version-giant-testing-basic-multi
- 09:28 AM Revision c5298fba (ceph): crush: fix get_full_location_ordered
- This should return -ENOENT when an id is not present. Broken by
746069ee62c74ecf04ed45988029d5c3382a38d2.
Signed-of... - 09:28 AM Revision ed1a8014 (ceph): crush/CrushWrapper: simplify get_full_location_ordered()
- Just ascend the hierarchy; it is much less complicated.
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked f... - 09:19 AM Revision c70331db (ceph): Merge pull request #2583 from ceph/wip-7648
- crush: backport newer get_full_location
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 09:16 AM Bug #9592: librados: Not able to create Large Files with Librados
- Looking at librados.hpp
- 01:57 AM Bug #9592 (Fix Under Review): librados: Not able to create Large Files with Librados
- https://github.com/ceph/ceph/pull/2584 should be enough. Unless there is a good reason to write an object with chunks...
- 08:48 AM Revision 33501d24 (ceph): librados: cap the rados*{write,append} buffer length
- When the caller submits a payload that will end up being rejected with
rados.Error: Ioctx.write(rbd): failed to wr... - 05:53 AM Bug #7648 (Resolved): ceph-mon corner case denial of service
- 02:32 AM Bug #7648 (Fix Under Review): ceph-mon corner case denial of service
- emperor backport https://github.com/ceph/ceph/pull/2585
- 02:22 AM Bug #7648 (Pending Backport): ceph-mon corner case denial of service
- the backport needs to be on emperor also
- 04:27 AM RADOS Bug #9606: mon: ambiguous error_status returned to user when type is wrong in a command
- ceph.in "uses ceph_argparse":https://github.com/ceph/ceph/blob/giant/src/ceph.in#L67 to validate the arguments client...
- 01:09 AM Revision c1d9d8c5 (ceph): crush/CrushWrapper: simplify get_full_location_ordered()
- Just ascend the hierarchy; it is much less complicated.
Signed-off-by: Sage Weil <sage@inktank.com>
(cherry picked f... - 01:09 AM Revision 1d3c0c6e (ceph): crush: fix get_full_location_ordered
- This should return -ENOENT when an id is not present. Broken by
746069ee62c74ecf04ed45988029d5c3382a38d2.
Signed-of... - 12:49 AM Revision 3f05fbf5 (ceph): Merge pull request #2580 from cernceph/wip-scientific
- ceph-disk: add Scientific Linux as a Redhat clone
Reviewed-by: Sage Weil <sage@redhat.com> - 12:46 AM Revision f8ac2248 (ceph): ceph-disk: add Scientific Linux as a Redhat clone
- Scientific Linux is a RHEL clone and needs to use partx.
Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>... - 12:12 AM Revision 67a3608b (ceph): Merge pull request #153 from ceph/wip-9511-wusui
- Use the correct versions of existing tests
- 12:07 AM Revision e68f6578 (ceph): Use the correct versions of existing tests
- The test-upgrade-firefly.sh script is used from the dumpling suite
Fixes: 9511
Signed-off-by: Warren Usui <warren.us... - 12:02 AM RADOS Bug #9492 (Need More Info): Crush Mapper crashes when number of replicas is less than total numbe...
- What happens with indep ?
09/26/2014
- 11:44 PM Revision dc9d5294 (ceph): Merge pull request #152 from dachary/wip-9608-ec-default-parallel-giant
- ec-rados-default is not fit for dumpling parallel upgrade
- 11:28 PM Revision 5b8fa621 (ceph): ec-rados-default is not fit for dumpling parallel upgrade
- Because it may race with the upgrade and end up being unable to perform.
http://tracker.ceph.com/issues/9608 Fixes: ... - 11:27 PM Revision fa72591c (ceph): Merge pull request #151 from dachary/wip-9608-ec-default-parallel
- ec-rados-default is not fit for dumpling parallel upgrade
- 11:23 PM Revision b77ef02b (ceph): ec-rados-default is not fit for dumpling parallel upgrade
- Because it may race with the upgrade and end up being unable to perform.
http://tracker.ceph.com/issues/9608 Fixes: ... - 10:25 PM Revision 9c3e01a2 (ceph): Merge pull request #2568 from johnugeorge/wip-9492-crush
- Crush: Ensuring at most num-rep osds are selected for any rule
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 08:04 PM Revision 6b4d1aa9 (ceph): Crush: Ensuring at most num-rep osds are selected
- Crush temporary buffers are allocated as per replica size configured
by the user.When there are more final osds (to b... - 07:13 PM rgw Bug #9588: Keystone s3 auth integration lacking access_key = tenant:user ability supported by swi...
- So, actually talking to a swift s3 proxy with:
access_key = 'demo:demo'
secret_key = 'password'
results in:
... - 06:41 PM Revision 24a52127 (ceph): Merge pull request #135 from ceph/wip_9398
- First draft of firefly-giant-x suite
- 06:14 PM Bug #7648 (Fix Under Review): ceph-mon corner case denial of service
- https://github.com/ceph/ceph/pull/2583
- 08:49 AM Bug #7648 (In Progress): ceph-mon corner case denial of service
- works for any osd that exists but is not in the crush map, it seems
- 05:53 PM Bug #9570 (In Progress): osd crash in FileJournal::WriteFinisher::entry() aio
- 04:01 PM Revision 3ea4b1dc (ceph): smoke: fix pjd debug conf syntax
- Needs the section there or else ceph.py barfs at runtime
Signed-off-by: Sage Weil <sage@redhat.com> - 04:01 PM Revision 5ca7ea5b (ceph): ceph-disk: add Scientific Linux as a Redhat clone
- Scientific Linux is a RHEL clone and needs to use partx.
Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch> - 03:32 PM CephFS Bug #8427: ceph-fuse: Dumpling "cache still has 0+1 items, waiting (for caps to release?)" on shu...
- Sage believes this is a bug with readahead that got fixed in subsequent releases.
- 06:51 AM CephFS Bug #8427 (Won't Fix): ceph-fuse: Dumpling "cache still has 0+1 items, waiting (for caps to relea...
- 03:26 PM RADOS Bug #9492 (Resolved): Crush Mapper crashes when number of replicas is less than total number of o...
- 01:56 PM RADOS Bug #9492: Crush Mapper crashes when number of replicas is less than total number of osds to be s...
- Ran valgrind with the patch and no errors were found with different rule combinations of num_rep and number of osds t...
- 12:44 PM Bug #9417: "Segmentation fault" in upgrade:dumpling-giant-x-master-distro-basic-vps run
- Same issue in job http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-26_10:44:24-upgrade:dumpling-giant-x:paralle...
- 12:03 PM devops Bug #9607 (Resolved): wrong epel-release version present in misc-ceph repo
- That epel release 7 RPM should not have ever been put in that repo. It is removed and to its correct location and cep...
- 11:31 AM devops Bug #9607 (Resolved): wrong epel-release version present in misc-ceph repo
- In a CentOS 6 box where we run `yum install epel-release` it now sees that it needs to update to use the epel-release...
- 11:35 AM devops Bug #9603: No package ceph-debuginfo-0.67.10 available in upgrade:dumpling-firefly-x-giant-distro...
- Would be helpful to include:...
- 11:08 AM devops Bug #9603: No package ceph-debuginfo-0.67.10 available in upgrade:dumpling-firefly-x-giant-distro...
- Same issue in suite:upgrade:dumpling-giant-x
http://pulpito.front.sepia.ceph.com/teuthology-2014-09-26_10:44:24-up... - 08:12 AM devops Bug #9603 (Rejected): No package ceph-debuginfo-0.67.10 available in upgrade:dumpling-firefly-x-g...
- In run http://pulpito.front.sepia.ceph.com/teuthology-2014-09-25_19:25:02-upgrade:dumpling-firefly-x-giant-distro-bas...
- 11:34 AM rbd Feature #2466 (Resolved): librbd: add invalidate_cache function to interface
- This was added a while back in commit:5d340d26dd70192eb0e4f3f240e3433fb9a24154
- 11:18 AM RADOS Bug #9606 (New): mon: ambiguous error_status returned to user when type is wrong in a command
- ...
- 10:57 AM Revision 83fb32ca (ceph): Merge pull request #2572 from ceph/wip-9562
- osdc/Filer: drop probe/purge locks before calling objecter
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: ... - 10:10 AM Bug #9592: librados: Not able to create Large Files with Librados
- Nice catch Pavan Rallabhandi ;-) I had trouble reproducing the problem because I forgot the "LD_LIBRARY_PATH=.libs" ...
- 09:37 AM Bug #9592: librados: Not able to create Large Files with Librados
- The minimal script...
- 08:05 AM Bug #9592 (In Progress): librados: Not able to create Large Files with Librados
- 04:44 AM Bug #9592: librados: Not able to create Large Files with Librados
- ...
- 10:01 AM devops Bug #9548 (Rejected): ceph mon creation failed for centOS
- 09:39 AM Feature #9302 (Fix Under Review): mon: 'ceph osd pool ls' command
- https://github.com/ceph/ceph/pull/2581
- 09:34 AM devops Bug #9232: disk zap doesnt remove the dmcrypt settings on disk
- I think that `disk zap` would certainly have to clear the dmcrypt flags in the disk.
Can you make sure that it doe... - 08:56 AM rgw Bug #9605 (Won't Fix): rgw: need to have shadow objects named after head object
- 08:55 AM rgw Feature #9604 (Resolved): rgw: create a tool for orphaned objects cleanup
- 08:07 AM devops Bug #9567 (New): Missing packages in upgrade:dumpling-firefly-x-giant-distro-basic-vps run
- Still see in today's run:
http://pulpito.front.sepia.ceph.com/teuthology-2014-09-25_19:25:02-upgrade:dumpling-fire... - 07:41 AM Fix #9601: erasure-code: ErasureCode::encode overhead is too high
- The handling got more complicated due to the updated padding handling.
It's a bit little faster. jerasure_matrix_e... - 05:17 AM Fix #9601: erasure-code: ErasureCode::encode overhead is too high
- The overhead has shifted but looks globaly the same with https://github.com/ceph/ceph/pull/2558
!{width: 100%}jannau... - 03:52 AM Fix #9601: erasure-code: ErasureCode::encode overhead is too high
- Applying https://github.com/ceph/ceph/pull/2558 and benchmarking again
- 03:34 AM Fix #9601 (New): erasure-code: ErasureCode::encode overhead is too high
- When encoding 4KB buffers it is ~15% of the total CPU being used although it is only preparing the buffers.
!{width:... - 05:26 AM rbd Bug #9602 (Closed): rbd export -> nc ->rbd import = memory leak
- I see a memory leak when importing raw devi?e.
Export Scheme:
[rbd@rbdbackup ~]$ rbd --no-progress -n client.rbdb... - 01:58 AM Cleanup #9600 (New): rework bufferlist::*aligned* functions
- The align function should allow 32 byte alignment (for SIMD instructions) or page alignment (for I/O). There should b...
- 01:00 AM Bug #8592: sgdisk no longer likes `--change-name` when creating partitions
- I have fixed this by add --zap-disk option, hope this will help you.
- 12:44 AM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- Thanks for explaining. Since alloc hint is optional it does not matter if it is activated and deactivate later.
- 12:22 AM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Please disregard #15. I just fell victim to inaccurate documentation about the @incomplete@ PG state.
-Sam's hunch... - 12:02 AM Revision 0a6f6a49 (ceph): Merge pull request #2575 from ceph/wip-zafman-cleanup
- osd: Remove unused PG functions queue_notify(), queue_info(), queue_log(...
Reviewed-by: Loic Dachary <loic@dachary....
09/25/2014
- 11:59 PM Bug #9592: librados: Not able to create Large Files with Librados
- A modified script to debug this issue:-
####################################
import rados
import sys
try:
cluste... - 10:02 AM Bug #9592: librados: Not able to create Large Files with Librados
- If I were to guess, something in the stack is converting the size value down to an int32 and then back up to int64, s...
- 09:51 AM Bug #9592 (Can't reproduce): librados: Not able to create Large Files with Librados
- ...
- 06:28 AM Bug #9592 (Resolved): librados: Not able to create Large Files with Librados
- I find this issue while i was trying to run a 1GB Write Cosbench Workload using librados.(My 1MB write & read run was...
- 11:15 PM Revision 7827e003 (ceph): os: io_event.res is the size written
- And not an error code to be converted with cpp_strerror()
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 11:08 PM rgw Bug #9588: Keystone s3 auth integration lacking access_key = tenant:user ability supported by swi...
- Despite asking for swift I am actually getting the nova object store doing the s3 stuff it seems. I'll comment gaian ...
- 10:16 PM rgw Bug #9588: Keystone s3 auth integration lacking access_key = tenant:user ability supported by swi...
- Hmm - maybe not tested enough, as it looks like the way devstack sets up the swift s3 layer is a bit screwy, and almo...
- 10:19 PM Revision b8562959 (ceph): Merge pull request #2524 from ceph/wip-5768
- rbd-fuse: Fix memory leak in enumerate_images
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 09:05 PM Revision 5c2984e6 (ceph): Merge pull request #2531 from dachary/wip-9536-isa-alignment
- erasure-code: isa plugin alignment fixes
Reviewed-by: Sage Weil <sage@redhat.com> - 08:51 PM Revision d851c3f2 (ceph): osd: improve debug output for do_{notifies,queries,infos}
- Hunting #9389
Signed-off-by: Sage Weil <sage@redhat.com> - 08:01 PM Revision 2ba5ed57 (ceph): Merge pull request #2540 from ceph/wip-giant-messenger-fixes
- giant messenger fixes
Reviewed-by: Sage Weil <sage@redhat.com> - 07:45 PM Documentation #9542: Error link:"Ceph Object Gateway"->"Manual Install"
- I know it "*is the way the doc is generated*", and I know "*it's not a bug in a link*",too.(Guess it's Sphinx?). But ...
- 09:39 AM Documentation #9542 (Won't Fix): Error link:"Ceph Object Gateway"->"Manual Install"
- This is the way the doc is generated, it's not a bug in a link. And it actually makes more logical sense to jump from...
- 07:34 PM Revision 126d0b30 (ceph): osdc/Objecter: only post_rx_buffer if no op timeout
- If we post an rx buffer and there is a timeout, the revocation can happen
while the reader has consumed the buffers b... - 06:23 PM CephFS Feature #541 (Resolved): mds: tempsync
- this is implemented... TSYN and related states
- 06:21 PM Feature #1092 (Rejected): mon: checkpointing
- 06:19 PM Feature #131 (Resolved): bring wireshark plugin is up to date
- 05:54 PM Revision 00a504fb (ceph): rados/singleton-nomsgr/all/objectstore: clean up
- ceph_test_objectstore scribbles in $CWD. Keep it inside $TESTDIR and
clean up when we are done.
Fixes: #8395
Signed... - 05:47 PM CephFS Feature #630 (Resolved): release caps on inodes unlinked by other clients
- 05:47 PM CephFS Feature #630: release caps on inodes unlinked by other clients
- dup of #5039. already fixed by commit f8a947d92 client: trim deleted inode
- 05:30 PM Revision b07abf5b (ceph): suites/fs: update to latest mdtest
- They appear to have (accidentally?) fixed whatever
was crashing.
Fixes: #9177
Signed-off-by: John Spray <john.spray... - 05:08 PM Revision 1fafd6bf (ceph): Merge pull request #2576 from ceph/wip-9593
- osd/ReplicatedPG: fix objecter locking in start_flush
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 05:03 PM Feature #9568 (Resolved): Add test case to test #9419 (ceph wip-9419)
- 05:02 PM Revision 71005c1b (ceph): osd/ReplicatedPG: fix objecter locking in start_flush
- Broken in backport fd96eb62ece27f5c660429584c2ff2e058bc6e94.
Signed-off-by: Sage Weil <sage@redhat.com> - 04:34 PM CephFS Bug #9514: ceph-fuse pjd test is failing in giant nightlies
- This hasn't reproduced since we turned on debug logging. :(
But I did see it on a run without any logging: /a/gregf-... - 04:26 PM Revision 0115a55a (ceph): Merge pull request #2574 from ceph/wip-msgr-shutdown
- msg: allow calling dtor immediately after ctor
Reviewed-by: Sage Weil <sage@redhat.com> - 04:09 PM Feature #9580: ceph-disk, ceph-osd: make journal [partition] creation conditional based on osd_ob...
- While we are thinking about this, note that some of the keyvalue backends have facility to have their "wal" aka journ...
- 04:05 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- I don't see how it could be related to a problem in align_bl or bufferlist::rebuild_align. The worst these could do i...
- 03:28 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- Maybe "iterating the bufferptr":https://github.com/ceph/ceph/blob/dumpling/src/os/FileJournal.cc#L1297 can return buf...
- 03:03 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- ...
- 02:43 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- Sheldon, could you upload the full log somewhere if you still have it ?
- 01:53 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- * align_bl related pull request https://github.com/ceph/ceph/pull/2501
* rebuild_align fix (back from 2013) https:/... - 04:05 PM Revision ba02a5e6 (ceph): erasure-code: test isa encode/decode with various object sizes
- Create an encode_decode() helper method to be called from the
encode_decode test function with various object size ar... - 04:04 PM Revision eb8fdfa4 (ceph): erasure-code: add test for isa chunk_size method
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 04:01 PM Revision 7a468f35 (ceph): msg: allow calling dtor immediately after ctor
- Asserting on reaper_stop only made sense if the
messenger had ever been started: as it stood,
one couldn't create and... - 03:39 PM Revision 668c3527 (ceph): erasure-code: do not use typed tests for isa
- Because there only is one type.
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 03:39 PM Revision ed77178e (ceph): erasure-code: run isa tests via libtool and valgrind
- Because running valgrind with no libtool does not test the binary but
the enclosing shell script.
Signed-off-by: Loi... - 03:39 PM Revision aa9d70be (ceph): erasure-code: isa test compare chunks with memcmp instead of strncmp
- Because they may contain null characters.
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 03:39 PM Revision af07d29e (ceph): erasure-code: isa encode tests adapted to per chunk alignment
- The encode tests use the alignment constraints. It has been changed to
be aligned on a per chunk basis instead of com... - 03:39 PM Revision 28c2b6e4 (ceph): erasure-code: isa uses per chunk alignment constraints
- Copy code from the jerasure plugin to enforce alignment constraints per
chunk instead of using the total object size.... - 03:37 PM Revision 6f4909ae (ceph): erasure-code: [ISA] modify get_alignment function to imply a platform/c...
- 02:17 PM Revision 75525712 (ceph): doc/release-notes: v0.67.11
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:14 PM Bug #9203 (In Progress): ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < l...
- 02:14 PM Feature #9598 (Resolved): re-enable Objecter fast dispatch
- We had to nix fast dispatch on the Objecter because it could deadlock in conjunction with mark_down() calls.
Fixin... - 02:06 PM Bug #9536: erasure-code: ISA plugin alignment must be constant
- (parts of) this will need to be backported with the rest of the ISA plugin stuff
- 02:05 PM Bug #9536 (Pending Backport): erasure-code: ISA plugin alignment must be constant
- 01:52 PM Bug #9389 (Need More Info): ec pg stuck peering, did not send query for one shard
- commit:d851c3f2338e8d17dfd78d631b9f7977365356aa adds better debug output (and cleans up a bit)
- 01:37 PM Revision 0124d8ee (ceph): Merge pull request #2571 from dachary/wip-9579-isa-documentation
- documentation: erasure-code plugin isa does not require k/m
Reviewed-by: Andreas Peters <andreas.joachim.peters@cern... - 01:21 PM rbd Bug #9595 (Resolved): librbd: internal methods can operate on extra objects when non-default stri...
- ...
- 01:04 PM Bug #9295 (Resolved): osd/OSD.cc: 5501: FAILED assert(session) in ms_fast_dispatch
- 01:03 PM Bug #9295 (Duplicate): osd/OSD.cc: 5501: FAILED assert(session) in ms_fast_dispatch
- 01:03 PM Bug #9295: osd/OSD.cc: 5501: FAILED assert(session) in ms_fast_dispatch
- dup of #9462
- 01:01 PM Bug #9462 (Resolved): msgr deadlock: osd reply vs mark_down vs fault
- 12:53 PM Revision 8dc94a2d (ceph): osdc/Filer: drop probe/purge locks before calling objecter
- Fixes: #9562
Signed-off-by: John Spray <john.spray@redhat.com> - 12:44 PM Bug #8910 (Duplicate): ceph_test_objectstore: ObjectStore/StoreTest.ManyObjectTest/0 failure on f...
- pretty sure this is a dup of #8395
- 12:27 PM Bug #9582: librados: segmentation fault on timeout
- i'm going to see if we can just skip the rx_buffers zero-copy paths when a timeout is present
- 12:20 PM Bug #9388: osd/PG.cc: 2945: FAILED assert(r == 0) in update_snap_map
- import/export related
- 11:44 AM Bug #9571 (Resolved): rocksdb testing with powercycling fails on trusty
- this was an issue with the code fix and not a product bug.
resolved now. - 11:14 AM Bug #9419: dumpling->firefly upgrade, sending setallochint?
On any change of pg configuration peering happens, so a new collection of feature bits from the peers is collected....- 10:37 AM Bug #9419 (Fix Under Review): dumpling->firefly upgrade, sending setallochint?
- 12:46 AM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- What happens if
* all OSDs in a PG support setallochint
* one secondary OSD goes down
* the secondary is replac... - 11:09 AM Bug #8395: ceph-test-objectstore doesn't clean up
- backported to firefly branch
- 11:08 AM Feature #9594 (New): stop backfill when osd becomes too full
- We will currently refuse the reservation, but we don't actually stop backfill once it is started.
- 10:45 AM Bug #9480: OSD is crashing while object deletion
- 10:42 AM Bug #9390 (In Progress): EEXIST on split due to import/export
- 10:38 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- 10:37 AM Bug #9584: OpTracker segfault on shutdown (firefly)
- shutdown race is not so important
- 10:28 AM Revision 8f49a7d8 (ceph): tasks: wait for active after mds restart
- May have been causing spurious failures on
trying to read session state after MDS restart (
session list isn't popula... - 10:17 AM devops Tasks #8366: Update ceph.com/docs to default to the latest major release (0.80)
- John Wilkins wrote:
> We need to review this a bit further. Pointing to the latest major release is fine, but we nee... - 10:09 AM Bug #9593 (Resolved): osdc/Objecter.cc: 1225: FAILED assert(client_lock.is_locked()) (firefly)
- 10:02 AM Bug #9593 (Fix Under Review): osdc/Objecter.cc: 1225: FAILED assert(client_lock.is_locked()) (fir...
- https://github.com/ceph/ceph/pull/2576
- 09:57 AM Bug #9593 (Resolved): osdc/Objecter.cc: 1225: FAILED assert(client_lock.is_locked()) (firefly)
- ...
- 10:01 AM Revision 9593d876 (ceph): documentation: erasure-code plugin isa does not require k/m
- http://tracker.ceph.com/issues/9579 Refs: #9579
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 09:24 AM Feature #9532 (Duplicate): rados.py should export omap interface
- #6114
- 06:58 AM Revision 68862242 (ceph): mailmap: Yan Zheng affiliation
- Also adding Yan Zheng to .peoplemap to track org. change
Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gma... - 06:57 AM Revision fc1380b1 (ceph): mailmap: Thorsten Glaser affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Reviewed-by: Thorsten Glaser <tg@mirbsd.de> - 06:31 AM rgw Bug #9469: RadosGW performance degrades with high concurrency workload.
- Debugging further I was able to root cause the issue further. I enable debug logs for radosgw (20/20) , enabled acces...
- 03:31 AM CephFS Bug #9562 (Fix Under Review): Lockdep assertion in Filer purge
- https://github.com/ceph/ceph/pull/2572
- 02:58 AM Bug #9579: Default parameters are not getting initialized for EC profile using isa EC plugin
- Documentation update on k/m, good catch ! https://github.com/ceph/ceph/pull/2571
- 02:52 AM Bug #9579: Default parameters are not getting initialized for EC profile using isa EC plugin
- This is confusing and I added http://tracker.ceph.com/issues/9589 to work on improving the user experience. Thanks fo...
- 02:49 AM Bug #9579: Default parameters are not getting initialized for EC profile using isa EC plugin
- From the code, it seems default value of k & m for "isa" profile are 7 & 3 respectively.
class ErasureCodeIsaDefau... - 02:41 AM Bug #9579 (Won't Fix): Default parameters are not getting initialized for EC profile using isa EC...
- There are no defaults for k/m for the isa plugin, the parameters need to be set explicitly as documented at http://ce...
- 02:49 AM Feature #9589 (Resolved): erasure-code: query plugin for erasure-code-profile defaults
- When a parameter is missing from an erasure-code-profile (ruleset-failure-domain for instance) it falls back to the d...
- 02:14 AM Bug #8863: osd: second reservation rejection -> crash
- Hi Sage,
We are still getting this issue, even thought commit is included in our build. Any Updates? - 01:24 AM RADOS Bug #9492: Crush Mapper crashes when number of replicas is less than total number of osds to be s...
- Running in debug mode with https://github.com/ceph/ceph/pull/2568 (using the crushmap created as in the description):...
- 12:56 AM CephFS Bug #9563 (Resolved): kcephfs crash in ceph_mdsc_do_request
- 12:55 AM CephFS Bug #9564 (Resolved): kcephfs crash in _nfs4_do_open
- the bug is fixed upstream commit f39c0104 (NFS: remove BUG possibility in nfs4_open_and_get_state). I rebased the tes...
- 12:55 AM Revision 7973280a (ceph): osd: Remove unused PG functions queue_notify(), queue_info(), queue_log()
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 12:47 AM Revision 0f884fdb (ceph): For pgls OP, get/put budget on per list session basis, instead of per O...
- Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
- 12:19 AM Bug #9485: Monitor crash due to wrong crush rule set
- Thanks so much.
BTW:
I repeat this in my dev environment with 60 osds on one host. I create 6 virtual racks. (you... - 12:09 AM Bug #9485: Monitor crash due to wrong crush rule set
- Thanks for the detailed instructions. I'll try them to repeat the problem.
09/24/2014
- 11:47 PM Revision 9adf7ec0 (ceph): Merge pull request #147 from dachary/wip-ec-feature-plugin-v2-giant
- erasure-code: assert that firefly-x/stress-split cannot load lrc
- 11:30 PM Revision 67ca8441 (ceph): erasure-code: assert that firefly-x/stress-split cannot load lrc
- Because it would crash existing plugins.
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked from c... - 10:33 PM Revision 7f87cf1b (ceph): ReplicatedPG: clean out completed trimmed objects as we go
- Also, explicitely maintain a max number of concurrently trimming
objects.
Fixes: 9113
Backport: dumpling, firefly, g... - 10:27 PM Revision 7a39e7cb (ceph): Merge remote-tracking branch 'origin/giant'
- 10:23 PM rgw Bug #9588 (Rejected): Keystone s3 auth integration lacking access_key = tenant:user ability suppo...
- For instance according to http://docs.openstack.org/grizzly/openstack-object-storage/admin/content/configuring-openst...
- 08:40 PM Bug #9485: Monitor crash due to wrong crush rule set
- K=8 M=4 doesn't work.
I rebuild the cluster and do the following steps.
(delete all pools)
1. create a profile... - 04:37 AM Bug #9485: Monitor crash due to wrong crush rule set
- Could you please let me know if it always work with *K=8 M=4* ?
- 01:17 AM Bug #9485: Monitor crash due to wrong crush rule set
- I know that I need 11 and the rule provide 12 and It looks CRUSH will do thetruncate.
It doesn't seem to be an iss... - 12:42 AM Bug #9485: Monitor crash due to wrong crush rule set
- You have *K=8 M=3* which means your pool needs 11 OSDs. However the rule you defined will always provide 12 OSDs and ...
- 12:22 AM Bug #9485: Monitor crash due to wrong crush rule set
- The profile used for the ecpool is K=8 M=3.
If I set the min_size = 3, max_size = 12(as default), the monitor cras... - 12:04 AM Bug #9485: Monitor crash due to wrong crush rule set
- Could you also attach the log of monitor crash you are seeing ? Note that if you change a crush rule that is currentl...
- 08:16 PM Bug #9585: ceph assertion using rocksdb store in master branch
- It looks like that powercycle will make header's bitmap inconsistence with actual data keys.
- 11:10 AM Bug #9585 (Can't reproduce): ceph assertion using rocksdb store in master branch
- ceph version 0.85-980-gc5906ec (c5906eca2ffa837891ba7d84775ece7b91f6c5c8)
ceph assertion when rocksdb is used for ... - 07:47 PM CephFS Bug #6613: samba is crashing in teuthology
- Still happening
/a/teuthology-2014-09-22_23:14:01-samba-giant-testing-basic-multi/50607 - 07:43 PM CephFS Bug #8427: ceph-fuse: Dumpling "cache still has 0+1 items, waiting (for caps to release?)" on shu...
- /a/teuthology-2014-09-22_19:06:01-fs-dumpling-testing-basic-multi/505408
Grabbed all the logs out of /var/log/ceph... - 06:29 PM Revision 989b4065 (ceph): Merge pull request #128 from dachary/wip-ec-feature-plugin-v2
- erasure-code: CEPH_FEATURE_ERASURE_CODE_PLUGINS_V2 integration tests
- 04:21 PM Bug #6697 (Resolved): strncmp(3) must not be used on binary data
- 07:02 AM Bug #6697 (Fix Under Review): strncmp(3) must not be used on binary data
- https://github.com/ceph/ceph/pull/2567
- 06:51 AM Bug #6697: strncmp(3) must not be used on binary data
- 03:53 PM Bug #8910 (In Progress): ceph_test_objectstore: ObjectStore/StoreTest.ManyObjectTest/0 failure on...
- reopening this bug as it seems to happen in the nightlies,
log: http://qa-proxy.ceph.com/teuthology/teuthology-... - 02:51 PM devops Bug #9489 (Rejected): --zap-disk does not clear enough
- ...
- 10:22 AM devops Bug #9489 (Can't reproduce): --zap-disk does not clear enough
- 10:19 AM devops Bug #9489: --zap-disk does not clear enough
- I believe the original cause of report was likely in error unrelated to ceph-disk. Loic, you had mentioned you might ...
- 09:43 AM devops Bug #9489 (Need More Info): --zap-disk does not clear enough
- A bit more context is needed here, how/what doesn't work as expected? Is it possible to reproduce?
When zap disk d... - 02:30 PM Revision c5906eca (ceph): Merge pull request #2567 from dachary/wip-6697-strncmp-vs-memcmp
- tests: use memcmp to compare binary buffers
Reviewed-by: Sage Weil <sage@redhat.com> - 02:27 PM Fix #3180: use of strerror() for possibly-negative return values
- Yeah, I actually fixed this, and forgot the bug still existed.
- 05:51 AM Fix #3180 (Rejected): use of strerror() for possibly-negative return values
- I could not find an instance where strerror is used instead of cpp_strerror in the current master...
- 02:27 PM Feature #4611: cephtool: set-quota, no get-quota
- heh, bug 4611 duplicates bug 8523, does it? :)
- 05:18 AM Feature #4611 (Duplicate): cephtool: set-quota, no get-quota
- 02:22 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- 2014-09-22 16:00:20.680448 7fee6abcf700 0 -- 10.10.10.7:6808/25820 >> 10.10.10.16:0/1007485 pipe(0xba12a00 sd=628 :6...
- 02:21 PM rgw Bug #9587 (Resolved): ceph-radosgw sysvinit script on EL6 cannot set ulimit
- The script tries to set ulimit -n 32768 as the apache user. It errors to:
bash: line 0: ulimit: open files: cannot m... - 02:18 PM CephFS Bug #8576: teuthology: nfs tests failing on umount
- https://github.com/ceph/teuthology/pull/336
- 02:00 PM Revision 2cd9b5f9 (ceph): tests: use memcmp to compare binary buffers
- instead of strncmp because it will stop at the first \0
http://tracker.ceph.com/issues/6697 Fixes: #6697
Signed-off... - 01:58 PM Bug #9113: osd: snap trimming eats memory, linearly
- 01:57 PM Feature #9568: Add test case to test #9419 (ceph wip-9419)
- Tests case:
@0-cluster / start.yaml@... - 01:25 PM Revision bc8b67be (ceph): 0.67.11
- 12:36 PM Bug #9582: librados: segmentation fault on timeout
- Okay, looks like this is another race:
1) The message is coming in over the wire, and the Pipe grabs a preallocated ... - 07:38 AM Bug #9582 (Resolved): librados: segmentation fault on timeout
- Summary: If you configure librados with rados_osd_op_timeout, timeouts will result sometimes in a segmentation fault....
- 11:35 AM Revision 468d245a (ceph): Merge pull request #2506 from dachary/wip-9304-unintended-implicit-ruleset
- erasure-code: pool create must not always create a ruleset
Reviewed-by: João Eduardo Luís <joao@redhat.com> - 10:52 AM Bug #9584: OpTracker segfault on shutdown (firefly)
- /a/samuelj-2014-09-23_14:40:50-rados-firefly-wip-testing-old-vanilla-basic-multi/507309 (once it times out)
- 10:52 AM Bug #9584 (Can't reproduce): OpTracker segfault on shutdown (firefly)
- #0 0x00007f5ec74baf07 in _dl_map_object_deps (map=map@entry=0x7f5ec76bc4e8, preloads=preloads@entry=0x0, npreloads=n...
- 10:40 AM Revision b8e6a6b1 (ceph): Merge remote-tracking branch 'origin/giant'
- 10:37 AM Messengers Bug #1803 (New): msgr: behave better when ending TCP connections
- This has been greatly improved with the addition of our socket timeouts and things, but I don't think it's properly r...
- 03:12 AM Messengers Bug #1803 (Resolved): msgr: behave better when ending TCP connections
- Not sure at which point this problem was fixed but it is doubtful that it stayed around for the past three years unno...
- 10:21 AM Bug #9554 (Can't reproduce): "FAILED assert(0 == "hit suicide timeout")" in upgrade:firefly-firef...
- Looks like just an overloaded node.
- 10:17 AM RADOS Feature #4650: osd: separate OSD names from their IDs
- We expose OSD IDs in lots of places — like error reporting. But users can't specify those IDs (although they could on...
- 05:25 AM RADOS Feature #4650: osd: separate OSD names from their IDs
- From a system administration point of view there is no need to know about the OSD id. Naming the OSDs with human read...
- 10:05 AM RADOS Bug #8984 (Won't Fix): creating erasure-code pool when not having a root item default
- The recommended way to deal with the absence of a *default* root is to define an erasure-code-profile that "specifies...
- 10:00 AM Bug #8942: Bad JSON output in ceph osd tree
- 10:00 AM CephFS Cleanup #2378 (Resolved): "ceph -s" MDS output is confusing
- We don't print mds status if there's not an FS any more.
- 09:43 AM Revision abcbd8db (ceph): Merge branch 'master' of github.com:ceph/ceph-qa-suite
- 09:42 AM Revision 8852355c (ceph): added multi-version for giant-clients vs firefly-servers
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 09:42 AM RADOS Feature #6114: Complete python binding interfaces for librados
- It went stale as I couldn't keep up with the changes to the modules themselves as the modifications where significant...
- 08:34 AM RADOS Feature #6114: Complete python binding interfaces for librados
- What has become of https://github.com/ceph/ceph/commits/wip-5900 ? Is there a reason why it was not merged ? Or am I ...
- 09:40 AM Bug #9556 (Duplicate): Segmentation fault in upgrade:dumpling-firefly-x-giant-distro-basic-multi ...
- 09:39 AM Bug #9556: Segmentation fault in upgrade:dumpling-firefly-x-giant-distro-basic-multi run
- From Sam's advice to look for something related to "Read Timeout" and from the log, this seems to be a duplicate of #...
- 09:29 AM RADOS Feature #6421: FileStore: Op unit tests
- change the %Done to reflect the fact that there is work done already.
- 09:16 AM devops Fix #8508: packaging: deb repository key should be @redhat.com
- The deb repository key just needs to be re-created with a @redhat.com email
- 09:08 AM Bug #8323 (Duplicate): mon_osd_allow_primary_affinity Can not be Injected
- 08:36 AM Feature #5511 (Duplicate): rados.py support for object locking
- #6114
- 08:30 AM Revision 2ac486d3 (ceph): added upgrade suite for dumpling-firefly-giant
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
(cherry picked from commit c7603c0b7871477110946d2d47ea67dda1b7a... - 08:26 AM Bug #7843 (Can't reproduce): OSD fails to start
- Feel free to re-open if you have a HOWTO reproduce the issue. If you figured out what was wrong, it would be nice if ...
- 08:18 AM Revision c7603c0b (ceph): added upgrade suite for dumpling-firefly-giant
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 08:16 AM rgw Feature #7680: Use new civetweb git repo for ceph
- The repository copy is useful when fixes are needed. They can diverge from upstream while the change is proposed.
- 08:12 AM Feature #7664 (Resolved): systemd service files
- https://github.com/ceph/ceph/tree/giant/systemd
- 08:10 AM Bug #7623 (Resolved): local 'best' uninitialized in Objecter
- Fixed by 605e645026487519d4195358330832b3369b531d
- 08:05 AM Bug #6101: ceph-osd crash on corrupted store
- Bumping so it does not get to the bottom of the list for the next bug scrub.
- 08:01 AM Bug #7368: ceph osd repair * blocks after some minutes and prevent other ceph pg repair commands
- Another mention of things slowing down when repair is almost complete : http://tracker.ceph.com/issues/9566 . Not sur...
- 07:52 AM Bug #7409 (Can't reproduce): "make check" doesn't work without --with-radosgw
- ...
- 07:43 AM Bug #9362: librados, rados_read corrupts memory on timeout
- Update: The patch branch I used did not contain the complete code that has been merged to the dumpling branch. Using ...
- 07:29 AM Feature #7340 (Duplicate): rados.py does not expose object locking
- 07:18 AM Cleanup #7105 (Closed): There are three different ways to retrieve an authentication key
- It is not necessary indeed. However, now that it has been published it would be non backward compatible to remove any...
- 07:09 AM Bug #6834 (Can't reproduce): nightlies: monitor crashed in emperor
- It either showed up again and has been associated with another issue or it has been fixed.
- 06:59 AM rgw Feature #9581 (New): Ability to move objects to a second storage tier based on policy
- To be compatible with AWS S3 API like bucket lifecyle, ceph should have the ability to move the object from standard ...
- 06:48 AM Feature #6687: Ability to set up/down/in/out based on CRUSH hierarchy
- +1
- 06:47 AM Feature #3604 (Resolved): print lookup path when reporting -ENOENT to user-space
- 06:45 AM Feature #6567 (Rejected): emit warning on unknown/ invalid configuration directives
- This is unfortunately not possible as there is no central place to query to know what is a valid option and what is n...
- 06:26 AM Bug #6371 (Duplicate): rados bench segfaults when read --block-size < write --block-size
- 06:09 AM devops Bug #9506: Pass monitor SSH addresses via CLI flag
- This will be *very* tricky to do with CLI flags, so after discussing this with Kyle, it was decided that using the ce...
- 05:28 AM devops Bug #9506: Pass monitor SSH addresses via CLI flag
- The use case is ceph-deploy is being executed on a management node, homed on a management network. The monitors are m...
- 06:09 AM Feature #5521: Enhance PGLS or new op to list all namespace/objects in a pool.
- 06:08 AM Feature #5521 (Duplicate): Enhance PGLS or new op to list all namespace/objects in a pool.
- 06:02 AM Feature #9580 (Resolved): ceph-disk, ceph-osd: make journal [partition] creation conditional base...
- or example, with keyvaluestore-dev ceph-disk makes a journal parititon and general screws things up. see http://artic...
- 05:46 AM Bug #9579 (Won't Fix): Default parameters are not getting initialized for EC profile using isa EC...
When created an EC profile using erasure code plugin "isa", default values for parameters k, m and technique are ...- 05:43 AM Feature #4771 (Rejected): Snippet / included configuration
- Loic Dachary wrote:
> The ceph.conf file tends to disapear almost entirely. The mons can contain all the information... - 05:33 AM Feature #4771: Snippet / included configuration
- The ceph.conf file tends to disapear almost entirely. The mons can contain all the information and are a central poin...
- 05:36 AM Feature #4230 (Resolved): librados: node.js bindings
- https://github.com/ksperis/node-rados
- 05:30 AM Revision e1af168b (ceph): Merge pull request #142 from ceph/wip-9284
- Tests for #9284, FUSE client capability handling
Reviewed-by: Greg Farnum <greg@inktank.com> - 05:25 AM devops Bug #9510 (Closed): ceph-deploy: Move mon keyring generation 'mon create-initial'
- 05:12 AM Feature #2158 (Duplicate): cephtool: helpful error/timeout when no monitor quorum
- 04:54 AM Subtask #4306 (Resolved): make the new snap trimmer design work with split
- 04:46 AM Feature #2147 (Resolved): objclass: add CLS_ERR macro
- https://github.com/ceph/ceph/blob/giant/src/objclass/objclass.h#L31
- 03:18 AM Feature #4005: Add perftools to the kernel debian package script
- Any progress ?
- 03:17 AM Feature #1810 (Resolved): monclient: timeouts?
- Implemented by 671a76d64bc50e4f15f4c2804d99887e22dcdb69
- 03:04 AM Bug #4206 (Resolved): concurrent rados bench processes don't work well for seq reads
- Implemented by 308758b7878c48ab64caf71ff646e057c2c1c5aa
- 03:01 AM Fix #4202: osd: pg delete
- a command that deletes a designated pg ? If so it would help to have a use case.
- 02:56 AM Support #3902 (Closed): S3-tests need to cleanup after themselves
- Tests are run on short lived machines and this won't be an issue.
- 02:54 AM Feature #3855 (Resolved): Making Scrubs Nicer
- 02:52 AM Documentation #3846 (Resolved): Debian install has incorrect gitbuilder URL
- The install pages have been reworked.
- 02:49 AM Feature #3202 (Resolved): tools: coverity clean
- An on going effort by Danny Al-Gaaf
- 02:42 AM Feature #3241 (Resolved): qa: integration tests for mon, osd, and mds caps
- There now are caps tests run by teuthology : https://github.com/ceph/ceph/blob/giant/qa/workunits/mon/caps.py https:/...
- 02:29 AM Feature #3095 (Resolved): rbd tool resize improvements
- ...
- 02:27 AM Feature #3083 (Resolved): Provide separate APT repos for argonaut, bobtail, etc; stable would alw...
- Not as suggested but the stable repositories are organized in a sensible way.
- 02:23 AM Feature #2953 (Resolved): append() in librados is not exposed to python API
- Implemented by 39bf68c3ceee3f62960d0866f35835325cca5660
- 02:19 AM Bug #2848: OSDMap: pool_id is 64-bit, but pool_max is 32-bit
- "still valid":https://github.com/ceph/ceph/blob/giant/src/osd/OSDMap.h#L206
- 02:16 AM Feature #2812 (Resolved): automated CentOS testing
- RPM based operating systems are now part of the teuthology runs.
- 02:14 AM Feature #2776 (Resolved): rados tool: bulk removal of objects
- Implemented by cc8df29e19a1fc441ad903aeeb59f7d3e15a5e7c
- 02:08 AM Feature #2755 (Resolved): ceph-conftool: optionally return the default for a config option if no ...
- Marking as resolved since there now is a way to get the default value, although not as suggested....
- 02:00 AM Cleanup #2671 (Resolved): buffer.h: do efficient buffer comparisons
- Resolved by 2a46564158ebf519ae6e7ee318b97c61cf032692 with content_equals
- 01:53 AM Tasks #2529 (Resolved): debian: Merge packaging changes from Ubuntu 12.04
- There is no longer a difference.
- 01:50 AM Feature #2519 (Resolved): rados: allow setting pg_num and pgp_num when creating a pool
- Using a mon cmd to create the pool instead of the specialized function supports setting pg_num / pgp_num.
- 01:40 AM Bug #2154 (Resolved): rados: bench seq should not segfault when blocksize doesn't match write blo...
- ...
- 01:32 AM Feature #2112 (Resolved): msgr fault injection
- Starting 90f66980bfb1f2541dcb11be2c358a9832a291b1 in november 2012 a number of *OPTION(ms_inject_...* options have be...
- 01:07 AM Feature #1583 (Resolved): osd: bound pg log memory usage
- Memory consumption has improved/changed a lot since this ticket was open and I believe this issue is no longer relevant.
- 01:04 AM Feature #1619 (Resolved): libvirt: test with selinux/apparmour enabled
- I believe this has been extensively tested in the context of OpenStack
- 12:59 AM Feature #1525 (Resolved): qa: check out fio, add to ceph-qa-suite if it's good
- https://github.com/ceph/ceph-qa-suite/blob/giant/suites/tgt/basic/tasks/fio.yaml and https://github.com/ceph/ceph/blo...
- 12:52 AM Tasks #1418: set up a no-atomic-ops gitbuilder
- gitbuilders currently use *--with-libatomic-ops*
- 12:34 AM Feature #543 (Resolved): PG::search_for_missing: don't iterate over all missing
- The code base changed significantly and does not have this problem anymore.
- 12:30 AM Feature #1091 (Duplicate): librados: support pgls filter
- http://tracker.ceph.com/issues/9262
- 12:24 AM Cleanup #1042: need const iterator for bufferlist
- "still valid":https://github.com/ceph/ceph/blob/giant/src/include/buffer.h#L240
09/23/2014
- 11:58 PM Bug #9485: Monitor crash due to wrong crush rule set
- What probably happens is that you created an erasure code profile with k+m that is lower than the number of OSDs prov...
- 07:03 PM Bug #9485: Monitor crash due to wrong crush rule set
- Because the monitor crash and it can not be restarted, so currently I can not get "ceph osd dump".
I checked the i... - 08:40 AM Bug #9485: Monitor crash due to wrong crush rule set
- Could you also please add the output of *ceph osd dump* ? It looks like you have run into http://tracker.ceph.com/iss...
- 11:30 PM Revision ec4291f8 (ceph): Merge pull request #146 from dachary/wip-9298-rgw-ec-starvation-giant
- rgw: increase the number of OSDs for erasure coded pools
- 11:28 PM Revision 255b430a (ceph): osd: initialize purged_snap on backfill start; restart backfill if change
- If we backfill a PG to a new OSD, we currently neglect to initialize
purged_snaps. As a result, the first time the s... - 11:28 PM Revision c17ac03a (ceph): ReplicatedPG: don't move on to the next snap immediately
- If we have a bunch of trimmed snaps for which we have no
objects, we'll spin for a long time. Instead, requeue.
Fix... - 10:47 PM Revision e36c888c (ceph): rgw: increase the number of OSDs for erasure coded pools
- So that there never is less than 4 osd which is the minimum when trying
erasure coded workloads that only work with 4... - 09:22 PM Revision bc8fb2bc (ceph): adjusting crush tunables when running a mixed version of mons and osds....
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 08:55 PM Revision dab7a4f6 (ceph): rgw: push hash calculater deeper
- This might have been the culprit for #9307. Before we were calculating
the hash after the call to processor->handle_d... - 08:55 PM Revision 21673022 (ceph): crypto: don't hash zero sized buffer
- libnss returns an error and we assert in that case.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked f... - 08:55 PM Revision ee02cfd2 (ceph): rgw: calculate hash after writing data
- Since data is written asynchronously, we should do the hash calculation
while it's pending.
Signed-off-by: Yehuda Sa... - 08:55 PM Revision 6fee7115 (ceph): rgw: don't try to authenticate a CORS preflight request
- Fixes: #8718
Backport: firefly
CORS preflight requests don't need to be authenticated. Treat them as
coming from ano... - 08:55 PM Revision b8fa2ed6 (ceph): rgw: fix test to identify whether object has tail
- Fixes: #9226
Reported-by: Sylvain Munaut <s.munaut@whatever-company.com>
Backport: firefly
We need to identify wheth... - 08:55 PM Revision 1bdcc079 (ceph): rgw: convert header field underscores into dashes
- Fixes: 9206
Backport: firefly
Certain web servers filter out underscores in the header field name.
Convert them into... - 08:55 PM Revision a777562b (ceph): civetweb: update submodule
- Update submodule to include multiple fixes.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 08:55 PM Revision bd0a9134 (ceph): rgw: separate civetweb log from rgw log
- The civetweb log now is independent from the rgw log.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked... - 08:55 PM Revision c75a79cb (ceph): rgw: fix compilation
- RGWRadosPutObj couldn't refer to the ceph context.
Reviewed-by: Sage Weil <sage@redhat.com>
Signed-off-by: Yehuda Sa... - 08:42 PM Revision 0e90b04d (ceph): cephfs-java: build against older jni headers
- Older versions of the JNI interface expected non-const parameters
to their memory move functions. It's unpleasant, bu... - 08:38 PM Bug #9558: Both op threads and dispatcher threads get hung even for few minutes during peering stage
- More info:
When OSD daemon/host is down, some PGs becomes active+degrade, while others are still active+clean. As ... - 08:33 PM Revision b0d08aab (ceph): rgw: call throttle_data() even if renew_state() failed
- Otherwise we're going to leak the aio callback handle.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picke... - 08:33 PM Revision a953b313 (ceph): rgw: disable civetweb url decoding
- Fixes: #8621
We want to have the raw request uri, as we do the decoding ourselves.
Signed-off-by: Yehuda Sadeh <yeh... - 08:31 PM Revision ba535771 (ceph): rgw: copy_obj_data() uses atomic processor
- Fixes: #9089
copy_obj_data was not using the current object write infrastructure,
which means that the end objects w... - 08:26 PM Revision d73dbc3a (ceph): rgw: clear bufferlist if write_data() successful
- Fixes: #9201
Backport: firefly
We sometimes need to call RGWPutObjProcessor::handle_data() again,
so that we send th... - 07:55 PM Revision f833f12a (ceph): rgw: rados->set_attrs() updates bucket index
- Fixes: #5595
Backport: dumpling, firefly
We need to update the bucket index when updating object attrs, otherwise
we'... - 07:54 PM Revision 44cfd88d (ceph): rgw: PutObjMetadata, clear bufferlist before appending into it
- Fixes: #9576
Backport: firefly, dumpling
We need to completely rewrite the bufferlist, not append into it.
Signed-o... - 07:53 PM Revision 4be53d5e (ceph): PG: check full ratio again post-reservation
- Otherwise, we might queue 30 pgs for backfill at 0.80 fullness
and then never check again filling the osd after pg 11... - 06:40 PM Revision f711819d (ceph): Merge pull request #2561 from athanatos/wip-9293
- Wip 9293
Reviewed-by: Sage Weil <sage@redhat.com> - 06:16 PM CephFS Bug #9562: Lockdep assertion in Filer purge
- can we just unlock the PurgeRange/Probe locks before using the objecter?
- 06:21 AM CephFS Bug #9562 (In Progress): Lockdep assertion in Filer purge
- 06:21 AM CephFS Bug #9562: Lockdep assertion in Filer purge
So I think this bug already existed with the Probe lock, but it was triggered by the new PurgeRange lock, because t...- 05:48 PM Bug #9528: RadosModel assertion failure in firefly
- sam, please mention the parent bug.
- 01:22 PM Bug #9528 (Duplicate): RadosModel assertion failure in firefly
- 05:37 PM Revision ebb59954 (ceph): Merge pull request #2548 from dachary/wip-9547-python-rados-truncate-fi...
- python radio aio_read must not truncate on \000 (firefly)
Reviewed-by: Samuel Just <sam.just@inktank.com> - 05:31 PM Revision f4662679 (ceph): added workload tag
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 05:27 PM Revision 2abd7a3f (ceph): Merge branch 'wip-sam-testing-firefly' into firefly
- 05:07 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- fix is in wip-9487 and wip-sam-testing
- 02:28 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- I'm not 100% sure, so I'd thought I'd ask: what's the exact reason for the PG being marked incomplete here? Is it the...
- 02:05 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- 01:58 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- https://github.com/ceph/ceph/pull/2525
The num_trimmed does not seem to be reset. I think you are not trimming at... - 06:12 AM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Whoa. While a cluster with this patch applied doesn't spin like crazy in snap_trim anymore, killing an OSD seems to i...
- 05:01 PM Revision 5fab7db3 (ceph): Merge pull request #2414 from dachary/wip-cephtool-test
- tests: make qa/workunits/cephtool/test.sh more robust
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com> - 04:30 PM Revision 2c7ada49 (ceph): Merge pull request #2417 from dachary/wip-placement-group
- doc: revise placement group number guide.
Reviewed-by: John Wilkins <jowilkin@redhat.com> - 04:06 PM Bug #9113: osd: snap trimming eats memory, linearly
- It's not just dumpling, the repops set in the snap trimmer is just wonky. We need to trim a bounded set of objects, ...
- 03:59 PM Revision 34e66586 (ceph): Merge pull request #2557 from ceph/wip-mon-fix-checks
- ceph-mon: check fs stats just before preforking
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 03:47 PM Bug #9554: "FAILED assert(0 == "hit suicide timeout")" in upgrade:firefly-firefly-testing-basic-v...
The crashed osd.5 was on a node that had a load average of 38. osd.1 didn't see ping responses although it saw o...- 03:27 PM rbd Bug #8187: librbd: list_children() reports duplicates with cache pools
- Never mind, figured it out. Apparently it's not enough to set pool2 up as a tier of pool1, it also has to be an over...
- 02:51 PM rbd Bug #8187: librbd: list_children() reports duplicates with cache pools
- Josh, I'm having trouble reproducing this. Do you have a test case?
- 10:55 AM rbd Bug #8187 (In Progress): librbd: list_children() reports duplicates with cache pools
- 02:54 PM Revision 9c825ec2 (ceph): documentation: revise placement group number guide
- When a cluster has few OSDs (less than 50) propose a preselection of
values: as long as the number of placement group... - 02:50 PM Revision a5a1cce3 (ceph): included an option to ceph_objectstore_tool, whenever we have keyvalues...
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 02:31 PM CephFS Bug #9564: kcephfs crash in _nfs4_do_open
- /a/teuthology-2014-09-22_23:10:02-knfs-giant-testing-basic-multi/506055/teuthology.log
- 02:26 PM Bug #9462: msgr deadlock: osd reply vs mark_down vs fault
- Finally got through a suite run and it looks pretty good, but need to check the few failures:
http://pulpito.ceph.co... - 02:26 PM devops Bug #9268 (Resolved): Recipe errors in rgw:multifs-dumpling-testing-basic-vps
- Fixed this in ceph-qa-chef. I hought there was another issue open so in teuthology and assigned to me, this was maybe...
- 02:23 PM devops Bug #9267 (Resolved): "Gem::DependencyError" in upgrade:dumpling-dumpling-distro-basic-vps
- Problematic images now include chef.
- 02:20 PM devops Bug #9489: --zap-disk does not clear enough
- 02:15 PM devops Bug #9567: Missing packages in upgrade:dumpling-firefly-x-giant-distro-basic-vps run
- Was caused when moving to new rhel7 gitbuilder firefly was comitted but not built on the new one when the old one was...
- 02:14 PM devops Bug #9567 (Resolved): Missing packages in upgrade:dumpling-firefly-x-giant-distro-basic-vps run
- 02:09 PM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- I had some comments on that pull request.
- 02:08 PM Bug #9545: filestore stuck in journal->should_commit_now() loop on shutdown
- 02:07 PM devops Bug #9548 (Need More Info): ceph mon creation failed for centOS
- the command `mon create-initial` does not take any hosts as arguments. It doesn't take any at all.
It will look at... - 02:07 PM devops Bug #8976 (Resolved): httpd on RHEL7 (RHEL repo) incompatible with mod_fastcgi (ceph repo)
- Closing as Tamil tested and said it was good.
- 02:07 PM Bug #8629: cache_evict needs to prevent make_writeable from creating a snapdir
- 02:05 PM Bug #9285: osd: promoted object can get evicted before promotion completes
- I left a comment on a simpler approach.
- 02:00 PM Linux kernel client Bug #8568: libceph: kernel BUG at net/ceph/osd_client.c:885
- BUG_ON(!list_empty(&req->r_req_lru_item)) in __kick_osd_requests()
Can't reproduce but need to look harder into ho... - 01:39 PM Bug #9472 (Duplicate): osd crash in -upgrade:dumpling-dumpling-distro-basic-vps suite
- 01:38 PM Bug #9476 (Duplicate): "Segmentation fault (core dumped)" in upgrade:dumpling-giant-x:parallel-gi...
- 01:35 PM Bug #9570: osd crash in FileJournal::WriteFinisher::entry() aio
- what was the assert?
- 01:33 PM Bug #9501 (Rejected): Assertion in FileJournal::do_write
- 01:27 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- 01:26 PM Bug #9422 (Can't reproduce): librados: client.admin authentication error (110) Connection timed out
- 01:25 PM Bug #9274 (Can't reproduce): "AssertionError: failed to recover before timeout expired" in upgrad...
- 01:21 PM Bug #9544 (Pending Backport): osd: pg deletion vs create race leads to EEXIST on mkcoll (dumpling)
- 01:18 PM rgw Bug #8587 (Fix Under Review): rgw: subuser object not created correctly
- 01:15 PM Bug #9418: mon: drop internal-purpose messages from clients without proper caps
- 01:14 PM Bug #9546 (Rejected): LibRadosWatchNotify.WatchNotifyTest failure
- 01:13 PM Bug #9293 (Pending Backport): _collection_move_rename EEXIST
- 01:12 PM Bug #9293 (Fix Under Review): _collection_move_rename EEXIST
- 01:12 PM rgw Feature #7467 (Fix Under Review): Make radosgw work with multiple hostnames
- 09:08 AM rgw Feature #7467 (In Progress): Make radosgw work with multiple hostnames
- 01:11 PM rgw Bug #5595: object has a Content-Type, but its content_type property is not shown in Swift object ...
- Needs review, can't set status on this tracker.
- 11:21 AM rgw Bug #5595: object has a Content-Type, but its content_type property is not shown in Swift object ...
- I think this happens if the object was created before, and then its metadata was modified. It's similar to another is...
- 01:06 PM Bug #9574: Backfill: recheck full status once reservation is granted
- 12:07 PM Bug #9574 (Resolved): Backfill: recheck full status once reservation is granted
- Otherwise, we queue many backfill reservations while we are not full and then each one is granted in turn without che...
- 01:05 PM Revision 7f71c116 (ceph): ceph-mon: check fs stats just before preforking
- Otherwise statfs may fail if mkfs hasn't been run yet or if the monitor
data directory does not exist. There are che... - 01:05 PM Bug #9443 (Rejected): btrfs pwrite returns EEXIST on journal FileJournal::write_bl
- Not our bug.
- 12:57 PM rgw Bug #9575: s3tests.functional.test_s3.test_region_copy_object fails (races with radosgw-agent?)
- teuthology@teuthology:/a/teuthology-2014-09-22_23:02:01-rgw-giant-testing-basic-multi/505881
- 12:56 PM rgw Bug #9575: s3tests.functional.test_s3.test_region_copy_object fails (races with radosgw-agent?)
- teuthology@teuthology:/a/teuthology-2014-09-22_23:02:01-rgw-giant-testing-basic-multi/505875
- 12:48 PM rgw Bug #9575: s3tests.functional.test_s3.test_region_copy_object fails (races with radosgw-agent?)
- Seem to me like timing out due to slow ec backend.
- 12:34 PM rgw Bug #9575 (Duplicate): s3tests.functional.test_s3.test_region_copy_object fails (races with rados...
- ...
- 12:49 PM rgw Bug #9576 (Resolved): rgw: update object content-length doesn't work correctly
- This only applies to the swift POST object metadata api call.
- 11:51 AM devops Tasks #8366 (Fix Under Review): Update ceph.com/docs to default to the latest major release (0.80)
- We need to review this a bit further. Pointing to the latest major release is fine, but we need to have a way to cher...
- 11:43 AM Bug #8885 (Resolved): SIGABRT in TrackedOp::dump() via dump_ops_in_flight()
- 11:37 AM Revision 9d3fbe92 (ceph): Merge pull request #2551 from dachary/wip-9343-erasure-code-feature
- erasure code feature
Reviewed-by: João Eduardo Luís <joao@redhat.com> - 11:34 AM Revision 9687150c (ceph): erasure-code: isa/lrc plugin feature
- There are two new plugins (isa and lrc). When upgrading a cluster, there
must be a protection against the following s... - 11:08 AM Bug #9547 (Resolved): python rados aio_read truncates returned buffer on \000
- 10:33 AM Bug #9482 (Resolved): osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log....
- 10:33 AM Bug #9339 (Resolved): ReplicatedPG crash in hitset_create
- 10:32 AM Bug #8777 (Resolved): osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == log.rbegin())
- 10:32 AM Bug #9054 (Resolved): ceph_test_rados: FAILED assert(!old_value.deleted())
- 10:32 AM Bug #9326 (Resolved): osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- Does not need to be backported!
- 10:30 AM Bug #9240 (Resolved): osd_max_backfills = 1 can cause reserver deadlock for EC
- 10:30 AM Bug #9179 (Resolved): unfound objects, recovery timeout
- 10:30 AM Bug #9481 (Resolved): osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- 10:30 AM Bug #9497 (Resolved): choose_acting has to let the pg be down any time acting < min_size even if ...
- 09:42 AM Revision f51d21b5 (ceph): erasure-code: restore jerasure BlaumRoth default w
- Changing from W=7 to W=6 by default for the BlaumRoth technique is
correct but introduces a regression. The content t... - 09:39 AM Linux kernel client Bug #9573 (New): krbd: investigate a dd-in-a-loop slowdown
- Reported at the bottom of #8818.
- 09:38 AM rbd Bug #5768 (Fix Under Review): rbd-fuse: leak in enumerate_images()
- https://github.com/ceph/ceph/pull/2524
- 09:34 AM rbd Bug #6926 (Fix Under Review): rbd: diff output includes previously non-existent objects as zeroed...
- https://github.com/ceph/ceph/pull/2523
- 09:27 AM Feature #8188: librados: interface to inspect pool properties
- https://github.com/ceph/ceph/pull/2552
- 09:27 AM Feature #8188 (Fix Under Review): librados: interface to inspect pool properties
- 09:26 AM rgw Bug #7796 (Won't Fix): RGW Keystone token auth fails with '411 Length Required' when Keystone usi...
- The recommendation is to work around the issue using the afformentioned apache configuration.
- 09:14 AM rgw Bug #8676: md5sum check failed during readwrite.py
- This might have been fixed, downgrading it for now until it's dis/proved.
- 08:59 AM rgw Bug #8676: md5sum check failed during readwrite.py
- There's a chance this one is the same as #9307
- 09:07 AM rgw Bug #6611 (Won't Fix): RGW: Using underscores when setting headers returns 403
- The cgi interface prevents us from doing anything about it. With civetweb it'd be different, but at this point there'...
- 09:02 AM devops Bug #6592 (Can't reproduce): 3.8 kernel + /dev/cciss/c0d1 + precise : fail to show in /dev/disk/b...
- I lost access to the hardware before being able to properly reproduce / diagnose this border case.
- 08:59 AM rgw Bug #9307 (Pending Backport): "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dump...
- Should have been fixed by commit:d41c3e858c6f215792c67b8c2a42312cae07ece9
Note that when backporting also need to ... - 08:57 AM Bug #9408: erasure-code: misalignment
- gitbuilder is all green
- 08:56 AM Bug #9408 (Fix Under Review): erasure-code: misalignment
- Corresponding pull request https://github.com/ceph/ceph/pull/2558
- 08:52 AM rgw Bug #9529 (Resolved): ./common/ceph_crypto.h: 83: FAILED assert(s == SECSuccess)
- 08:52 AM rgw Bug #9529: ./common/ceph_crypto.h: 83: FAILED assert(s == SECSuccess)
- Fixed by commit:7b137246b49a9f0b4d8b8d5cebfa78cc1ebd14e7
- 08:45 AM Bug #9381 (Resolved): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in ...
- All rpm packages were eventually updated.
- 08:42 AM Bug #9224 (Can't reproduce): osd: segv in dlopen
- 08:29 AM Bug #9470 (Resolved): daemon pid file is not being created when running service ceph
- 08:29 AM Bug #9509 (Resolved): init script cannot stop OSDs
- 08:15 AM Bug #9572 (Fix Under Review): erasure-code: BlaumRoth default encoding regression
- 02:45 AM Bug #9572: erasure-code: BlaumRoth default encoding regression
- https://github.com/ceph/ceph/pull/2556
- 02:35 AM Bug #9572 (In Progress): erasure-code: BlaumRoth default encoding regression
- 02:10 AM Bug #9572 (Resolved): erasure-code: BlaumRoth default encoding regression
- Fixing the "bug on BlaumRoth w constraint":https://github.com/ceph/ceph/commit/9e2d04f7631cc7cd8444e7329890c2429a2d94...
- 06:31 AM Feature #9420: erasure-code: tools and archive to check for non regression of encoding
- 05:54 AM Revision e270216f (ceph): mailmap: William A. Kennington III affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 05:51 AM Revision 9f0202b0 (ceph): mailmap: Jason Dillaman affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 04:37 AM Feature #9343 (Resolved): erasure-code: allow upgrades for lrc and isa plugins
- 02:16 AM Revision 7354165c (ceph): Merge pull request #2538 from ceph/wip-mon-data-space-die
- mon: die if 'mon data' fs has critically low available disk space & fix logging issues
Reviewed-by: Sage Weil <sage@... - 12:13 AM devops Bug #9506 (Rejected): Pass monitor SSH addresses via CLI flag
- There probably is something to be done to clarify the confusion between mon id and hostnames but it is another topic ;-)
09/22/2014
- 10:43 PM CephFS Bug #9563: kcephfs crash in ceph_mdsc_do_request
- the bug came from "ceph: use pagelist to present MDS request data". I force updated the testing branch, please test it.
- 05:04 AM CephFS Bug #9563 (Resolved): kcephfs crash in ceph_mdsc_do_request
From serial console:...- 07:50 PM Bug #9571 (Resolved): rocksdb testing with powercycling fails on trusty
- This is when osd_objectstore is using rocksdb,...
- 07:19 PM Bug #9503 (Fix Under Review): Dumpling: removing many snapshots in a short time makes OSDs go ber...
- 07:32 AM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- OK, that seems to have done it. After installing the updated autobuild with Dan's patch and keeping the snap_trim lim...
- 07:18 PM Bug #9502 (Pending Backport): mon: does not verify disk is not full on startup
- 07:16 PM Bug #9455 (Resolved): mon: audit log read events should be debug level
- 04:36 PM Revision 9996d446 (ceph): mon: DataHealthService: use get_fs_stats() instead
- and relieve the DataStats struct from clutter by using
ceph_data_stats_t instead of multiple fields.
Signed-off-by: ... - 04:36 PM Revision 2da1a291 (ceph): ceph_mon: check available storage space for mon data dir on start
- error out if available storage space is below 'mon data avail crit'
Fixes: #9502
Signed-off-by: Joao Eduardo Luis <... - 04:36 PM Revision 6c378aeb (ceph): common: LogEntry: if channel is missing, default to "cluster"
- Keeps backward compatibility when there are entities that do not know
what a channel is. This way we ensure that tho... - 04:36 PM Revision 3760bc1b (ceph): mon: LogMonitor: appropriately expand channel meta variables
- We must only expand the log file's channel meta variables upon requiring
a channel's log file. As we may have a 'def... - 04:36 PM Revision 9686044a (ceph): mon: LogMonitor: add debug message upon logging to a channel's file
- Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
- 04:36 PM Revision 282bac79 (ceph): mon: LogMonitor: adjust debug messages output levels
- Reduce the noise.
Signed-off-by: Joao Eduardo Luis <joao@redhat.com> - 04:36 PM Revision f1b814e5 (ceph): mon: Monitor: log RO admin socket commands on 'debug' level
- Reduces the noise caused by read-only operations via the admin socket.
RW commands are still logged at 'info' level.
... - 04:36 PM Revision bb558620 (ceph): mon: Monitor.h: add 'requires_perm()' function to MonCommand struct
- Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
- 04:36 PM Revision 2c5b12d9 (ceph): mon: Monitor: use MonCommand::requires_perm() when checking perms
- Signed-off-by: Joao Eduardo Luis <joao@redhat.com>
- 04:36 PM Revision 89fceb3c (ceph): mon: Monitor: log RO commands on 'debug' level, RWX on 'info'
- Fixes: #9455
Signed-off-by: Joao Eduardo Luis <joao@redhat.com> - 04:36 PM Revision 3d74230d (ceph): common: util: add get_fs_stats() function
- simplifies the task of obtaining available/used disk space, as well as
used available percentage.
Signed-off-by: Joa... - 03:57 PM devops Feature #9050: Calamari builds for ceph.com
- Yes, we need a ceph.com/<something>/calamari repo which contains the various packages.
What needs some discussion... - 03:55 PM Bug #9570 (Rejected): osd crash in FileJournal::WriteFinisher::entry() aio
- h3. Workaround
Try with a kernel newer than 3.13 - as new as the environment allows.
h3. Collect more informati... - 02:31 PM Revision f421d5cc (ceph): documentation: comment the CompatSet data members
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 02:18 PM Feature #9420: erasure-code: tools and archive to check for non regression of encoding
- * Created the repository https://github.com/ceph/ceph-erasure-code-corpus
* Asked Sandon if having such a reposito... - 01:37 PM Feature #9568 (Resolved): Add test case to test #9419 (ceph wip-9419)
- 12:03 PM Bug #9538 (Resolved): mon crashes on some --format=plain commands
- 10:32 AM devops Bug #9567: Missing packages in upgrade:dumpling-firefly-x-giant-distro-basic-vps run
- and http://pulpito.front.sepia.ceph.com/teuthology-2014-09-21_19:25:01-upgrade:dumpling-firefly-x-giant-distro-basic-...
- 09:10 AM devops Bug #9567 (Rejected): Missing packages in upgrade:dumpling-firefly-x-giant-distro-basic-vps run
- In run http://pulpito.front.sepia.ceph.com/teuthology-2014-09-21_19:25:01-upgrade:dumpling-firefly-x-giant-distro-bas...
- 10:13 AM Feature #9343 (Fix Under Review): erasure-code: allow upgrades for lrc and isa plugins
- Rebased the pull request against giant https://github.com/ceph/ceph/pull/2551
- 07:45 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- The "logs of the failed test":http://qa-proxy.ceph.com/teuthology/ubuntu-2014-09-19_04:50:17-rados:monthrash-wip-9343...
- 07:33 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- The "monthrash against giant":http://pulpito.ceph.com/ubuntu-2014-09-20_00:35:01-rados:monthrash-giant-testing-basic-...
- 10:08 AM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- Samuel Just wrote:
> I'd need the corresponding logs from osd.5 to be sure, but I believe the problem is that osd.5,... - 10:05 AM Bug #9515 (Duplicate): "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:...
- 10:04 AM Bug #9515: "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:parallel-gia...
- Seems to be related to http://tracker.ceph.com/issues/9508 and recently resolved
- 10:01 AM Bug #9515: "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:parallel-gia...
- The stack trace is:...
- 09:17 AM Bug #9515: "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:parallel-gia...
- Also seeing in suite:upgrade:dumpling-firefly-x
http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-21_19:25:01... - 07:53 AM Bug #9515: "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:parallel-gia...
- Also shows in http://tracker.ceph.com/issues/9343#note-9
- 08:52 AM Feature #9161: Cache warmup and ejection
- I started to work on this.
Is there a chance it could go into Hammer release? - 08:51 AM Feature #9161: Cache warmup and ejection
- I started to work on this.
Is there a change it could go into Hammer release? - 08:43 AM Fix #9566 (Need More Info): osd: prioritize recovery of OSDs with most work to do
Assume 72 hours for host replacement/reprovisioning SLA. When host goes down (hardware failure), we expect complete...- 07:36 AM devops Bug #9510: ceph-deploy: Move mon keyring generation 'mon create-initial'
- Would adding a separate command for keyring creation be better?
Would moving it to `create-initial` mean that it ... - 05:12 AM devops Bug #9506 (Need More Info): Pass monitor SSH addresses via CLI flag
- Could you give me a use case? In what context something like this would happen, and at what point in the deployment p...
- 05:08 AM CephFS Bug #9564: kcephfs crash in _nfs4_do_open
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-19_23:10:01-knfs-giant-testing-basic-multi/500158/...
- 05:07 AM CephFS Bug #9564 (Resolved): kcephfs crash in _nfs4_do_open
- 04:47 AM CephFS Bug #9562: Lockdep assertion in Filer purge
- ...
- 04:46 AM CephFS Bug #9562 (Resolved): Lockdep assertion in Filer purge
- 04:08 AM Linux kernel client Bug #8979 (Resolved): GPF kernel panics - auth?
- Landed in 3.17-rc5. Opened #9560 and #9561 for the issues mentioned above.
- 04:04 AM Linux kernel client Bug #9561 (Rejected): libceph: do not crash if auth reply is not understood
- 04:02 AM Linux kernel client Bug #9560 (Rejected): libceph: msg kmalloc failure handling on the reply path
- 02:55 AM Bug #9077: Cluster is up in MON node even if Ceph is uninstalled in OSD node
- Issue reproduced, find the following info
Attaching mon and dmesg log of monitor node
Executed following comman... - 12:51 AM rbd Bug #8000: SLAB: Unable to allocate memory on node 0
- RAM frequency, interesting. Something to keep in mind..
09/21/2014
- 11:56 PM Bug #9559: ?off-by-one vulnerability?ceph-0.80.5/src/common/fd.cc dump_open_fds() function
- ceph-0.80.5/src/common/fd.cc dump_open_fds() function allows attackers to cause buffer overflow via vectors related t...
- 11:47 PM Bug #9559 (Resolved): ?off-by-one vulnerability?ceph-0.80.5/src/common/fd.cc dump_open_fds() func...
- ceph-0.80.5/src/common/fd.cc dump_open_fds() function allows attackers to cause buffer overflow via vectors related...
- 11:28 PM Bug #7995: osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
- Please ignore the previous update. Here is the correct one:
While some osds where in nearfull situation, shutdown ... - 10:56 PM Revision ce8eefca (ceph): osd/ReplicatedPG: do not clone or preserve snapdir on cache_evict
- If we cache_evict a head in a cache pool, we need to prevent
make_writeable() from cloning the head and finish_ctx() ... - 10:54 PM Revision 398c74ea (ceph): ceph_test_rados_api_tier: add EvictSnap2 test case
- Verify an evict doesn't create a snapdir object. Reproduces #8629
Signed-off-by: Sage Weil <sage@redhat.com> - 10:27 PM Bug #8863: osd: second reservation rejection -> crash
- Yes Sage it is included.
commit 2b13de16c522754e30a0a55fb9d072082dac455e
Author: Sage Weil <sage@redhat.com>
Dat... - 10:24 PM Bug #9558 (Can't reproduce): Both op threads and dispatcher threads get hung even for few minutes...
- During peering stage, op threads will handle peering event and check the missing objects in this function: bool PG::M...
- 09:56 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Thanks a lot! I'll report back once there is an update to share.
- 07:26 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- That log shows me PGs with huge snap_trimq, which is very unfriendy to the snap trimmer. I've added Dan's patch on t...
- 12:51 PM Bug #9503 (Need More Info): Dumpling: removing many snapshots in a short time makes OSDs go berserk
- 09:23 PM Revision e2791ad5 (ceph): Merge pull request #145 from dachary/wip-erasure-code-workloads
- erasure-code: fix bugous ec-rados-default.yaml in some suites
- 08:41 PM Revision 682b9daa (ceph): os/FileJournal: do not request sync while shutting down
- FileStore calls should_commit_now() to determine whether it should
loop and do a second sync (among other things). D... - 07:16 PM Bug #8752: firefly: scrub/repair stat mismatch
- Sage Weil wrote:
> Is it possible the inconsistencies are correlated with the kernel (vs userspace) client? That wo... - 07:03 PM Bug #8752: firefly: scrub/repair stat mismatch
- Dmitry Smirnov wrote:
> On 0.80.5 inconsistencies disappear from pool 20 (CephFS caching pool) although I also stopp... - 06:12 PM Bug #8752: firefly: scrub/repair stat mismatch
- On 0.80.5 inconsistencies disappear from pool 20 (CephFS caching pool) although I also stopped using kernel FS client...
- 07:11 PM rbd Bug #8000 (Closed): SLAB: Unable to allocate memory on node 0
- No particular access pattern seems to provoke this issue and frankly I have no clue what's causing it apart from "dee...
- 06:26 PM Revision f5256f14 (ceph): ceph_manager: reset timeout if we are making progress in wait_for_recovery
- Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 8d9d724b7f2f6c2984b0ee58a5a91f79820719f2) - 05:19 PM Revision 544b8c7f (ceph): ReplicatedPG::on_removal: clear rollback info
- Fixes: #9293
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com> - 05:10 PM Revision 781f05c9 (ceph): Revert "PG::init: clear rollback info for backfill as well"
- Log here is already empty, need to rollback the log when we start
removal in on_removal().
This reverts commit 1e69f... - 05:04 PM Revision 80f2f9cf (ceph): Merge remote-tracking branch 'origin/wip-9240' into wip-sam-testing-fir...
- 05:04 PM Revision bb9e65bd (ceph): PG: wait until we've build the missing set to discover_all_missing
- Fixes: #9179
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 970d9830a... - 05:04 PM Revision fa645b21 (ceph): PG: mark_log_for_rewrite on resurrection
- Fixes: #8777
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 8346e1075... - 05:04 PM Revision fd96eb62 (ceph): ReplicatedPG:start_flush send a second delete
- Suppose we start with the following in the cache pool:
30:[29,21,20,15,10,4]:[22(21), 15(15,10), 4(4)]+head
The obj... - 05:04 PM Revision 5aa35ac2 (ceph): ReplicatedPG::start_flush: remove superfluous loop
- Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 66c7439ea0888777b5cfc08bcb0fbd7bfd8653c3) - 05:03 PM Revision 14fb643b (ceph): Merge remote-tracking branch 'origin/wip-9339' into wip-sam-testing-fir...
- 04:58 PM Revision 55e5160d (ceph): erasure-code: fix bugous ec-rados-default.yaml in some suites
- The ec-rados-default.yaml was linked to suites as if it was a task
although it is a workload intended for parallel up... - 04:26 PM CephFS Feature #9557 (Resolved): mds: verify backtrace on fetch_dir
- Verify that the backtrace is valid when we finish fetch_dir. That is, that we would have been able to locate the dir...
- 04:13 PM Bug #9285 (Fix Under Review): osd: promoted object can get evicted before promotion completes
- 04:03 PM Bug #8629 (Fix Under Review): cache_evict needs to prevent make_writeable from creating a snapdir
- https://github.com/ceph/ceph/pull/2550
- 02:25 PM Bug #9556 (Duplicate): Segmentation fault in upgrade:dumpling-firefly-x-giant-distro-basic-multi ...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-21_10:14:47-upgrade:dumpling-firefly-x-giant-distr...
- 01:42 PM Bug #9545 (Fix Under Review): filestore stuck in journal->should_commit_now() loop on shutdown
- https://github.com/ceph/ceph/pull/2549
- 12:50 PM Bug #9389: ec pg stuck peering, did not send query for one shard
- At least on that one, looks like do_queries doesn't send the query. That can happen if the osd is down as of the osd...
- 12:41 PM Bug #9389: ec pg stuck peering, did not send query for one shard
- /a/samuelj-2014-09-20_19:00:23-rados-wip-sam-testing-firefly2-wip-testing-old-vanilla-basic-multi/501557
probably ... - 12:33 PM Revision 2e60c2f1 (ceph): test: check python rados aio_read with buffers containing null
- http://tracker.ceph.com/issues/9547 Refs: #9547
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
(cherry picked... - 12:33 PM Revision 72a46dec (ceph): pybind: Fix aio_read handling of string buffer
- Read data may contain \0, and buf.value interprerts them as string terminator.
Signed-off-by: Mohammad Salehe <saleh... - 11:22 AM Revision 3cf38bbb (ceph): Merge pull request #2547 from wkennington/master
- Cleanup boost optionals for boost 1.56
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 11:06 AM Bug #9555 (Resolved): msg/Pipe.cc: 1513: FAILED assert(0 == "old msgs despite reconnect_seq featu...
- firefly
/a/samuelj-2014-09-20_19:00:23-rados-wip-sam-testing-firefly2-wip-testing-old-vanilla-basic-multi/501749/r... - 10:08 AM Bug #9293: _collection_move_rename EEXIST
- 08:20 AM Revision a53ead14 (ceph): osd: Cleanup boost optionals
- Signed-off-by: William A. Kennington III <william@wkennington.com>
- 07:38 AM Revision 86bdd044 (ceph): Merge pull request #2545 from dachary/wip-9547-python-rados-truncate
- python radio aio_read must not truncate on \000
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 05:44 AM Bug #9547: python rados aio_read truncates returned buffer on \000
- firefly backport https://github.com/ceph/ceph/pull/2548
- 03:18 AM Bug #9547: python rados aio_read truncates returned buffer on \000
- The example from the description was not right but fixing it to have the expected length does not change the result o...
- 03:16 AM Bug #9547 (Pending Backport): python rados aio_read truncates returned buffer on \000
- 01:26 AM Revision e8ac9d86 (ceph): Merge pull request #2546 from ceph/wip-log-msgs
- mds: remove spurious logging
09/20/2014
- 11:55 PM Revision 28b7b933 (ceph): mds: remove spurious logging
- This is from 4f3b8032d, it should not have been
included. Harmless but noisy.
Signed-off-by: John Spray <john.spray... - 06:46 PM Bug #9554 (Can't reproduce): "FAILED assert(0 == "hit suicide timeout")" in upgrade:firefly-firef...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-20_15:08:15-upgrade:firefly-firefly-testing-basic-...
- 05:53 PM devops Bug #9460: mira004, mira036. mira017 unresponsive
- mira004 is bad again - 2014-09-20T17:31:32.251 INFO:teuthology.provision:Downburst completed on ubuntu@vpm024.front.s...
- 03:29 PM Linux kernel client Bug #9432: kcephfs: null pointer deref in posix_acl_create
- 03:04 PM Bug #9551 (Duplicate): "Segmentation fault" in upgrade:firefly-firefly-testing-basic-vps run
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-20_13:44:11-upgrade:firefly-firefly-testing-basic-...
- 11:08 AM Revision 226c0c7a (ceph): test: check python rados aio_read with buffers containing null
- http://tracker.ceph.com/issues/9547 Refs: #9547
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 10:43 AM Revision 8bda44ff (ceph): pybind: Fix aio_read handling of string buffer
- Read data may contain \0, and buf.value interprerts them as string terminator.
Signed-off-by: Mohammad Salehe <saleh... - 06:32 AM devops Bug #9548 (Rejected): ceph mon creation failed for centOS
- Trying to deploy ceph in centOS. But every time execute the below command I'm getting failed response.
[ceph@ceph-... - 04:15 AM Bug #9547 (Fix Under Review): python rados aio_read truncates returned buffer on \000
- 04:15 AM Bug #9547: python rados aio_read truncates returned buffer on \000
- running wip-9547-python-rados-truncate from https://github.com/ceph/ceph/pull/2545 on http://ceph.com/gitbuilder.cgi
- 03:44 AM Bug #9547: python rados aio_read truncates returned buffer on \000
- "need firefly backport":https://github.com/ceph/ceph/blob/firefly/src/pybind/rados.py#L1093
- 03:40 AM Bug #9547: python rados aio_read truncates returned buffer on \000
- Proposed fix https://github.com/ceph/ceph/pull/2544
- 03:36 AM Bug #9547 (Resolved): python rados aio_read truncates returned buffer on \000
- ...
- 03:51 AM Revision 7fe6e105 (ceph): Merge pull request #2431 from ceph/wip-6613
- Wip 9178
- 02:16 AM Bug #9535 (Duplicate): monitor crashed after restarting
- 02:14 AM Bug #9455 (Fix Under Review): mon: audit log read events should be debug level
- https://github.com/ceph/ceph/pull/2538
- 02:14 AM Bug #9502 (Fix Under Review): mon: does not verify disk is not full on startup
- https://github.com/ceph/ceph/pull/2538
- 12:37 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- What "was supposed to be the baseline":http://pulpito.ceph.com/sage-2014-09-18_17:42:51-rados:monthrash-wip-9301-dist...
09/19/2014
- 10:58 PM Revision bddeb954 (ceph): sharedptr_registry.hpp: removed ptrs need to not blast contents
- See the included unit test update. Consider:
1) x = lookup_or_create(1, 1)
2) remove(1)
3) y = lookup_or_create(1, 2... - 10:58 PM Revision 49e8e466 (ceph): SharedPtrRegistry: get_next must not delete while holding the lock
- bool get_next(const K &key, pair<K, VPtr> *next)
may indirectly delete the object pointed by next->second when
doing... - 10:58 PM Revision 478ec880 (ceph): sharedptr_registry: add a variant of get_next() and the empty() method
- The SharedPtrRegistry::get_next() method with a value of type VPtr
instead of V is added because it is sometime more ... - 10:58 PM Revision e0f4162e (ceph): replace in_method_t with a counter
- A single counter ( waiting ) accurately reflects the number of
waiters, regardless of the method waiting. It is enoug... - 10:57 PM Revision d236a93b (ceph): unit tests for sharedptr_registry
- Covers 100% of the LOC and all the expected behavior, including thread
safety.
The sharedptr_registry is made friend... - 10:25 PM Revision 2a2711da (ceph): Merge pull request #2542 from ceph/wip-9539
- Filer: add lock to protect struct PurgeRange
- 10:09 PM Revision 9d9c8c73 (ceph): Filer: add lock to protect strcut PurgeRange
- Fix: #9539
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 09:40 PM Revision 77b9f4a5 (ceph): Merge pull request #2529 from dachary/wip-doc-troubleshooting
- documentation: fix bugous osd stop/start example
- 09:40 PM Revision 927f14b9 (ceph): Merge pull request #2532 from dachary/wip-doc-tiering
- documentation: tiering typo
- 09:40 PM Revision 859bf308 (ceph): Merge pull request #2534 from dachary/wip-9537-erasure-code-assert
- erasure-code: fix assert overflow
Reviewed-by: Sage Weil <sage@redhat.com> - 09:27 PM Revision bdb1f0f0 (ceph): Merge remote-tracking branch 'gh/wip-9301'
- Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
- 09:16 PM Revision cdcb315c (ceph): fixed indentation in the yaml
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 09:02 PM CephFS Bug #9178 (Resolved): samba: ENOTEMPTY on "rm -rf"
- 08:58 PM Bug #9546 (Rejected): LibRadosWatchNotify.WatchNotifyTest failure
- ...
- 08:53 PM Revision b47fdd40 (ceph): rbd: Use a rolling average to compute RBD write throughput
- Replace the cumulative average with a rolling average
to better expose variations within IOS/sec and bytes/sec.
Fixe... - 08:42 PM Revision 437db44b (ceph): Merge remote-tracking branch 'wonzhq/req-pending' into giant
- 08:34 PM Revision 184773d6 (ceph): Merge remote-tracking branch 'somnathr/wip-optracker-optimization' into...
- 06:35 PM Revision f3acae41 (ceph): rgw_main.cc: add missing virtual destructor for RGWRequest
- CID 1160858 (#1 of 1): Non-virtual destructor (VIRTUAL_DTOR)
nonvirtual_dtor: Class RGWLoadGenRequest has a destruct... - 06:35 PM Revision eeb74a1c (ceph): os/GenericObjectMap.cc: pass big parameter by reference
- CID 1188142 (#1 of 1): Big parameter passed by value (PASS_BY_VALUE)
pass_by_value: Passing parameter header of type... - 06:35 PM Revision 1f134bb0 (ceph): mds/Beacon.*: fix UNINIT_CTOR cases
- CID 1238905 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member want_state is... - 06:34 PM Revision f2a7d62c (ceph): ErasureCodeLrc.h: fix UNINIT_CTOR
- Fix Coverity issue, preinit with 0:
uninit_member: Non-static class member chunk_count is not initialized
in this c... - 06:34 PM Revision 1a748222 (ceph): ErasureCodeLrc.h: fix data_chunk_count UNINIT_CTOR
- CID 1238903 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member data_chunk_co... - 06:34 PM Revision 0e15df43 (ceph): LogEntry.h: init LogEntry::seq with 0
- CID 1238904 (#1 of 1): Uninitialized scalar variable (UNINIT)
uninit_use_in_call: Using uninitialized value e.seq wh... - 06:34 PM Revision e34b8ad5 (ceph): Merge pull request #2537 from ceph/wip-coverity-20140919
- Various fixes for Coverity issues
Reviewed-by: Sage Weil <sage@redhat.com> - 06:30 PM Revision 2675b0d1 (ceph): Merge pull request #2535 from dachary/wip-9470-pidfile-firefly
- daemons: write pid file even when told not to daemonize (firefly)
Reviewed-by: Sage Weil <sage@redhat.com> - 06:13 PM Revision 7f1d9e8a (ceph): Merge pull request #2539 from ceph/wip-9529
- Wip 9529
Reviewed-by: Sage Weil <sage@redhat.com> - 05:48 PM Revision 5bb94ede (ceph): rgw: calculate hash after writing data
- Since data is written asynchronously, we should do the hash calculation
while it's pending.
Signed-off-by: Yehuda Sa... - 05:48 PM Revision 7b137246 (ceph): crypto: don't hash zero sized buffer
- libnss returns an error and we assert in that case.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 05:33 PM Revision 76eff950 (ceph): include/util.h: prevent multiple inclusion of header
- Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
- 05:32 PM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- 05:31 PM Revision 705abd9f (ceph): Merge pull request #2517 from athanatos/wip-9326
- DBObjectMap: restructure map_header_in_use locking
Reviewed-by: Sage Weil <sage@redhat.com> - 05:18 PM Bug #9545: filestore stuck in journal->should_commit_now() loop on shutdown
- ...
- 05:18 PM Bug #9545: filestore stuck in journal->should_commit_now() loop on shutdown
- sync_entry is looping on the same seq while the main thread waits for umount. journal should_commit_now() is stuck r...
- 05:18 PM Bug #9545 (Resolved): filestore stuck in journal->should_commit_now() loop on shutdown
- 04:47 PM Revision 305e4c49 (ceph): Merge pull request #2526 from ceph/wip-6228
- common: Add cctid meta variable
Backport: giant, firefly
Reviewed-by: Sage Weil <sage@redhat.com> - 04:45 PM Bug #9390: EEXIST on split due to import/export
- Not precisely sure how to approach this. We can make the OSD robust to this situation or we can adjust the test to a...
- 04:44 PM Bug #9390: EEXIST on split due to import/export
- Tricky. I think that we saw the following sequence:
stop osd N
export pg X at epoch e
split pg X at epoch e+3
... - 04:43 PM Bug #8011 (Can't reproduce): osd/ReplicatedPG.cc: 5244: FAILED assert(soid < scrubber.start || so...
- Pinged Dmitry to see if he is sitll seeing this or has a log
- 04:35 PM Bug #9384 (Resolved): OSD is crashing while io is running and querying withadmin socket
- 04:32 PM Revision b82ceda7 (ceph): rgw_main.cc: add missing virtual destructor for RGWRequest
- CID 1160858 (#1 of 1): Non-virtual destructor (VIRTUAL_DTOR)
nonvirtual_dtor: Class RGWLoadGenRequest has a destruct... - 04:32 PM Revision 13b8c92a (ceph): os/GenericObjectMap.cc: pass big parameter by reference
- CID 1188142 (#1 of 1): Big parameter passed by value (PASS_BY_VALUE)
pass_by_value: Passing parameter header of type... - 04:32 PM Revision 45e0b33d (ceph): common/buffer.cc: catch exception by reference
- CID 1135926 (#1 of 1): Big parameter passed by value (PASS_BY_VALUE)
pass_by_value: Catching an exception object of s... - 04:32 PM Revision 7e829509 (ceph): test/kv_store_bench.cc: fix coverity REVERSE_INULL
- CID 1021214 (#1 of 1): Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking cb_args sugges... - 04:31 PM Revision 48389d07 (ceph): test/libradosstriper/striping.cc: remove dead initialization
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 04:31 PM Revision ff614832 (ceph): mds/Beacon.*: fix UNINIT_CTOR cases
- CID 1238905 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member want_state is... - 04:27 PM Revision ea02dc37 (ceph): ErasureCodeLrc.h: fix UNINIT_CTOR
- Fix Coverity issue, preinit with 0:
uninit_member: Non-static class member chunk_count is not initialized
in this c... - 04:27 PM Revision 2bb35bba (ceph): ErasureCodeLrc.h: fix data_chunk_count UNINIT_CTOR
- CID 1238903 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member data_chunk_co... - 04:27 PM Revision 2c2b413e (ceph): LogEntry.h: init LogEntry::seq with 0
- CID 1238904 (#1 of 1): Uninitialized scalar variable (UNINIT)
uninit_use_in_call: Using uninitialized value e.seq wh... - 04:26 PM Revision af039ce8 (ceph): test/librados/aio.cc: close resource leak
- Release my_completion4.
CID 1219581 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable my_completion... - 04:13 PM Bug #9502: mon: does not verify disk is not full on startup
- 04:03 PM Bug #9544: osd: pg deletion vs create race leads to EEXIST on mkcoll (dumpling)
- wip-sharedptr-registry-backport
- 03:39 PM Bug #9544 (Resolved): osd: pg deletion vs create race leads to EEXIST on mkcoll (dumpling)
- ...
- 03:59 PM Revision 6cbf95b6 (ceph): Merge pull request #2533 from dachary/wip-9538-format-plain
- mon: osd find / metadata --format plain fallback
Reviewed-by: John Spray <john.spray@redhat.com> - 03:42 PM Bug #7120 (Duplicate): osd: EEXIST on mkcoll on dumpling
- 03:34 PM Bug #7120: osd: EEXIST on mkcoll on dumpling
- /a/sage-2014-09-18_22:33:58-rados-dumpling-distro-basic-multi/496304/remote
- 03:34 PM CephFS Bug #9539 (Resolved): struct PurgeRange in Filer.cc needs lock to protect
- 06:32 AM CephFS Bug #9539 (Resolved): struct PurgeRange in Filer.cc needs lock to protect
- send two requests to delete 1000026dfe3.00000067, but no request to 1000026dfe3.00000068...
- 03:30 PM Revision 5cb994d7 (ceph): Merge pull request #2536 from dachary/wip-doc-reweight
- documentation: explain ceph osd reweight vs crush weight
Reviewed-by: Laurent Guerby <laurent@guerby.net> - 03:19 PM Revision 639c9818 (ceph): documentation: explain ceph osd reweight vs crush weight
- Using the wording from Gregory Farnum at
http://lists.ceph.com/pipermail/ceph-users-ceph.com/2014-June/040961.html
S... - 02:50 PM rgw Bug #9543 (Rejected): AssertionError(s) in upgrade:dumpling-dumpling-distro-basic-vps run
- All in http://pulpito.front.sepia.ceph.com/teuthology-2014-09-19_11:48:54-upgrade:dumpling-dumpling-distro-basic-vps/...
- 02:47 PM CephFS Bug #8576: teuthology: nfs tests failing on umount
- Been playing around with this some.
- 02:47 PM CephFS Bug #9177 (Resolved): ceph-fuse: failing MPI mdtest runs
- John fixed this by updating mdtest in ceph-qa-suite as of commit:b1365a80982dba4160e861c28d887b066ca451b6.
- 02:32 PM Revision d535fab6 (ceph): daemons: write pid file even when told not to daemonize
- systemd wants to run daemons in foreground, but daemons wouldn't write
out the pid file with -f. Fixed.
Signed-off-... - 02:27 PM Bug #9301 (Pending Backport): paxos: off by one w/ versions in forming quorum
- 02:16 PM Revision f51e368f (ceph): erasure-code: fix assert overflow
- If the file size does not fit in 32 bits the (unsigned) cast will
overflow. Cast to uint64_t which is the type of the... - 01:31 PM Revision 13780d75 (ceph): mon: osd find / metadata --format plain fallback
- ceph --format plain osd find 1 (and metadata) are not implemented and
must fallback to the default (json-pretty).
ht... - 01:16 PM Revision d9ec7f2f (ceph): tasks: wait for mds active before mounting clients
- To make the logs clearer when trying to work out
if/when something went wrong, rather than always
having client logs ... - 01:16 PM Revision 9b1c3fe2 (ceph): suites: enable mds_client_limits test
- Requires same cluster etc as 'recovery', though
we could now do with better name than that.
Signed-off-by: John Spra... - 01:16 PM Revision 3e07bd1a (ceph): tasks: rename FuseMount.get_client_id to get_global_id
- 'client_id' was ambiguous because in other places it
meant the '0' in client.0, whereas here it means
the runtime-gen... - 01:15 PM Revision 72742895 (ceph): tasks: add mds_client_limits
- New CephFS tests for the behaviour of the system while
enforcing its resource limits.
Signed-off-by: John Spray <joh... - 01:13 PM Revision d777d712 (ceph): tasks: generalise CephFSTestCase
- Some of this stuff could be even more general for embedding
unittest-style suites, but for the moment let's keep the ... - 01:13 PM Revision 6f36269d (ceph): tasks: generalize config writing for Filesystem
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:05 PM Revision 0fb0967e (ceph): documentation: tiering typo
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 12:42 PM Bug #9537: OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo.get_total_chunk_si...
- Please note that while this ought to work in the technical sense, you are unlikely to be happy with RADOS if you make...
- 11:51 AM Bug #9537: OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo.get_total_chunk_si...
- OSD log of the primary OSD which crashed
- 07:17 AM Bug #9537 (Fix Under Review): OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo...
- https://github.com/ceph/ceph/pull/2534 to be confirmed by the OSD logs
- 07:05 AM Bug #9537 (Need More Info): OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo.g...
- Could you please attach the last 20,000 (twenty thousand) lines of the logs of the crashed primary OSD ?
- 03:52 AM Bug #9537: OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo.get_total_chunk_si...
- Config:
OSD nodes: 3
Monitor nodes: 2
Number of OSD's: 24
This is observed on 0.84 and is consistently getting ... - 03:33 AM Bug #9537 (Resolved): OSD crash after writing 10GB file onto EC Pool: FAILED assert(hinfo.get_tot...
On freshly created cluster, created an EC pool with default ec profile.
Wrote 5MB of object file using rados put...- 12:16 PM CephFS Feature #9284 (Resolved): mds: warn when clients are not responding to cache pressure
- Merged in giant...
- 12:03 PM rbd Feature #6228: image name metavariable
- Glad to see this feature added. Thank you! Mark, Adam, Sage, and Loic!
Assuming it wouldn't be too difficult, coul... - 09:47 AM rbd Feature #6228 (Pending Backport): image name metavariable
- 02:01 AM rbd Feature #6228 (Fix Under Review): image name metavariable
- 11:43 AM Revision c8885aec (ceph): Merge pull request #2505 from dachary/wip-8914-unlink-fdcache
- os: FileStore::lfn_unlink always clears FDCache
Reviewed-by: Sage Weil <sage@redhat.com> - 11:23 AM Revision 7ac60d8c (ceph): tests: check osd health in cephtool/test.sh
- Add a trivial osd health test at the beginning of each group of
tests. When facing an intermittent failure, it is dif... - 11:09 AM Revision 4d75c4c3 (ceph): tests: qa/workunits/cephtool/test.sh ! and -e
- From the bash man page:
set -e exit immediately ... The shell does not exit ... if the
command's return value ... - 11:04 AM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- I'd need the corresponding logs from osd.5 to be sure, but I believe the problem is that osd.5, due to 9497 and this ...
- 10:51 AM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- wip-sam-testing-firefly
- 10:51 AM Bug #9497: choose_acting has to let the pg be down any time acting < min_size even if there are b...
- wip-sam-testing-firefly
- 10:50 AM Bug #9481: osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- wip-sam-testing-firefly
- 10:50 AM Bug #9326 (Pending Backport): osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- 10:50 AM Bug #9240: osd_max_backfills = 1 can cause reserver deadlock for EC
- wip-sam-testing-firefly
- 10:50 AM Bug #9293: _collection_move_rename EEXIST
- wip-sam-testing-firefly
- 10:49 AM Bug #9179: unfound objects, recovery timeout
- wip-sam-testing-firefly
- 10:49 AM Bug #8777: osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == log.rbegin())
- wip-sam-testing-firefly
- 10:49 AM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- wip-sam-testing-firefly
- 10:49 AM Bug #9339: ReplicatedPG crash in hitset_create
- wip-sam-testing-firefly
- 09:23 AM Documentation #9542 (Won't Fix): Error link:"Ceph Object Gateway"->"Manual Install"
- In the page "Install Ceph Object Gateway"(http://ceph.com/docs/master/install/install-ceph-gateway/index.html), the "...
- 09:05 AM Documentation #8995 (Resolved): Preflight Checklist Clarifications
- http://ceph.com/docs/master/start/ Preflight was revamped significantly to address the comments and anticipate others.
- 09:05 AM CephFS Bug #9540 (Rejected): Crash during FS upgrade: assert(o->get_num_ref() == 0)
- Never mind, seems like this was just another manifestation of the original segment reference bug -- giant HEAD is OK.
- 06:37 AM CephFS Bug #9540: Crash during FS upgrade: assert(o->get_num_ref() == 0)
- The crash hits at the last ceph.restart (after upgrade from firefly to 83bd3430e3a17b77265e696095904b7a9032d2ee).
... - 06:33 AM CephFS Bug #9540 (Rejected): Crash during FS upgrade: assert(o->get_num_ref() == 0)
- ...
- 08:29 AM Revision be97b7d5 (ceph): documentation: fix bugous osd stop/start example
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 08:08 AM Bug #8863: osd: second reservation rejection -> crash
- does your build include commit:2b13de16c522754e30a0a55fb9d072082dac455e ?
- 07:44 AM RADOS Bug #9492 (Fix Under Review): Crush Mapper crashes when number of replicas is less than total num...
- https://github.com/ceph/ceph/pull/2528
- 07:35 AM Bug #9470 (Pending Backport): daemon pid file is not being created when running service ceph
- firefly backport : https://github.com/ceph/ceph/pull/2535
- 06:58 AM Linux kernel client Bug #9533 (Duplicate): kcephfs: fail to send requests initiated during mds restart
- this was an old bug, patch was missing from running kernel.
ceph: fix kick_requests()
- 06:56 AM Bug #9362: librados, rados_read corrupts memory on timeout
- I did another test today using the build from http://gitbuilder.ceph.com/ceph-deb-wheezy-x86_64-basic/ref/wip-dumplin...
- 06:36 AM Bug #9538: mon crashes on some --format=plain commands
- Checked all other uses of new_formatter allocated pointer in OSDMonitor
- 06:31 AM Bug #9538 (Fix Under Review): mon crashes on some --format=plain commands
- https://github.com/ceph/ceph/pull/2533
- 06:11 AM Bug #9538: mon crashes on some --format=plain commands
- 05:28 AM Bug #9538 (Resolved): mon crashes on some --format=plain commands
Mentioned by bens on IRC, creating ticket in case we forget:...- 06:33 AM rgw Feature #8911: RGW doesn't return 'x-timestamp' in header which is used by 'View Details' of Open...
- Hello Luis, et al..
I have a customer who's requesting status for this Feature.. They view it as a bug since it c... - 05:17 AM Bug #9536 (Fix Under Review): erasure-code: ISA plugin alignment must be constant
- * giant backport https://github.com/ceph/ceph/pull/2531
- 05:07 AM Bug #9536 (In Progress): erasure-code: ISA plugin alignment must be constant
- 02:57 AM Bug #9536 (Resolved): erasure-code: ISA plugin alignment must be constant
commit:28c2b6e4f2bc6d77b9150fcf9a917d85c69c9ed1
"EC_ISA_VECTOR_OP_WORDSIZE":https://github.com/ceph/ceph/blob/ma...- 04:52 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- scheduled "a monthrash":http://pulpito.ceph.com/ubuntu-2014-09-19_04:50:17-rados:monthrash-wip-9343-erasure-code-feat...
- 04:43 AM Fix #8914 (Resolved): osd crashed at assert ReplicatedBackend::build_push_op
- 04:13 AM Revision fb5a2449 (ceph): osd/Watch: handle con without sessions
- Signed-off-by: Sage Weil <sage@redhat.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 04:13 AM Revision 681a7548 (ceph): Pipe: stop delayed delivery fast_dispatch in stop_and_wait()
- If we don't, we can keep fast_dispatching messages after a Pipe has been
mark_down()ed. That breaks things right now.... - 04:13 AM Revision 3f23709c (ceph): Objecter: disable fast dispatch of CEPH_MSG_OSD_OPREPLY messages
- If they fast_dispatch they can deadlock right now. We'll need to change
the semantics around mark_down() or around th... - 04:13 AM Revision 5d37850c (ceph): osd: drop dead shutdown message handler
- There is no code anywhere in the code base that generates a MSG_SHUTDOWN.
Kill this. The get_priv() logic here looks... - 04:13 AM Revision 93dccdb3 (ceph): osd/ReplicatedPG: handle sessionless op con in do_osd_op_effects
- Signed-off-by: Sage Weil <sage@redhat.com>
Reviewed-by: Greg Farnum <greg@inktank.com> - 02:59 AM Bug #9485: Monitor crash due to wrong crush rule set
- Hi loic:
log, "ceph osd tree" output and crush map added.
log:
0> 2014-09-19 09:43:08.462737 7f92d9674700 -... - 01:58 AM Bug #9485: Monitor crash due to wrong crush rule set
- Hi,
It should not crash, it should give you an error of some kind maybe. Could you please attach to this ticket a ... - 02:11 AM Bug #9408: erasure-code: misalignment
- Running under the branch wip-9408-buffer-alignment in http://ceph.com/gitbuilder.cgi
- 01:55 AM Bug #7995: osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
- It reproed. PFA logs attached.
here is the snippet:
2014-09-19 10:27:02.228364 7f86d73a2700 0 log_channel(de... - 01:41 AM Bug #9535 (Duplicate): monitor crashed after restarting
- recently when i restarted my ceph cluster , the monitor crashed , below is the output of monitor log
2014-09-19 ... - 01:39 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Hi Sage, Thanks for the quick patch. I tried wip-9487-dumpling on our test cluster and now there is no snap trimming ...
09/18/2014
- 11:57 PM Revision bb45621c (ceph): common: Add cctid meta variable
- Fixes: #6228
Signed-off-by: Adam Crume <adamcrume@gmail.com> - 10:41 PM Revision 5fc04144 (ceph): Update giant.yaml
- Added comments(cherry picked from commit 3dd1314c6fa70fefc4ea8c65c33bac1ad3ba267a)
- 10:41 PM Revision a06d6747 (ceph): Update giant.yaml(cherry picked from commit 90b6eec23fad6b3af1305de9c59...
- 10:41 PM Revision 1af3281b (ceph): Update giant.yaml
- fixed typos(cherry picked from commit 1f950d2bf4dd76d671ab1c85a87a5f1175305635)
- 10:41 PM Revision 1140e18d (ceph): dumpling-giant-x: use giant version of workunits
- The firefly version of test_librbd_python at least does
not work on giant.
Signed-off-by: Josh Durgin <josh.durgin@i... - 10:41 PM Revision 403a8595 (ceph): Update giant.yaml(cherry picked from commit 35b3f4617c54dcb9705ca31489e...
- 10:41 PM Revision 84a63faa (ceph): Create dumpling-giant-x stress-split upgrade suite
- Fixes: 9357
Signed-off-by: Warren Usui <warren.usui@inktank.com>
(cherry picked from commit c729372aaae82b8fb0a31a42f... - 10:03 PM Revision 278f4dc7 (ceph): Merge pull request #143 from ceph/wip-migration-test
- tasks: more substantial IO for journal migration
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:01 PM Revision 9442336f (ceph): Merge pull request #2514 from ceph/wip-acl-test
- qa/workunits/fs/misc: Add a workunit for ACL
Reviewed-by: Greg Farnum <greg@inktank.com> - 09:43 PM Linux kernel client Bug #9533 (Duplicate): kcephfs: fail to send requests initiated during mds restart
- mds sees...
- 09:37 PM Bug #9202: Performance degradation during recovering and backfilling
- New ticket here - http://tracker.ceph.com/issues/9523
- 01:53 AM Bug #9202: Performance degradation during recovering and backfilling
- Hi Samuel,
Thanks for the short-term fix by tuning that 2 parameters of backfill scan. With tuning other backfill/... - 09:23 PM Revision c421b55e (ceph): mon: re-bootstrap if we get probed by a mon that is way ahead
- During bootstrap we verify that our paxos commits overlap with the other
mons we will form a quorum with. If they do... - 09:16 PM Bug #9481: osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- ceph cluster with 8 osd nodes each having 64 osds, few osds were crashing with this assert .As one node had timestamp...
- 11:01 AM Bug #9481 (Pending Backport): osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- 09:44 AM Bug #9481 (Fix Under Review): osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- 09:11 PM Revision d81cd7f8 (ceph): mon/Paxos: fix off-by-one in last_ vs first_committed check
- peon last_committed + 1 == leader first_committed is okay. Note that the
other check (where I clean up whitespace) g... - 09:03 PM Revision 9132ca47 (ceph): rbd-fuse: Fix memory leak in enumerate_images
- Fixes: #5768
Signed-off-by: Adam Crume <adamcrume@gmail.com> - 08:42 PM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- Captured debug log with wip-log-crash-firefly branch and attached.
- 01:18 PM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- sjust believes I may have hit the same bug, running 0.80.5. Attached is the log from an OSD with settings:...
- 12:55 PM Bug #9482 (Pending Backport): osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head...
- 09:44 AM Bug #9482 (Fix Under Review): osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head...
- Can't set info.last_epoch_started there, going to just use history.last_epoch_started as lower bound.
- 08:03 PM Revision 9a1ab951 (ceph): rbd: Fix rbd diff for non-existent objects
- Fixes: #6926
Signed-off-by: Adam Crume <adamcrume@gmail.com> - 07:55 PM Revision 0519225b (ceph): Merge pull request #2521 from athanatos/wip-9453
- osd_types: fix pg_log_t tail dump
Reviewed-by: Sage Weil <sage@redhat.com> - 07:55 PM Revision b4eb0e8f (ceph): Merge pull request #2519 from athanatos/wip-9482
- PG::find_best_info: let history.last_epoch_started provide a lower bound
Reviewed-by: Sage Weil <sage@redhat.com> - 07:16 PM Bug #9485: Monitor crash due to wrong crush rule set
- Hi, loic.
Currently I'm running some tests on my dev envrionment, after the tests are finished, I will reproduce i... - 06:42 PM CephFS Feature #9189 (Resolved): Expose client identifying metadata to MDS, e.g. hostname
- 06:42 PM Revision 9b8bbc04 (ceph): Merge pull request #2520 from athanatos/wip-9497
- PG::choose_acting: let the pg go down if acting is smaller than min_size
Reviewed-by: Sage Weil <sage@redhat.com> - 06:33 PM Feature #9532 (Duplicate): rados.py should export omap interface
- IWBN to be able to manipulate omap values with Python
- 06:03 PM Feature #8188 (In Progress): librados: interface to inspect pool properties
- 06:02 PM Revision e7a099b2 (ceph): Merge pull request #2518 from athanatos/wip-9481
- PGLog::claim_log_and_clear_rollback_info: fix rollback_info_trimmed_to
Reviewed-by: Sage Weil <sage@redhat.com> - 06:00 PM Revision 336d905e (ceph): Merge pull request #2522 from athanatos/wip-8315
- ReplicatedPG: cancel cb on blacklisted watcher
Reviewed-by: Sage Weil <sage@redhat.com> - 05:52 PM Revision d6913ae4 (ceph): mds: fix not journaling client metadata
- Previously the code was there for storing in
the SessionMap table, but not for the ESession
logevent.
Fixes: #9518
... - 05:09 PM rgw Bug #9529 (Resolved): ./common/ceph_crypto.h: 83: FAILED assert(s == SECSuccess)
- ...
- 05:01 PM Bug #9487 (Fix Under Review): dumpling: snaptrimmer causes slow requests while backfilling. osd_s...
- wip-9487
wip-9487-dumpling for backport - 03:23 PM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Nevermind, I've reproduced it!
- 03:21 PM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Thanks Sage. There's a log with debug_osd=20 attached to this issue. I'll try tomorrow to get one with debug_ms=1 too.
- 03:08 PM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Okay, I can't seem to reproduce this.
Dan or Florian, can you attach a log? What I need is debug ms = 1 and debug... - 02:45 PM Bug #9487 (In Progress): dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_t...
- Dan van der Ster wrote:
> I also noticed that before the snap trimmer starts, purge_snaps is [] for 5.318. Is that n... - 02:52 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Please comment on https://github.com/ceph/ceph/pull/2516.
Thanks! - 04:47 PM Revision c4685075 (ceph): PGLog::claim_log_and_clear_rollback_info: fix rollback_info_trimmed_to
- We have been setting it to the old head value. This is usually
harmless since the new head will virtually always be ... - 04:46 PM Revision 112f6f02 (ceph): Merge remote-tracking branches 'origin/wip-9497' and 'origin/wip-9482' ...
- 04:28 PM Revision 5c8143e3 (ceph): Merge pull request #2457 from ceph/wip-7767-b
- msg: prefetch data when doing recv()
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Samuel Just <sam.just@... - 04:26 PM Revision a8e6de30 (ceph): Merge pull request #2440 from somnathr/wip-optracker-optimization
- Wip optracker optimization
Reviewed-by: Samuel Just <sam.just@inktank.com> - 04:25 PM Revision 95768963 (ceph): Merge pull request #2445 from wonzhq/req-pending
- Cache tiering: do not allow flush/evict when there are requests pending
Reviewed-by: Samuel Just <sam.just@inktank.com> - 04:17 PM Bug #9528: RadosModel assertion failure in firefly
- also this one,
log: http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-11_23:20:03-multi-version-master-testi... - 04:14 PM Bug #9528 (Duplicate): RadosModel assertion failure in firefly
- This is basically firefly client running against the dumpling cluster.
logs: http://qa-proxy.ceph.com/teuthology/t... - 03:43 PM Bug #9517 (Resolved): Errors in test_rbd.* tests in upgrade:dumpling-giant-x:parallel-giant-distr...
- This was due to ceph-qa-suite updates not being on the giant branch.
- 03:29 PM rbd Feature #6228: image name metavariable
- Yeah, that is probably a good idea anyway.. we've had uniqueness issues like this before! That is an easy thing and ...
- 03:21 PM rbd Feature #6228: image name metavariable
- It's not perfect, but we could add a cctid variable so users could specify something like "admin socket = /var/run/ce...
- 03:11 PM rbd Feature #6228: image name metavariable
- This assumes that each image has its own cct, but a process could have multiple images open in one cct. (In fact, co...
- 02:07 PM rbd Feature #6228 (In Progress): image name metavariable
- 03:01 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Sage, a log is at https://www.dropbox.com/s/f2xyx12y2zr7fid/ceph-osd.14.log.xz -- behold the awesomeness of xz; that ...
- 02:49 PM Bug #9503 (Duplicate): Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Florian Haas wrote:
> Sage, I do have logs (@debug osd=20@, though not @debug ms=1@), but after the discussion with ... - 02:46 PM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Sage, I do have logs (@debug osd=20@, though not @debug ms=1@), but after the discussion with Dan on the -devel list,...
- 02:35 PM Bug #9503 (Need More Info): Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Hi Florian-
Can you generate some OSD logs (debug ms = 1, debug osd = 20) and attach them to the bug? The message... - 02:23 PM Bug #9301: paxos: off by one w/ versions in forming quorum
- 02:07 PM rbd Bug #5768: rbd-fuse: leak in enumerate_images()
- 01:49 PM rbd Bug #5768 (In Progress): rbd-fuse: leak in enumerate_images()
- 02:05 PM Bug #9462: msgr deadlock: osd reply vs mark_down vs fault
- 12:55 PM Bug #9453 (Resolved): ceph_objectstore_tool incorrect log tail output for --op log
- 09:45 AM Bug #9453 (Fix Under Review): ceph_objectstore_tool incorrect log tail output for --op log
- 12:53 PM Revision ab545ad0 (ceph): Merge pull request #2515 from ceph/wip-9518
- mds: fix not journaling client metadata
Reviewed-by: Yan, Zheng <ukernel@gmail.com> - 11:51 AM rbd Feature #7746 (In Progress): Capacity Management: rbd df
- see wip-7746
- 11:46 AM Feature #9526 (Resolved): mon: 'osd crush rename-bucket <old> <new>'
- 11:42 AM Bug #9497 (Pending Backport): choose_acting has to let the pg be down any time acting < min_size ...
- 09:43 AM Bug #9497 (Fix Under Review): choose_acting has to let the pg be down any time acting < min_size ...
- 11:26 AM rgw Bug #9525 (Duplicate): Deleted object shows in object listing
What appears to happen is that a request to delete an object comes in while the cluster is in a terrible state perf...- 11:16 AM rgw Bug #9169: 100-continue broken for centos/rhel
- Similar issue in suite:upgrade:firefly
http://pulpito.front.sepia.ceph.com/teuthology-2014-09-17_19:00:01-upgrade:... - 11:04 AM rgw Bug #9479: ETag is not included in the XML response to put object copy operation
- This is under v0.67.10
- 11:04 AM rgw Bug #9478: Incorrect content type in response header
- This is under v0.67.10
- 11:00 AM Bug #8315 (Pending Backport): osd: watch callback vs callback funky
- 09:40 AM Bug #8315 (Fix Under Review): osd: watch callback vs callback funky
- 09:40 AM rbd Bug #6926 (In Progress): rbd: diff output includes previously non-existent objects as zeroed extents
- 09:40 AM Bug #9326 (Fix Under Review): osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- 09:37 AM Feature #7767 (Resolved): messenger:buffer reads
- 09:29 AM Revision 1395275e (ceph): mds: fix not journaling client metadata
- Previously the code was there for storing in
the SessionMap table, but not for the ESession
logevent.
Fixes: #9518
... - 07:58 AM CephFS Feature #9437 (In Progress): make 'ceph tell mds.* ...' work, deprecate 'ceph mds tell * ...'
- 07:56 AM Revision 346c06c1 (ceph): client: preserve ordering of readdir result in cache
- Preserve ordering of readdir result in a list, so that the result of cached
readdir is consistant with uncached readd... - 07:55 AM Revision 600af254 (ceph): client: introduce a new flag indicating if dentries in directory are so...
- When creating a file, Client::insert_dentry_inode() set the dentry's offset
based on directory's max offset. The offs... - 06:20 AM Revision 59c89760 (ceph): qa/workunits/fs/misc: Add a workunit for ACL
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 06:16 AM CephFS Feature #9477: Handle kclient shutdown with dead network more gracefully
- In the general case (e.g. root filesystem is cephfs) there's nothing we can do: the system can't shut down until the ...
- 05:56 AM CephFS Bug #9518 (Resolved): client metadata get lost after mds restart
- ...
- 02:30 AM CephFS Bug #9518 (Fix Under Review): client metadata get lost after mds restart
- Well, I also shouldn't have missed it while writing the code :-)
https://github.com/ceph/ceph/pull/2515 - 04:01 AM RADOS Bug #9523 (Closed): Both op threads and dispatcher threads could be stuck at acquiring the budget...
- When OSD is rejoining and peering, we still see some slow requests and performance downgradation in about 5 to 10 min...
- 02:55 AM Revision 051cc7d5 (ceph): Merge pull request #2503 from ceph/wip-xattr-fix
- Wip xattr fix
- 02:08 AM Bug #8863: osd: second reservation rejection -> crash
- Two osds were down and out due to that crash, I was not able to start those osds again. So removed those osds and add...
- 01:53 AM Revision 2bd7ceef (ceph): client: trim unused inodes before reconnecting to recovering MDS
- So the recovering MDS does not need to fetch these ununsed inodes during
cache rejoin. This may reduce MDS recovery t... - 01:13 AM Linux kernel client Bug #9507: calling llistxattr(2) on a symlink crashes the client
- ...
09/17/2014
- 11:59 PM CephFS Bug #9504 (Duplicate): failed to decode message of type 24 v2: buffer::end_of_buffer
- looks like this is duplicate of #9458
- 08:23 AM CephFS Bug #9504 (Duplicate): failed to decode message of type 24 v2: buffer::end_of_buffer
- root@burnupi21:~# less /var/log/upstart/ceph-mds-ceph_burnupi21.log
... - 11:57 PM Linux kernel client Bug #9458: client wrongly fenced
- is the client using 3.16 kernel? possibly due to missing following commit...
- 02:45 PM Linux kernel client Bug #9458: client wrongly fenced
- The kernel client is definitely doing something wrong here, but I don't know what — the userspace messenger is not in...
- 02:38 PM Linux kernel client Bug #9458: client wrongly fenced
- The MDS went into reconnect at 4:59:50...
- 11:09 AM Linux kernel client Bug #9458: client wrongly fenced
- Taking a look; luckily we have at least *some* of the logging...
- 08:17 AM Linux kernel client Bug #9458: client wrongly fenced
- mds restarted and teuthology failed to reconnect again, 07:30:34.485721
- 07:18 AM Linux kernel client Bug #9458: client wrongly fenced
- teuthology was fenced again. not sure it was during a mds restart this time, either. notably the monitors went offl...
- 10:52 PM Bug #8863: osd: second reservation rejection -> crash
- Even i got the above crash, when few osds were in nearfull situation.
Snippet of logs:
2014-09-17 17:29:41.69... - 10:05 PM Revision d1f9f102 (ceph): suites: turn on debugging for smoke cfuse pjd test
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 09:07 PM Revision 1f950d2b (ceph): Update giant.yaml
- fixed typos
- 09:01 PM Revision 90b6eec2 (ceph): Update giant.yaml
- 09:01 PM Revision 3dd1314c (ceph): Update giant.yaml
- Added comments
- 08:46 PM CephFS Bug #9518: client metadata get lost after mds restart
- Dur, shouldn't have missed that in review. :(
- 07:44 PM CephFS Bug #9518 (Resolved): client metadata get lost after mds restart
- 07:23 PM Revision 82998179 (ceph): Merge remote-tracking branch 'origin/giant'
- 07:02 PM Revision 95ee6998 (ceph): client: Replace client readahead logic with Readahead
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 07:02 PM Revision ad45d8c4 (ceph): librbd: Add read-ahead
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 07:00 PM Revision 48094187 (ceph): doc: Change listed RBD cache defaults to actual defaults
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 07:00 PM Revision 830373ef (ceph): osdc: Fix ObjectCounter::_readx perf counters
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 06:07 PM Revision cef34f42 (ceph): Merge pull request #2498 from ceph/wip-objecter-cancel
- osdc/Objecter: cancel timeout before clearing op->session
Reviewed-by: John Spray <john.spray@redhat.com> - 05:15 PM Revision f27ccd80 (ceph): Merge pull request #2513 from ceph/wip-decode-dump
- msg: dump messages we can' decode at a low devel level
Reviewed-by: Greg Farnum <greg@inktank.com> - 04:53 PM Revision 40896842 (ceph): msg: hexdump bad messages at debug level 1 (not 30)
- These are rare, and when they do happen, we want to see them in the
logs.
Signed-off-by: Sage Weil <sage@redhat.com> - 04:52 PM Revision 54955702 (ceph): msg: hexdump messages we can't decode at configurable debug level
- We had this hard-coded to 30. Make it tunable.
Signed-off-by: Sage Weil <sage@redhat.com> - 04:25 PM Bug #9517 (Resolved): Errors in test_rbd.* tests in upgrade:dumpling-giant-x:parallel-giant-distr...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-17_13:53:14-upgrade:dumpling-giant-x:parallel-gian...
- 03:31 PM Bug #9452 (Resolved): All tests failed in upgrade:dumpling-giant-x:parallel-master-distro-basic-m...
- Looks like we passed those issues
#9515 might be realted - 03:29 PM Bug #9515 (Duplicate): "Segmentation fault (ceph_test_rados_api_io)" in upgrade:dumpling-giant-x:...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-17_13:53:14-upgrade:dumpling-giant-x:parallel-gian...
- 03:05 PM CephFS Bug #9514 (Resolved): ceph-fuse pjd test is failing in giant nightlies
- commit:0ea20a668cf859881c49b33d1b6db4e636eda18a
http://qa-proxy.ceph.com/teuthology/sage-2014-09-14_18:23:49-smoke... - 02:43 PM rbd Bug #9513 (Resolved): rbd_cache=true default setting is degading librbd performance ~10X in Giant
- We are experiencing severe librbd performance degradation in Giant over firefly release. Here is the experiment we di...
- 02:33 PM Bug #8885: SIGABRT in TrackedOp::dump() via dump_ops_in_flight()
- It's the same issue as #9384
Here is the pull request for the same.
https://github.com/ceph/ceph/pull/2440 - 01:17 PM Bug #9508 (Resolved): objecter: segv on timeout/cancel (LibRadosIo ReadTimeout)
- commit:cef34f429972267061fc0e730ef976887ccb78a9
- 10:22 AM Bug #9508 (Fix Under Review): objecter: segv on timeout/cancel (LibRadosIo ReadTimeout)
- https://github.com/ceph/ceph/pull/2498
- 09:59 AM Bug #9508 (Resolved): objecter: segv on timeout/cancel (LibRadosIo ReadTimeout)
- ...
- 01:05 PM Revision 7e13ac8e (ceph): rgw: Export user stats in get-user-info Adminops API
- Fixes: #9359
Signed-off-by: Ray Lv <raylv@yahoo-inc.com> - 12:59 PM Revision d8630cf4 (ceph): Merge pull request #2512 from dachary/wip-9490-crushtool-num-rep
- Revert "crushtool: safeguard for missing --num-rep when --test"
Reviewed-by: John Spray <john.spray@redhat.com> - 12:56 PM Revision 480c3727 (ceph): Revert "crushtool: safeguard for missing --num-rep when --test"
- This reverts commit 10b8966c8da38f8ae5ac7dbb2e16a5e45e502ba3.
- 12:50 PM Revision 65a4141e (ceph): Merge remote-tracking branch 'origin/giant'
- 12:47 PM Revision 51eeb334 (ceph): Merge pull request #2496 from ceph/wip-9189
- #9375 and #9189 - add client metadata and improve health messages
Reviewed-by: Greg Farnum greg@inktank.com - 12:37 PM Revision 7d086403 (ceph): tasks: escaping '*' when deleting files
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:21 PM Revision 32b51bb7 (ceph): mds: generate friendly name for client sessions
- ...from the client metadata, for use in health messages.
Signed-off-by: John Spray <john.spray@redhat.com> - 12:21 PM Revision 6320e535 (ceph): mds: implement const SessionMap::get_session
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:21 PM Revision 288f05a5 (ceph): mds: use new Session::human_name in health msgs
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:21 PM Revision bdda2377 (ceph): ceph_fuse: refactor init to make mount_point accessible
- ...so that we can read out the mount point between init
and start() in order to feed it to Client before it connects
... - 12:21 PM Revision bd8c4413 (ceph): client: send metadata in session open
- Populated with hostname and entity_id by
default, with interface for outer layers
like ceph_fuse, libcephfs to inject... - 12:21 PM Revision 920cac90 (ceph): ceph_fuse: include mount point in client metadata
- Signed-off-by: John Spray <john.spray@redhat.com>
- 11:05 AM Bug #9509: init script cannot stop OSDs
- Yep, it needs to be backported to Firefly
- 11:01 AM Bug #9509 (Pending Backport): init script cannot stop OSDs
- See #9470. I guess the commit probably needs to be backported to firefly?
- 10:57 AM Bug #9509: init script cannot stop OSDs
- Let me redo the last sentence...
One user reported the issue on CentOS 7 and I managed to reproduce it. I assume i... - 10:48 AM Bug #9509 (Resolved): init script cannot stop OSDs
- Running a @service ceph stop osd@ will not stop OSDs.
It seems the problem is that the OSDs are launched with the ... - 11:04 AM devops Bug #9510 (Closed): ceph-deploy: Move mon keyring generation 'mon create-initial'
- Right now the monitor keyring is generated with 'ceph-deploy new', in cases where an admin wants to use a pre-existin...
- 10:26 AM Revision 1a9e3cc9 (ceph): mds: store client metadata on Session
- Signed-off-by: John Spray <john.spray@redhat.com>
- 10:26 AM Revision 758a7cea (ceph): mds: include client metadata in `session ls`
- Signed-off-by: John Spray <john.spray@redhat.com>
- 10:26 AM Revision cd215c75 (ceph): messages: add cephfs client metadata to MClientSession
- Signed-off-by: John Spray <john.spray@redhat.com>
- 10:25 AM Revision 3025b5f0 (ceph): mds: collapse numerous client warnings into one
- ...to avoid sending O(N_clients) sized beacons.
Fixes: #9375
Signed-off-by: John Spray <john.spray@redhat.com> - 10:23 AM Bug #9501: Assertion in FileJournal::do_write
- Don't worry, Sam says this is some kernel bug in btrfs, but he hasn't told the rest of us about it yet.
- 04:03 AM Bug #9501: Assertion in FileJournal::do_write
- Urgh, I have stupidly just killed that job before making a copy of the logs.
- 03:56 AM Bug #9501 (Rejected): Assertion in FileJournal::do_write
- ...
- 09:48 AM Linux kernel client Bug #9507 (Resolved): calling llistxattr(2) on a symlink crashes the client
- The code hits a "BUG();" line at https://github.com/ceph/ceph-client/blob/7e8a295295775ec9e05411cefc578ff4bfc94740/fs...
- 09:33 AM devops Bug #9506 (Rejected): Pass monitor SSH addresses via CLI flag
- In some network configurations it is desirable to have ceph-deploy access monitors from one network, and use another ...
- 08:51 AM Linux kernel client Bug #9505 (Duplicate): kcephfs: client gets stuck in reconnect loop?
- ...
- 08:37 AM Bug #9503: Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Added issue #9487 as *possibly* related.
- 07:52 AM Bug #9503 (Resolved): Dumpling: removing many snapshots in a short time makes OSDs go berserk
- Back in March, there was a report from Craig Lewis on the users list that mentioned several OSDs going to 100% CPU fo...
- 07:10 AM Bug #9502 (Resolved): mon: does not verify disk is not full on startup
- mira040...
- 07:06 AM Revision 66deda4b (ceph): Merge pull request #2508 from dachary/wip-9490-crushtool-num-rep
- crushtool: safeguard for missing --num-rep when --test
Reviewed-by: Johnu George <johnugeorge109@gmail.com> - 06:09 AM rgw Feature #9359: rgw: Export user stats in get-user-info Adminops API
- Updated PR with a new commit to resolve Yehuda's comments. Please help to review it.
- 06:08 AM Bug #9490 (Rejected): crushtool crash if --num-rep is missing
- The root of the problem is #9492 : when --num-rep is missing it defaults to the range defined in the rule and does th...
- 05:58 AM Bug #9490 (In Progress): crushtool crash if --num-rep is missing
- 06:03 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- I also noticed that before the snap trimmer starts, purge_snaps is [] for 5.318. Is that normal, or should (the compl...
- 05:48 AM CephFS Feature #9189: Expose client identifying metadata to MDS, e.g. hostname
- Userspace part merged:...
- 05:47 AM CephFS Feature #9375 (Resolved): Send single 'many clients' health warning instead of N warnings for N c...
- ...
- 03:56 AM Revision 0769310c (ceph): PGLog::claim_log_and_clear_rollback_info: fix rollback_info_trimmed_to
- We have been setting it to the old head value. This is usually
harmless since the new head will virtually always be ... - 03:52 AM Revision 07f54f28 (ceph): PG::find_best_info: let history.last_epoch_started provide a lower bound
- If we find a info.history.last_epoch_started above any
info.last_epoch_started, we must be missing updates and
min_la... - 03:39 AM rgw Bug #9500 (Duplicate): 0.80.5 on CentOS 6.5: radosgw-admin fails to correctly name subuser object
- System info: Firefly (0.80.5 on CentOS 6.5). radosgw is configured and working fine with s3cmd.
Symptom: despite t... - 03:37 AM Revision 92cfd370 (ceph): PG::choose_acting: let the pg go down if acting is smaller than min_size
- Even if the backfill peer would bring us up to min_size, we can't go
active since build_prior will not consider the i... - 03:32 AM Revision 1f450faa (ceph): client: request xattrs if xattr_version is 0
- Following sequence of events can happen.
- Client releases an inode, queues cap release message.
- A 'lookup' reply... - 01:38 AM Revision c837fb94 (ceph): mds: return xattrs when client explictly requests xattrs
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 01:23 AM Bug #8083 (In Progress): erasure-code: fix static code analysis errors found in gf-complete
- A number of fixes already are in gf-complete master and "added two":https://bitbucket.org/jimplank/gf-complete/pull-r...
- 12:37 AM Revision eb1b2e00 (ceph): Merge pull request #2507 from ceph/wip-9435
- #9435 - enforce rules for using EC/tiered pools with CephFS
Reviewed-by: Greg Farnum <greg@inktank.com> - 12:16 AM Revision 80441cda (ceph): mon: prevent cache pools being used CephFS
- Fixes two things:
* EC pools are now permissible if they have a cache overlay
* Pools are not permissible if they a... - 12:16 AM Revision 681a49c4 (ceph): mon: forbid tier changes when in use by FS
- * Removing tiers from a base pool in use by CephFS is forbidden.
* Using CephFS pools as tiers is forbidden.
Signed-...
09/16/2014
- 11:27 PM Bug #9488 (Rejected): Writing object onto EC pool created with customized ec profile getting hung
- k=1 m=1 is not supposed to work, it won't do anything useful. k=5 m=3 totals 8 osds and you only have 6 hence it blocks.
- 08:32 PM Bug #9488: Writing object onto EC pool created with customized ec profile getting hung
- Hi Loic,
I have 3 OSD hosts and total of 6 OSD's.
ems@rack6-client-5:~$ sudo ceph osd crush rule dump
[
{... - 08:36 AM Bug #9488 (Need More Info): Writing object onto EC pool created with customized ec profile gettin...
- It is the normal behavior when there are not enough hosts to satisfy the crush rules. Do you have 22 hosts available ...
- 05:14 AM Bug #9488: Writing object onto EC pool created with customized ec profile getting hung
- Attaching logs
- 05:09 AM Bug #9488: Writing object onto EC pool created with customized ec profile getting hung
- This issue is observed on ceph 0.84
- 05:07 AM Bug #9488 (Rejected): Writing object onto EC pool created with customized ec profile getting hung
- Writing object onto EC pool created with customized EC profile is getting hung.
But, writing object onto EC pool wit... - 11:15 PM Bug #9219 (Resolved): lost_unfound test got ENOENT: i don't have pgid 1.e
- 05:38 PM Bug #9219: lost_unfound test got ENOENT: i don't have pgid 1.e
- Merged into giant by commit:782848af596fdb0be57daa68481b3976b7119141.
- 10:36 PM Revision 86a4bed6 (ceph): FileStore: Race condition during object delete is fixed
- There was a race condition (hence OSD crash) between lfn_unlink
and lfn_open. The reason was FDCache lookup was calle... - 10:14 PM devops Bug #9499 (Can't reproduce): osds do not start after reboot (centos7, dm-crypt)
- most osds do not come up after reboot; only one does.
adding a 'sleep 10 ; ceph-disk activate-all' to /etc/rc.loca... - 09:57 PM devops Bug #9498 (Resolved): el7 still using crappy el6 udev rules
- 08:33 PM Bug #9497: choose_acting has to let the pg be down any time acting < min_size even if there are b...
- 08:33 PM Bug #9497 (Resolved): choose_acting has to let the pg be down any time acting < min_size even if ...
- Otherwise, build_prior won't realize that the interval was maybe_went_rw.
- 07:18 PM Revision 10b8966c (ceph): crushtool: safeguard for missing --num-rep when --test
- http://tracker.ceph.com/issues/9490 Fixes: #9490
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 07:15 PM Revision ea48a21e (ceph): Merge pull request #2475 from ceph/wip-lttng
- Update rbd-replay
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 07:05 PM RADOS Bug #9492: Crush Mapper crashes when number of replicas is less than total number of osds to be s...
- The issue is that crush temporary buffers(scratch array) are allocated as per size of num_replica configured by the ...
- 05:28 PM RADOS Bug #9492: Crush Mapper crashes when number of replicas is less than total number of osds to be s...
- Seg fault log:
CRUSH*** Caught signal (Segmentation fault) **
in thread 7f3dcb0007c0
ceph version 0.85-778-gb285... - 12:37 PM RADOS Bug #9492 (Resolved): Crush Mapper crashes when number of replicas is less than total number of o...
- 1. ./crushtool --outfn crushmap --build --num_osds 100 host straw 4 rack straw 10 default straw 0
2../crushtool -d c... - 06:47 PM Revision 13b82ded (ceph): Merge pull request #2473 from ceph/wip-doc-preflight
- doc: Revised the Preflight Doc
Reviewed-by: John Wilkins <jowilkin@redhat.com> - 06:36 PM Bug #9496 (Resolved): mon: pg scrub timestamps must be populated at pg creation
- logs: ubuntu@teuthology:/a/teuthology-2014-09-15_16:05:01-upgrade:firefly-giant-x:parallel-giant-distro-basic-multi/4...
- 06:28 PM Revision 9a65b390 (ceph): doc: Fixed broken hyperlink.
- Signed-off-by: John Wilkins <jowilki@redhat.com>
- 06:01 PM Revision 399fa806 (ceph): script/run-coverity: update submodules
- Signed-off-by: Sage Weil <sage@redhat.com>
- 05:37 PM CephFS Fix #9435 (Resolved): prevent use of cache pools as metadata or data pools
- Merged into giant branch in commit:eb1b2e0072bf605095f4104c2b6c2abfba216dbe
- 02:57 AM CephFS Fix #9435 (Fix Under Review): prevent use of cache pools as metadata or data pools
- https://github.com/ceph/ceph/pull/2507
- 03:46 PM Bug #9480: OSD is crashing while object deletion
- Created the following pull request for the fix.
https://github.com/ceph/ceph/pull/2510 - 02:50 PM rgw Feature #9493 (Resolved): Ability to disable keystone revocation polling when using UUID keystone...
- When using a UUID keystone provider revocation is handled by deleting the token from the persistence backend (ie. no ...
- 02:16 PM CephFS Feature #9466: kclient: Extend CephFSTestCase tests to cover kclient
- Got these passing at least once by hand using IPMI to work around #9477, suite scheduled:
http://pulpito.front.sep... - 02:14 PM Revision 366ee005 (ceph): tasks: more substantial IO for journal migration
- ...so that there will at least be multiple segments
in the log during the rewrite.
Also make the test stricter by ch... - 01:26 PM Revision cf17111c (ceph): Merge pull request #2394 from ceph/wip-da-fix-make_check
- Fix some issues with make check
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 01:22 PM Revision 52c5c32f (ceph): Merge pull request #2504 from dachary/wip-erasure-code-valgrind
- test: libtool --mode=execute valgrind
Reviewed-by: Sahid Orentino Ferdjaoui <sahid.ferdjaoui@redhat.com> - 01:12 PM Revision fdbfece8 (ceph): Merge pull request #2497 from ceph/wip-xfs-inode64
- ceph-disk: mount xfs with inode64 by default
Reviewed-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 11:49 AM Documentation #8995: Preflight Checklist Clarifications
- Addressed clarifications and also added preflight material for other distributions.
- 11:47 AM Documentation #9475 (Resolved): Broken links on downloads page
- Resolved by Ross Turk.
- 11:36 AM Documentation #9491 (Closed): Radosgw docs incorrectly state to disable print continue on centos ...
- https://ceph.com/docs/master/radosgw/config/ states:
"On CentOS/RHEL distributions, turn off print continue. If yo... - 11:31 AM Bug #9490 (Fix Under Review): crushtool crash if --num-rep is missing
- https://github.com/ceph/ceph/pull/2508
- 11:15 AM Bug #9490: crushtool crash if --num-rep is missing
- crash occurs when num-rep takes the value 1
- 11:00 AM Bug #9490 (Rejected): crushtool crash if --num-rep is missing
- ...
- 10:13 AM devops Bug #9489: --zap-disk does not clear enough
- it's worth noting that the OSD worked fine in the cluster after initial deployment, it's not until the node is reboot...
- 10:05 AM devops Bug #9489 (Rejected): --zap-disk does not clear enough
- sometime the partitions are resurected
- 10:07 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Andrei,
No, I haven't, but plan to try harder. I am however seeing an extreme slowdown, will open a ticket to tak... - 02:49 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Ilya,
I was wondering if you've managed to verify my findings? Has anyone experienced similar behaviour as I am?
... - 10:00 AM devops Bug #5929 (Resolved): debian: python-ceph should depend on libcephfs1
- This was comitted quite a long time ago during a bug scrub I believe.
- 10:00 AM Revision 123c3fbb (ceph): test/mon/mkfs.sh: add check for default keyring
- The auth_cephx_key check always fails if there is a default
keyring in /etc/ceph/ available. Check if keyring exists ... - 10:00 AM Revision 8ea86dfa (ceph): test/ceph-disk.sh: resolve symlinks before check
- Make sure symlinks are resolved in command_fixture()
before compare result of which command and the current
path.
Si... - 09:51 AM Revision 8b27997b (ceph): mon: pool create must not always create a ruleset
- The implicit creation of a ruleset when creating a pool is convenient
when nothing is specified. However, if the call... - 09:27 AM Revision d5084f3f (ceph): mon: add the get_crush_ruleset helper
- By factoring a code snippet from prepare_pool_crush_ruleset with no
modification.
Signed-off-by: Loic Dachary <loic-... - 09:12 AM RADOS Fix #6109: pg <pgid> mark_unfound_lost fails if a completely-gone OSD still in map
- I'm having a similar issue, I have one unfound object that I can't delete. I'm also getting the "Error EINVAL: pg has...
- 08:55 AM Revision f952f453 (ceph): Merge pull request #2487 from wonzhq/org
- mailmap: correcting Zhiqiang Wang's mailmap and org affiliation
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 08:46 AM Bug #9438: librados API generated doc broken
- I'd be happy to review : which pull request / branch is it ?
- 08:46 AM Revision daf63798 (ceph): mailmap: correcting Zhiqiang Wang's mailmap and org affiliation
- Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
- 08:38 AM Bug #9485 (Need More Info): Monitor crash due to wrong crush rule set
- Could you add the stack trace of the mon crash to the ticket ? I remember the discussion we had on the mailing list a...
- 08:36 AM Revision 4b8c50fe (ceph): tests: flush logs before grepping them
- Otherwise the test races with the daemon writing the logs and can
sometime fail.
Signed-off-by: Loic Dachary <loic-2... - 08:28 AM Revision 45731dbc (ceph): os: FileStore::lfn_unlink always clears FDCache
- Otherwise the FDCache will keep a file descriptor to a file that was
removed from the file system. This may create va... - 08:28 AM Revision 0ffc5ee5 (ceph): tests: set the failure domain to OSD by default
- So that tests do not need to do it to be able to use the default rbd
pool to store objects.
Signed-off-by: Loic Dach... - 08:28 AM Revision 191d67cb (ceph): tests: add get_osds() and get_pg() helpers
- To get the ordered list of OSD to which an object is mapped and the name
of the corresponding PG.
Signed-off-by: Loi... - 08:25 AM Revision 6d4e2ca0 (ceph): test: libtool --mode=execute valgrind
- otherwise valgrind will not actually run against the binary.
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 06:59 AM Revision 457790da (ceph): KeyValueStore: Reduce redundant "make_pair"
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 06:04 AM Revision 1a8b91b7 (ceph): doc: Add keyvaluestore config description
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 05:43 AM Revision bb49547d (ceph): KeyValueStore: Reduce redundancy set_header call
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 05:36 AM Revision 38f7af0a (ceph): s/tasks/workload
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 04:54 AM Revision 306fb2f5 (ceph): mds: set new inode's xattr version to 1
- set new inode's xattr version to 1 even if it has no xattr. This allow
client to differentiate no xattr in inode from... - 03:29 AM Bug #9434: rbd rm hangs
- Loic Dachary wrote:
> Version 0.71 was a development version. Are you observing the same version on a stable release... - 02:37 AM Bug #9304: pool create with invalid crush rule name succeeds
- "rebased against giant":https://github.com/ceph/ceph/pull/2506
- 02:33 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- Here is a bit more... I checked for "snap_trimmer entry" on other OSDs this morning. There were a few others, but all...
- 01:59 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- > I was able to isolate the cause of the backfilling to one single OSD
typo.. I was able to isolate the cause of ... - 01:47 AM Bug #9487: dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim_sleep not ...
- In case it wasn't clear, there is nothing special about osd.11. Each time I reweight 2 OSDs the slow requests are cau...
- 01:44 AM Bug #9487 (Resolved): dumpling: snaptrimmer causes slow requests while backfilling. osd_snap_trim...
- Hi,
using dumpling 0.67.10...
We are doing quite a bit of backfilling these days in order to make room for some n... - 12:40 AM Revision 782848af (ceph): Merge pull request #2499 from ceph/wip-9219-giant
- wip-9219: subscribe to the newest osdmap when reconnecting to a monitor
Reviewed-by: Sage Weil <sage@redhat.com> - 12:07 AM Revision 1b9226c7 (ceph): osd: subscribe to the newest osdmap when reconnecting to a monitor
- This is mostly relevant in testing clusters, but it ensures that an OSD
disconnecting from the monitor at the wrong t...
09/15/2014
- 11:45 PM Revision 56ba3411 (ceph): osdc/Objecter: fix command op cancellation race
- Cancel the command op timeout event before we clear out the op from the
session struct. This isn't strictly necessar... - 11:40 PM Revision baf7be9d (ceph): osdc/Objecter: cancel timeout before clearing op->session
- The C_CancelOp path assumes op->session != NULL. Cancel that op before
we clear it. This fixes a crash like
#0 pt... - 10:29 PM Revision 11496399 (ceph): ceph-disk: mount xfs with inode64 by default
- We did this forever ago with mkcephfs, but ceph-disk didn't. Note that for
modern XFS this option is obsolete, but f... - 10:23 PM Revision 8c23ef09 (ceph): Merge pull request #2492 from ceph/wip-9284
- #9284 - fix client RECALL handling and add health metrics
Reviewed-by: Greg Farnum <greg@inktank.com> - 10:23 PM Revision 47124f39 (ceph): Merge pull request #2482 from dachary/wip-erasure-code-documentation
- erasure code documentation
Reviewed-by: Sage Weil <sage@redhat.com> - 10:19 PM Revision 9d36d87c (ceph): Merge pull request #2476 from ceph/wip-9307
- rgw: push hash calculater deeper
Reviewed-by: Sage Weil <sage@redhat.com> - 08:37 PM Bug #9485 (Resolved): Monitor crash due to wrong crush rule set
- I create a customized crush rule for ec pool
1 set take default
2 choose firstn 6 type rack
3 chooseleaf firstn ... - 08:25 PM Revision 853ba2df (ceph): Merge pull request #2493 from ceph/wip-rbd-objectcacher-hang
- rbd: ObjectCacher reads can hang when reading sparse files
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 08:04 PM Feature #9222: annotate config options
- If we want to think in an internationalization direction, perhaps the right thing is to msg-catalog the help informat...
- 03:05 PM Feature #9222: annotate config options
- Yeah. I would also love to see min/max values for the numeric options.
- 02:56 PM Feature #9222: annotate config options
- If a fourth argument is set to a description string in config_opts.h, ceph.in could get access to it via a pybind/com...
- 07:59 PM Revision a67f9152 (ceph): librbd: fix crash using clone of flattened image
- The crash occurs due to ImageCtx->parent->parent being uninitialized,
since the inital open_parent() -> open_image(pa... - 06:36 PM Revision 3f020443 (ceph): librbd: fix crash using clone of flattened image
- The crash occurs due to ImageCtx->parent->parent being uninitialized,
since the inital open_parent() -> open_image(pa... - 06:26 PM Revision f2039c4e (ceph): Merge pull request #2495 from dachary/wip-erasure-code-preload
- erasure-code: preload fails if < 0
Reviewed-by: Sage Weil <sage@redhat.com> - 06:21 PM Revision ded1b303 (ceph): erasure-code: preload fails if < 0
- And not if < -1.
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 06:10 PM Revision 27208dba (ceph): doc: Added feedback.
- Signed-off-by: John Wilkins <jowilki@redhat.com>
- 05:39 PM Fix #9484: OSD: block until we have the same map as the client on pg commands
- Instead of blocking for *every* tell command (or even a subset), we can add one new command 'get_latest_osdmap' or si...
- 05:11 PM Fix #9484 (New): OSD: block until we have the same map as the client on pg commands
- Right now, if a client has a newer map than we do and sends a PG command (like list_missing, #9219) we can reply ENOE...
- 05:14 PM Bug #9219 (Fix Under Review): lost_unfound test got ENOENT: i don't have pgid 1.e
- I created a few other tickets for the specific pg command issue, and created a PR so the OSD will subscribe to any os...
- 04:49 PM Bug #9219: lost_unfound test got ENOENT: i don't have pgid 1.e
- Okay, so at the time osdmap 19 was created, we had two of three OSDs running (osd.1 was down and out, and teuthology ...
- 05:13 PM Revision 09beb4c7 (ceph): Merge pull request #141 from ceph/wip-9423
- Fix #9423 - race in mds_client_recovery
- 05:08 PM Feature #9483 (Resolved): OSD: add a get_newest_map command to the admin socket
- This could be useful in testing and to "unstick" clusters in some odd situations we've seen before.
- 04:43 PM Revision 2422f9fd (ceph): init-radosgw.sysv: Support systemd for starting the gateway
- When using RHEL7 the radosgw daemon needs to start under systemd.
Check for systemd running on PID 1. If it is then ... - 04:41 PM Revision 0eef2d1b (ceph): Merge pull request #2486 from jgalvez/master
- init-radosgw.sysv: Support systemd for starting the gateway
Reviewed-by: Sage Weil <sage@redhat.com> - 04:23 PM Revision 1941d7b6 (ceph): Merge pull request #2472 from dachary/wip-9429-bench
- erasure-code: fix erasure_code_benchmark goop (decode)
Reviewed-by: Janne Grunau <j@jannau.net> - 04:21 PM Revision b1365a80 (ceph): Merge pull request #140 from ceph/wip-9177
- suites/fs: update to latest mdtest
- 04:20 PM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- Yeah, pretty sure that's right, even if we only find backfill peers, we want to let those determine the min acceptabl...
- 04:06 PM Bug #9482: osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)
- Actually, I'm not sure that's right. Thinking.
- 04:02 PM Bug #9482 (Resolved): osd/PGLog.cc: 544: FAILED assert(log.head >= olog.tail && olog.head >= log....
- backfill peers are not setting info.last_epoch_started allowing subsequent primaries to erroneously conclude that it ...
- 03:39 PM Bug #9481 (Resolved): osd/PGLog.h: 87: FAILED assert(rollback_info_trimmed_to == head)
- Bug is PGLog::claim_log_and_clear_rollback_info sets rollback_info_trimmed_to before setting head.
- 03:27 PM Bug #9480: OSD is crashing while object deletion
- I have root caused it, it seems to be happening because one of my earlier changes :-( .. Here is the rot cause.
1.... - 03:00 PM Bug #9480 (Resolved): OSD is crashing while object deletion
- Reproducible step:
1. Run a command something like this.
rados bench -p rbench 200 write -t 32 -b 1024
The O... - 03:12 PM Bug #9109: ceph CLI: Help is missing -k keyring option
- Initial pull request:
https://github.com/ceph/ceph/pull/2483
Need to design a solution so that all clients can ... - 03:05 PM Bug #9109: ceph CLI: Help is missing -k keyring option
- not a low hanging fruit after all, johnu will try another ;-)
- 12:51 PM Bug #9109: ceph CLI: Help is missing -k keyring option
- So, really, this applies to all the "Ceph global" options that the frontend doesn't have reason to do anything specia...
- 03:03 PM Revision bda325b3 (ceph): suites/fs: update to latest mdtest
- They appear to have (accidentally?) fixed whatever
was crashing.
Fixes: #9177
Signed-off-by: John Spray <john.spray... - 02:05 PM Revision 00a00214 (ceph): client: fix trim_caps for inodes in root
- Previously client would fail to release caps for files
in the root directory in response to CEPH_SESSION_RECALL_STATE... - 02:05 PM Revision 8199f808 (ceph): xlist: implement const_iterator
- Signed-off-by: John Spray <john.spray@redhat.com>
- 02:05 PM Revision 8c0f2555 (ceph): mds: add state for tracking RECALL progress
- To be used later for generating health metrics
for clients which are failing to promptly service
CEPH_SESSION_RECALL_... - 02:05 PM Revision e6062b8d (ceph): mds: add a health metric for failure to recall caps
- Fixes: #9284
Signed-off-by: John Spray <john.spray@redhat.com> - 02:05 PM Revision 05d69580 (ceph): mon: trigger transaction on MDS health changes
- I think this was previously only working as a side effect
of other MDS map changes.
Signed-off-by: John Spray <john.... - 02:05 PM Revision fd04d5e6 (ceph): mds: health metric for late releasing caps
- Follow up on Yan Zheng's "mds: warn clients which
aren't revoking cap" to include a health metric
for this condition ... - 02:05 PM Revision a6a0fd81 (ceph): xlist: implement copy constructor
- ...so that I can have a std::map of them.
Signed-off-by: John Spray <john.spray@redhat.com> - 02:05 PM Revision bf590f8a (ceph): mds: keep per-client revoking caps list
- ...to avoid doing an O(caps) scan to find out
which clients are responsible for any late-revoking
caps during health ... - 02:05 PM Revision a140439f (ceph): mds: limit number of caps inspected in caps_tick
- This is to avoid hitting an O(caps) loop in the worst
cast scenario. This mechanism is a little crude but
should be ... - 02:05 PM Revision c328486f (ceph): client: fix crash in trim_caps
- In a75af4c2, procedure was added to invalidate root's dentries
if the trimming failed to free enough caps. This woul... - 02:05 PM Revision 90072172 (ceph): client: more precise cap trimming
- Two fixes:
* Client would unlink everything it could, instead of just
meeting its goal, because caps.size() doesn... - 02:05 PM Revision 21f5e18e (ceph): client: fix potentially invalid read in trim_caps
- trim_dentry can potentially free an inode, so get/put
it around the block where we use the inode's dn_set.
Signed-of... - 02:05 PM Revision 2b5bbab5 (ceph): client: failure injection for cap release
- Used for simulating a buggy client that trips
the error detection in #9282 (warn clients
which aren't revoking caps)
... - 02:02 PM CephFS Bug #9444 (Resolved): "unmatched rstat" exception after firefly->master upgrade
- if mds_verify_scatter isn't enabled, the MDS will fix rstat mismatch atomically.
- 10:45 AM CephFS Bug #9444: "unmatched rstat" exception after firefly->master upgrade
- I think you're right, John. I'm not sure why we never saw this before though — Zheng, what changed that we're looking...
- 02:45 AM CephFS Bug #9444: "unmatched rstat" exception after firefly->master upgrade
- Is this actually fixed, in the case of filesystems created using old code? It seems like the patch prevents creating...
- 01:40 PM Revision 68001fea (ceph): Merge pull request #2485 from Abioy/master
- bugfix: wrong socket address in log msg of Pipe.cc
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 01:34 PM Bug #9452: All tests failed in upgrade:dumpling-giant-x:parallel-master-distro-basic-multi run
- The main source of these problems should be fixed by commit:cdb7675a21c9107e3596c90c2b1598def3c6899f
- 01:33 PM rbd Bug #6494: High memory consumption of qemu/librbd with enabled cache
- FTR the commits fixing this are commit:4fc9fffc494abedac0a9b1ce44706343f18466f1 and commit:cdb7675a21c9107e3596c90c2b...
- 01:32 PM Revision 1d9101cf (ceph): tasks: fix race in test_stale_caps
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:04 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- First half here: https://github.com/ceph/ceph/tree/wip-9435 (no handling of tiering updates yet)
- 12:47 PM CephFS Fix #9435 (In Progress): prevent use of cache pools as metadata or data pools
- 12:44 PM Revision bc257677 (ceph): tasks: handle failure cleanly in test_stale_caps
- Previously would fail because the cap waiter
completed too soon, without noticing that the
reason it completed quickl... - 12:44 PM Revision 4daf2ddc (ceph): tasks: typo in mds_client_recovery
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:43 PM Revision 83fd1cf8 (ceph): bugfix: wrong socket address in log msg of Pipe.cc
- paddr was not yet set up for the socket address
Signed-off-by: Yongyue Sun abioy.sun@gmail.com - 12:39 PM rgw Bug #9479 (Resolved): ETag is not included in the XML response to put object copy operation
- User performs a put object copy operation, and the ETag is not included in the XML response.
- 12:37 PM rgw Bug #9478 (Resolved): Incorrect content type in response header
- User performs a put object copy operation, and seeing the content-type in the response header returned as "binary/oct...
- 12:32 PM CephFS Feature #9477: Handle kclient shutdown with dead network more gracefully
Ah, this *only* happens if I have some dirty state from userspace at the time. In this instance it's my Mount.open...- 11:59 AM CephFS Feature #9477 (Closed): Handle kclient shutdown with dead network more gracefully
- ...
- 10:44 AM Bug #9476 (Duplicate): "Segmentation fault (core dumped)" in upgrade:dumpling-giant-x:parallel-gi...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-14_15:05:01-upgrade:dumpling-giant-x:parallel-gian...
- 10:40 AM Linux kernel client Bug #4614 (Can't reproduce): Root cephfs does not mount at boot on Ubuntu 12.04
- 10:38 AM Documentation #9475 (Resolved): Broken links on downloads page
- The "View installation docs for..." links at the bottom of http://ceph.com/resources/downloads/ are broken, presumabl...
- 10:35 AM Bug #9470 (Resolved): daemon pid file is not being created when running service ceph
- This was fixed by commit:bccb0eb64891f65fd475e96b6386494044cae8c1, which will be in Giant.
- 05:01 AM Bug #9470 (Resolved): daemon pid file is not being created when running service ceph
- Hi,
We have been seeing some strange issues with the latest version(s) of ceph. I'm testing on 0.85 right now, an... - 10:24 AM Revision 92204287 (ceph): Merge pull request #2442 from dachary/wip-6754-jerasure-parameters
- erasure-code: fix BlaumRoth sanity check on w
Reviewed-by: Andreas Peters <andreas.joachim.peters@cern.ch> - 10:14 AM CephFS Bug #9423 (Resolved): failure in client_recovery task
- 10:14 AM CephFS Bug #9423: failure in client_recovery task
Fixed merged to giant....- 08:07 AM CephFS Bug #9423: failure in client_recovery task
- 09:50 AM CephFS Feature #9466 (In Progress): kclient: Extend CephFSTestCase tests to cover kclient
- 03:43 AM CephFS Feature #9466: kclient: Extend CephFSTestCase tests to cover kclient
- kclient instrumentation to enable implementing KernelClient::get_global_id (mapping local mount to the ID we see on t...
- 03:38 AM CephFS Feature #9466 (Resolved): kclient: Extend CephFSTestCase tests to cover kclient
Currently the mds_client_recovery and mds_client_limits tasks in ceph-qa-suite only run against the fuse client, be...- 09:42 AM devops Feature #9474 (Resolved): unify init-radosgw versions'
- there is a sysv version and a regular version. keep these in sync.
even better would be to unify with init-ceph .... - 09:39 AM Revision 8e625a00 (ceph): Merge pull request #2488 from cernceph/docfix
- doc: osd_backfill_scan_(min|max) are object counts
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 09:27 AM Revision 868b6b99 (ceph): doc: osd_backfill_scan_(min|max) are object counts
- osd_backfill_scan_min and osd_backfill_scan_max set the number of
items grabbed during a single backfill scan, not an... - 08:33 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- running "monthrash against master":http://pulpito.ceph.com/loic-2014-09-15_08:31:19-rados:monthrash-master-testing-ba...
- 08:33 AM Bug #9472 (Duplicate): osd crash in -upgrade:dumpling-dumpling-distro-basic-vps suite
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-14_17:00:01-upgrade:dumpling-dumpling-distro-basic...
- 08:21 AM devops Bug #9460: mira004, mira036. mira017 unresponsive
- For mira017 see : http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-14_17:00:01-upgrade:dumpling-dumpling-distro...
- 08:06 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
- https://github.com/ceph/ceph-qa-suite/pull/140
- 08:04 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
- 07:16 AM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- I've now installed an ubuntu 14.04 but could still not make it fail.
Even valgrind has a perfectly clean output.
I ... - 06:50 AM rgw Bug #8766: multipart minimum size error should be EntityTooSmall
- Starting to look into this...
- 06:42 AM Bug #9408 (In Progress): erasure-code: misalignment
- Now I see it, thanks for your patience.
- 06:12 AM Bug #9408: erasure-code: misalignment
- Hi Loic, I think Janne Grunau is right. For memory align, it depend on the bufferlist::c_str.
Using this patch:
... - 04:13 AM Bug #9408: erasure-code: misalignment
- With the following applied on dcc608d5d3f701315eaf0edee6f0a4796a4d97e1...
- 03:20 AM Bug #9408: erasure-code: misalignment
- jianpeng ma wrote:
> Can you tell met the result for this situation? I run with your command but it looks good.
I... - 04:53 AM Revision cdb7675a (ceph): rbd: ObjectCacher reads can hang when reading sparse files
- The pending read list was not properly flushed when empty objects
were read from a space file.
Signed-off-by: Jason ... - 04:52 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- Please help to review: https://github.com/ceph/ceph/pull/2489
- 04:33 AM rgw Bug #9469 (Rejected): RadosGW performance degrades with high concurrency workload.
- I am running COSbench as a performance benchmarking tool on a CEPH cluster(Swift API). Setup details are as follows:-...
- 04:29 AM Bug #9468 (Won't Fix): Unable to delete crush rule with blank space
- I am not sure how crush rule with blank space in beginning got created. But, I am not able to delete it.
ems@rack... - 04:15 AM Bug #9467 (Won't Fix): Delete default erasure coded profile getting succeeded
- Deleting default erasure coded profile is getting succeeded.
Also, re-creating erasure coded profile "default" with ... - 03:38 AM Revision ddd52e87 (ceph): init-radosgw.sysv: Support systemd for starting the gateway
- When using RHEL7 the radosgw daemon needs to start under systemd.
Check for systemd running on PID 1. If it is then ... - 03:24 AM Fix #6754 (Resolved): erasure-code: jerasure plugin does not check parameters properly
09/14/2014
- 05:43 PM Revision 973021cb (ceph): Merge pull request #137 from ceph/wip-9461
- Removed cuttlefish part foxes #9461
- 05:16 PM Revision 991a83c2 (ceph): Removed cuttlefish part foxes #9461
- Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
- 03:46 PM Revision d888753c (ceph): Merge pull request #2484 from sjahl/master
- doc: Added bucket management commands to ops/crush-map
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 03:31 PM devops Bug #9460: mira004, mira036. mira017 unresponsive
- For mira036 see - http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-13_17:00:04-upgrade:dumpling-dumpling-distro...
- 03:00 PM Linux kernel client Bug #4614: Root cephfs does not mount at boot on Ubuntu 12.04
- Howdy. I do not know. :) It has been over a year and I moved on long ago.
- 02:41 PM Revision d32b4286 (ceph): doc: Added bucket management commands to ops/crush-map
- Describes the CLI for adding and removing buckets, in addition to the
'moving' instructions which were already presen... - 01:22 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Done more testing. The hang tasks happen regardless of the block size (tried 4K, 1M ,4M). Without rbd cache enabled t...
- 11:24 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Okay, just tried with block size 4M and after about 20 dd runs I got a hang task.
Also, I've noticed that the spee... - 11:06 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Ilya,
Not sure as I've not left it to run for a long time. There was no progress for about 30 minutes, so I've reb... - 09:11 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Do those dds eventually complete?
- 09:04 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Ilya,
I've tried the kernel that was patched, but I am still experiencing hang tasks. Although less often than bef... - 04:20 AM Revision 3f0ca466 (ceph): Merge remote-tracking branch 'gh/giant'
09/13/2014
- 11:01 PM Revision 4fb3e292 (ceph): documentation: isa plugin is sub section of erasure code profile
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 11:01 PM Revision d446a656 (ceph): documentation: jerasure plugin is sub section of erasure code profile
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 11:01 PM Revision 75f0fb26 (ceph): documentation: erasure code profile update
- The plugins are made a sub-section of the erasure code profile
section. The k and m parameters are removed from erasu... - 10:32 PM Bug #9462 (Fix Under Review): msgr deadlock: osd reply vs mark_down vs fault
- see wip-9462.
tempted to push a patched ceph-mds to burnupi21 and see how it fares there. - 10:24 PM Bug #9462: msgr deadlock: osd reply vs mark_down vs fault
- Here is where we added it:...
- 10:19 PM Bug #9462: msgr deadlock: osd reply vs mark_down vs fault
- thread 160 (not fast dispatch) holds Objecter rwlock, blocks waiting for fast-dispatch to complete
thread 145 (and m... - 10:12 PM Bug #9462 (Resolved): msgr deadlock: osd reply vs mark_down vs fault
- ...
- 10:04 PM Bug #9459: osd: blocked request
- not affected by marking osd down.
- 12:12 PM Bug #9459 (Can't reproduce): osd: blocked request
- ...
- 09:29 PM Linux kernel client Bug #3087 (Can't reproduce): Hung ceph_msg_kfree
- 09:29 PM Linux kernel client Bug #4614: Root cephfs does not mount at boot on Ubuntu 12.04
- Is this still an issue? We have not been able to replicate. my fstab line looks like:
10.214.133.104:6789,10.21... - 09:25 PM Linux kernel client Bug #8979 (Pending Backport): GPF kernel panics - auth?
- 07:46 PM Revision b285788c (ceph): Merge pull request #2481 from sjahl/master
- doc: fixes a formatting error on ops/crush-map
- 07:31 PM Revision b8a1ec08 (ceph): doc: fixes a formatting error on ops/crush-map
- Signed-off-by: Stephen Jahl <stephenjahl@gmail.com>
- 05:23 PM devops Bug #9460 (Rejected): mira004, mira036. mira017 unresponsive
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-13_17:00:04-upgrade:dumpling-dumpling-distro-basic-vps/482303/...
- 03:56 PM Revision 8d066732 (ceph): Merge pull request #2467 from majianpeng/fix3
- buffer: In rebuild_page_aligned for the last ptr is page aligned, no need call rebuild().
Reviewed-by: Loic Dachary ... - 03:32 PM Revision 04e40737 (ceph): Merge pull request #2478 from ceph/wip-9445
- global: fix hang when segv happens inside logging code
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 02:40 PM Bug #8741 (In Progress): osd: ec plugin leak
- "added uninitialize function in jerasure":https://bitbucket.org/jimplank/jerasure/pull-request/24/define-galois_unini...
- 09:52 AM Fix #9413 (Resolved): erasure-code: disable erasure-code plugin version check when running from s...
- https://github.com/ceph/ceph/pull/2450
- 09:50 AM Bug #9434 (Need More Info): rbd rm hangs
- Version 0.71 was a development version. Are you observing the same version on a stable release such as 0.72 ? It woul...
- 09:32 AM Linux kernel client Bug #9458 (Resolved): client wrongly fenced
- ...
- 08:33 AM Bug #9445 (Resolved): deadlock in log code
- 04:07 AM Feature #9420 (Fix Under Review): erasure-code: tools and archive to check for non regression of ...
- 12:42 AM Revision 499a73b3 (ceph): Merge pull request #2477 from ceph/wip-client-msg-leak
- client: fix a message leak
- 12:39 AM Revision c3c6468c (ceph): mds: update segment references during journal rewrite
- ... to avoid leaving log events that reference log
segments by offsets which no longer exist.
Signed-off-by: John Sp... - 12:35 AM Revision e06f4251 (ceph): Merge pull request #2469 from ceph/wip-9427-rewrite
- mds: update segment references during journal rewrite
Reviewed-by: Greg Farnum <greg@inktank.com> - 12:31 AM Revision 6070383b (ceph): Merge pull request #2479 from ceph/wip-9444
- mds: fix root and mdsdir inodes' rsubdirs
Reviewed-by: Sage Weil <sage@redhat.com> - 12:27 AM Revision 7b25512d (ceph): mds: fix root and mdsdir inodes' rsubdirs
- inode rstat accounts inode itself.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
(cherry picked from commit da17... - 12:24 AM Revision 668cd135 (ceph): FileStore: report l_os_j_lat as commit latency
- l_os_commit_lat is actually the commit cycle latency.
Fixes: #9269
Backport: firefly
Signed-off-by: Samuel Just <sam... - 12:18 AM Revision a8c943a0 (ceph): log: add simple test to verify an internal SEGV doesn't hang
- Test that the segv injection works.
Test that a segv while logging something doesn't hang when the signal
handlers a...
09/12/2014
- 11:33 PM Revision 67a7fd27 (ceph): rgw: set debug rgw=20
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 11:14 PM Revision 2313ce1d (ceph): client: fix a message leak
- Signed-off-by: John Spray <john.spray@redhat.com>
- 10:32 PM Revision 7a0ef8e2 (ceph): First draft of firefly-giant-x suite
- Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
- 10:25 PM Revision e3fe18aa (ceph): global/signal_handler: do not log if SEGV originated inside log code
- Signed-off-by: Sage Weil <sage@redhat.com>
- 10:24 PM Revision 558463e8 (ceph): log: add Log::is_inside_log_lock()
- Signed-off-by: Sage Weil <sage@redhat.com>
- 10:05 PM Revision 386f2d7c (ceph): mds: update segment references during journal rewrite
- ... to avoid leaving log events that reference log
segments by offsets which no longer exist.
Signed-off-by: John Sp... - 09:07 PM Revision d41c3e85 (ceph): rgw: push hash calculater deeper
- This might have been the culprit for #9307. Before we were calculating
the hash after the call to processor->handle_d... - 08:40 PM Revision 28d4460a (ceph): DBObjectMap: restructure map_header_in_use locking
- Tieing map_header_in_use to a _Header is a mistake since ownership of
the map_header_in_use can move (clone). Instea... - 07:26 PM Revision 0a1abac6 (ceph): osd_types: fix pg_log_t tail dump
- Fixes: #9453
Signed-off-by: Samuel Just <sam.just@inktank.com> - 05:51 PM Revision 76869666 (ceph): Objecter::_recalc_linger_op: resend for any acting set change
- Fixes: #9220
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 1349383ac... - 05:50 PM Bug #9219: lost_unfound test got ENOENT: i don't have pgid 1.e
- The OSDs are getting the new OSDMap *very* late compared to the client
mon.c{2}: ... - 05:40 PM CephFS Bug #9427 (Resolved): osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_writte...
- Merged to master in commit:e06f4251ac36503d33f203567ada1b096119ab80.
Immediately cherry-picked to giant in commit:c3... - 11:35 AM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- For posterity, the manual test procedure for the changes to rewrite that fix this issue:...
- 06:57 AM CephFS Bug #9427 (Fix Under Review): osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= las...
- https://github.com/ceph/ceph/pull/2469
- 05:31 PM CephFS Bug #9444 (Resolved): "unmatched rstat" exception after firefly->master upgrade
- aha, yep! thanks
- 05:29 PM CephFS Bug #9444 (Fix Under Review): "unmatched rstat" exception after firefly->master upgrade
- git cherry-pick da17394941386dab88ddbfed4af2c8cb6b5eb72f
https://github.com/ceph/ceph/pull/2479
- 06:45 AM CephFS Bug #9444 (Resolved): "unmatched rstat" exception after firefly->master upgrade
Create filesystem with firefly, then restart system with master binaries plus wip-9427-rewrite....- 05:26 PM Bug #9456 (Rejected): rocksdb testing with powercycling fails on trusty
- teuthology needs to pass --type keyvaluestore-dev to ceph_objectstore_tool when using a key value store backend.
- 05:09 PM Bug #9456 (Rejected): rocksdb testing with powercycling fails on trusty
- when ceph_osd_objectstore is set to rocksdb with powercycling on trusty, it fails with the following message,
<pre... - 05:24 PM Bug #9269 (Resolved): ceph osd perf outputs the commit cycle latency for fs_commit_latency
- 05:19 PM Bug #9445 (Fix Under Review): deadlock in log code
- https://github.com/ceph/ceph/pull/2478
- 03:21 PM Bug #9445 (In Progress): deadlock in log code
- 08:21 AM Bug #9445 (Resolved): deadlock in log code
- ...
- 04:43 PM Bug #9455 (Resolved): mon: audit log read events should be debug level
- too much noise otherwise
while you're there, maybe we can change
2014-09-12 16:42:53.533788 mon.0 [INF] from='... - 04:30 PM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- Two steps:
1) During GetInfo, for actingbackfill peers, build up a feature set which is the intersection of the feat... - 04:15 PM Revision 4d3579d8 (ceph): doc: Added clarifications and added additional preflight steps for Cent...
- Fixes: #8995
Signed-off-by: John Wilkins <jowilki@redhat.com> - 04:14 PM Revision 1c8485e4 (ceph): doc: Deleted redundant text string.
- Signed-off-by: John Wilkins <jowilki@redhat.com>
- 03:47 PM Revision 9f4c6872 (ceph): Merge pull request #2471 from ceph/wip-9446
- mon: fix MDS health detail output
Reviewed-by: Sage Weil <sage@redhat.com> - 03:46 PM Revision ce7b2ecc (ceph): erasure-code: fix erasure_code_benchmark goop (decode)
- Using a stringstream that is only displayed on error when calling the
erasure code factory, instead of cerr. The user... - 03:43 PM Revision 9ba4e78f (ceph): mon: fix MDS health detail output
- I fat fingered a couple of things here.
Signed-off-by: John Spray <john.spray@redhat.com> - 03:00 PM devops Bug #9376 (Resolved): ceph-release package has '.centos' in its name, breaks installers
- Aparrantly centos changed this behavior with 7.0. Maybe they did it to better diferentiate between packages built on ...
- 02:18 PM Revision 2b45bfa2 (ceph): buffer: Add a test for bufferlist::rebuild_page_aligned
- Make the last prt of bufferlist which is page-aligned don't change anything
after rebuild_page_aligned.
Signed-off-b... - 02:14 PM Revision c268400b (ceph): buffer: In rebuild_page_aligned for the last ptr is page aligned, no ne...
- This only happen for the last ptr. Because rebuild() don't change the len
of ptr, so if last ptr isn't page-size alig... - 02:00 PM Revision 52834603 (ceph): Merge pull request #2468 from dachary/wip-always-create-pidfile
- daemons: write pid file even when told not to daemonize
Reviewed-by: Loic Dachary <loic-201408@dachary.org>
Reviewed... - 01:57 PM CephFS Bug #9280 (Resolved): valgrind failures in ceph-fuse
- commit:46bbe30e6895311e4ce5f9cf2dea3438db99188e
- 01:53 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- Yes, that's what I'm hoping as well. That's what _check_remove_pool() is; we'd need to add an equivalent for tiering....
- 01:48 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- I lean toward setting data pool to the base pool too. I worry about having to stand up so many guard rails, though. ...
- 01:43 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- The user pointed out that right now we prevent assigning EC pools to CephFS. I believe this is the result of a user w...
- 01:24 PM CephFS Bug #9423: failure in client_recovery task
- Can we pull out the fix so we can merge it and have it run against giant going forward?
- 01:18 PM Linux kernel client Bug #9426: kcephfs: soft lockup in handle mds map
- Added my Reviewed-by.
- 01:18 PM Linux kernel client Bug #9426: kcephfs: soft lockup in handle mds map
- 12:25 PM devops Bug #9265 (Resolved): ceph-deploy: don't force users to install `wget` if `wget` is available
- merged commit 66e0996 into ceph:master
- 10:46 AM devops Bug #9265 (Fix Under Review): ceph-deploy: don't force users to install `wget` if `wget` is avail...
- PR opened https://github.com/ceph/ceph-deploy/pull/239
- 09:47 AM devops Bug #9265 (In Progress): ceph-deploy: don't force users to install `wget` if `wget` is available
- 12:23 PM Bug #9453: ceph_objectstore_tool incorrect log tail output for --op log
- 12:23 PM Bug #9453 (Resolved): ceph_objectstore_tool incorrect log tail output for --op log
- { "head": "34'55",
"tail": "34'55",
"log": [
{ "op": "modify ",
"object": "c8be91cb\/rb.0.... - 11:44 AM Bug #9109 (In Progress): ceph CLI: Help is missing -k keyring option
- johnu will work on it
- 10:59 AM Bug #9452: All tests failed in upgrade:dumpling-giant-x:parallel-master-distro-basic-multi run
- I think it's the same problem in suite:upgrade:dumpling-firefly-x
http://qa-proxy.ceph.com/teuthology/teuthology-2... - 10:58 AM Bug #9452 (Resolved): All tests failed in upgrade:dumpling-giant-x:parallel-master-distro-basic-m...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-11_15:05:02-upgrade:dumpling-giant-x:parallel-mast...
- 10:52 AM Bug #9220 (Resolved): objecter doesn't reconnect watch on interval change w/ same primary
- 10:43 AM Bug #9446 (Resolved): mon deadlock
- commit:9ba4e78f00d6feaf97fcac6ec82950d981f5f80d
- 08:31 AM Bug #9446: mon deadlock
- another host:...
- 08:30 AM Bug #9446 (Resolved): mon deadlock
- ...
- 10:05 AM Revision bccb0eb6 (ceph): daemons: write pid file even when told not to daemonize
- systemd wants to run daemons in foreground, but daemons wouldn't write
out the pid file with -f. Fixed.
Signed-off-... - 09:39 AM Feature #9449 (Resolved): mon: make ceph -s break more things onto multiple lines (health blurbs,...
- 09:30 AM Revision 62ee5fc1 (ceph): Merge pull request #2464 from dachary/wip-9429-bench
- erasure-code: fix erasure_code_benchmark goop
Reviewed-by: Janne Grunau <j@jannau.net> - 08:30 AM Feature #9420 (In Progress): erasure-code: tools and archive to check for non regression of encoding
- 08:19 AM Documentation #8995 (In Progress): Preflight Checklist Clarifications
- 05:26 AM Linux kernel client Bug #9432 (In Progress): kcephfs: null pointer deref in posix_acl_create
- it seems that ACL code can't handle traceless reply.
- 05:01 AM Bug #9408: erasure-code: misalignment
- Can you tell met the result for this situation? I run with your command but it looks good.
- 02:10 AM Bug #9408: erasure-code: misalignment
- ./src/ceph_erasure_code_benchmark --plugin jerasure --workload encode --iterations 1048576 --size 4096 --erasures 0 -...
- 03:41 AM Revision e01e0c27 (ceph): Merge pull request #2416 from xiaoxichen/make_crush_private
- Change CrushWrapper::crush to private
Reviewed-by: Sage Weil <sage@redhat.com> - 02:56 AM Revision 69d3594f (ceph): Merge pull request #2450 from dachary/wip-9413-erasure-code-version-check
- erasure-code: mon, osd etc. depend on the plugins
- 02:31 AM Bug #9429 (Resolved): erasure-code: bench.sh fails because of load: jerasure
- 02:27 AM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- This is because giant is at tag v0.85 which does not include the fix. The fix is in the giant branch though so it wil...
- 12:19 AM Revision aa5234ed (ceph): doc: Add file system osd config settings
- Add documentation for osd mkfs and osd mount options.
Signed-off-by: Kevin Dalley <kevin@kelphead.org>
09/11/2014
- 11:48 PM Revision 8d23e8db (ceph): rbd-replay: Add --dump-perf-counters option
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 11:48 PM Revision f54d5c7b (ceph): rbd-replay: Add code documentation
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 11:48 PM Revision 398ebc0b (ceph): librbd: Add missing tracepoints
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 11:48 PM Revision 675a6a67 (ceph): rbd-replay: Add rbd-replay-many
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 11:47 PM Revision 3911354d (ceph): rbd-replay: Add --anonymize flag to rbd-replay-prep
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 10:46 PM Revision 0e0a8d4e (ceph): msg: prefetch data when doing recv()
- Since that when we're reading the header we do lot's of small IOs, we
can really benefit from doing bigger reads.
Set... - 09:45 PM Bug #9443 (Rejected): btrfs pwrite returns EEXIST on journal FileJournal::write_bl
- ...
- 09:14 PM Revision 16bd4577 (ceph): ReplicatedPG: cancel cb on blacklisted watcher
- Fixes: #8315
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com> - 08:24 PM Revision b7bdb93f (ceph): erasure-code: fix erasure_code_benchmark goop
- Using a stringstream that is only displayed on error when calling the
erasure code factory, instead of cerr. The user... - 07:56 PM Revision 60d65daa (ceph): Merge pull request #2409 from apeters1971/wip-ec-isa-table-cache-refac-...
- EC-ISA: add intelligent table cache
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 07:37 PM Bug #9438 (Fix Under Review): librados API generated doc broken
- 04:38 PM Bug #9438 (Resolved): librados API generated doc broken
The generated documentation which ends up in build-doc/output/html/rados/api/librados/index.html has some broken ou...- 06:56 PM Bug #6115: doc: asphyxiate does not support class
- It would be nice if we could create documentation based on the librados namespace. Doxygen comments would be in 2 fil...
- 06:22 PM Revision 1de43e00 (ceph): OpTracker: Removed _dump_op_descriptor_unlocked call for log level < 5
- _dump_op_descriptor_unlocked call within the function _mark_event()
was invoked unnecessarily every time for all log ... - 06:22 PM Revision 3945ead6 (ceph): OpTracker: The TrackedOp::events list was not protected while dumping ops
- OpTracker::Mutex::lock is there to protect the event list and it is
only used in the mark_event while pushing the eve... - 06:22 PM Revision 11082f7a (ceph): OpTracker: Race condition removed while dumping ops through admin socket
- OSD was crashing due to a race condition while IO was going on and
user wants to dump in flight ops. This was happeni... - 06:18 PM Revision 3fac7901 (ceph): OpTracker: Sharding logic is implemented to improve performance
- Enabling OpTracking is degrading performance significantly. The
main reason for this is the contention in the ops_in_... - 06:16 PM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- Note for re-testing:
Same issues on suite:upgrade:dumpling-giant-x
For example http://qa-proxy.ceph.com/teuthol... - 06:07 PM Revision d43850ee (ceph): Added RHEL7 to matrix.
- Signed-off-by: Sandon Van Ness <sandon@inktank.com>
- 05:51 PM rbd Feature #6228: image name metavariable
- ...which is hard because librbd layers on top of an existing librados ioctx, which has already long been initialized....
- 05:46 PM rbd Feature #6228: image name metavariable
- It's per-cct. We can probably add some methods like
cct->set_meta("image", "foo");
that would populate a map<... - 05:21 PM rbd Feature #6228: image name metavariable
- The code would be in src/common.cc. I don't think it would be as simple as just adding another variable, because all...
- 05:51 PM Bug #9441 (Closed): defgroup
- 05:40 PM Bug #9441 (Closed): defgroup
- 05:50 PM Bug #9442 (Resolved): Doxygen comments in @defgroup are not part of generated doc
Large sections of comments in the librados.h specified to describe a grouping with @defgroup are not part of the ge...- 05:50 PM Revision 2fbe82df (ceph): doc: Fixed syntax error.
- Signed-off-by: John Wilkins <jowilki@redhat.com>
- 05:50 PM Revision 7dbf750f (ceph): doc: Updated authentication notes. Fixed syntax error.
- Signed-off-by: John Wilkins <jowilki@redhat.com>
- 05:33 PM Revision 9e6d05af (ceph): Merge pull request #2459 from ceph/wip-7934
- test: Fix ceph_test_rados_watch_notify to delete the pools it creates
Reviewed-by: Loic Dachary <loic-201408@dachary... - 05:23 PM Bug #9408: erasure-code: misalignment
- Loic, although i'm not sure erasure whether need align memory.
int ErasureCodeJerasure::encode_chunks(const set<i... - 05:21 PM Feature #9440 (Resolved): mon: log all changes to health in the central log
- It would be awesome if the central log recorded every time the HEALTH status of the cluster changed.
- 04:40 PM Bug #9439 (Resolved): pg_op_must_wait() not checking FILTER variants
The filter variants CEPH_OSD_OP_PGLS_FILTER and CEPH_OSD_OP_PGNLS_FILTER should be checked too.
CEPH_OSD_OP_PGNL...- 04:19 PM Revision 25d25370 (ceph): erasure-code: CEPH_FEATURE_ERASURE_CODE_PLUGINS_V2 integration tests
- http://tracker.ceph.com/issues/9343 Refs: #9343
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 03:57 PM Revision d6f02d60 (ceph): erasure-code: assert that firefly-x/stress-split cannot load lrc
- Because it would crash existing plugins.
Signed-off-by: Loic Dachary <loic-201408@dachary.org> - 03:45 PM Revision dcc608d5 (ceph): Merge pull request #2463 from ceph/wip-mds-beacon
- mds: a couple fixes for the beacons
Reviewed-by: John Spray <john.spray@redhat.com> - 03:37 PM Bug #9274: "AssertionError: failed to recover before timeout expired" in upgrade:dumpling-x:stres...
- 03:35 PM Revision 6fb5769a (ceph): mds/Beacon: do not reconnect to mon in quick succession
- Wait at least one beacon interval between mon session resets.
Fixes: #9428
Signed-off-by: Sage Weil <sage@redhat.com> - 03:35 PM Revision 7281638d (ceph): mds: sleep in progress thread if laggy and waiting_for_nolaggy waiters
- If we have nolaggy waiters but are laggy we want to sleep. Otherwise,
we will just spin and spam the log ...
Signed... - 02:53 PM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- I can't reproduce it with debug and original sha1 is already gone.
- 02:42 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- my vote is to make the pgls continuation hold onto existing budget (and not take new budget). is that feasible?
- 02:34 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- Okay, this is because our object listing code is incorrect (in both Firefly and Giant-to-be). A pgls response has a c...
- 01:57 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- Well, the dispatcher doesn't normally take budget directly, but it could be doing something else farther down the cal...
- 06:08 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- I finally got some logs which revealed that the osd_op_reply message had been received by Pipe, but it stayed in Disp...
- 01:58 PM Bug #9272 (Duplicate): Test failed on wait_until_healthy in upgrade:dumpling-firefly-x-master-dis...
- 01:48 PM Bug #8315: osd: watch callback vs callback funky
- 01:32 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- This conversation is getting split across several mediums, but this shouldn't prevent specifying the use of a base po...
- 01:13 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- My vote is NAK on this. THis is exactly what I want to do on my cluster and I this is the only way EC can be used fo...
- 01:07 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- Yeah, that's the simple solution. I was also wondering though if we wanted to do something more sophisticated trying ...
- 12:48 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- I mean something like this (although I'm not positive I got all the requirements right):...
- 12:39 PM CephFS Fix #9435: prevent use of cache pools as metadata or data pools
- would checking the nature of pools during 'fs new' on the monitor and failing if any of the specified pools (data or ...
- 11:30 AM CephFS Fix #9435 (Resolved): prevent use of cache pools as metadata or data pools
- From the mailing list...
- 01:26 PM Bug #9429 (Fix Under Review): erasure-code: bench.sh fails because of load: jerasure
- "proposed fix":https://github.com/ceph/ceph/pull/2464
- 02:34 AM Bug #9429 (Resolved): erasure-code: bench.sh fails because of load: jerasure
- This message should not be sent |"directly to cerr":https://github.com/ceph/ceph/blob/master/src/test/erasure-code/ce...
- 01:16 PM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- The problem here appears to be that the user upgraded the clients before the osds were fully upgraded. librbd sends ...
- 12:56 PM Bug #9367 (Resolved): erasure-code: use LRU cache for decoding in ISA
- 12:55 PM Feature #9161: Cache warmup and ejection
- awesome, thanks Greg!
- 11:50 AM CephFS Feature #9437 (Resolved): make 'ceph tell mds.* ...' work, deprecate 'ceph mds tell * ...'
- 11:34 AM rgw Feature #9436 (New): qa: cosbench in rgw test suite
- 11:28 AM Bug #9434 (Can't reproduce): rbd rm hangs
- I'm using ceph 0.71(maybe it's a little old)
I did some performance measurements on ceph these days, but I met some ... - 11:22 AM Bug #7934 (Resolved): ceph_test_rados_watch_notify doesn't clean-up all pools it creates
- 11:02 AM rgw Feature #9433 (New): Support per-user container namespace
- Per:
http://docs.openstack.org/api/openstack-object-storage/1.0/content/object_storage_api_overview.html
"Your se... - 10:25 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- running "monthrash suite":http://pulpito.ceph.com/loic-2014-09-11_19:23:55-rados:monthrash-wip-9343-erasure-code-feat...
- 09:38 AM Bug #9327 (Rejected): Usability Issue: Ceph-deploy does not print all the commands which it is ex...
- What you are seeing is actually the output of the remote host (stormeap-0 in your case) that is caused by ceph-disk i...
- 09:33 AM devops Tasks #8366 (Resolved): Update ceph.com/docs to default to the latest major release (0.80)
- /docs/ now redirects to the latest (0.80.5 at the moment)
And will get updated as soon as there is a new release - 09:32 AM devops Feature #9103 (Resolved): create a (generic) webservice to handle Sphinx documentation versions
- Deployed to http://ayni.ceph.com
If a JSON add-on is installed in the browser, here are the projects for ceph: htt... - 09:26 AM CephFS Bug #9428 (Resolved): mds: tight mon reconnect loop
- 09:06 AM Linux kernel client Bug #9432 (Resolved): kcephfs: null pointer deref in posix_acl_create
- ...
- 09:06 AM Revision c9306241 (ceph): Merge pull request #2460 from ceph/wip-client-ll-ref
- client: include ll_ref when printing inode
- 09:03 AM Revision 4ad5db00 (ceph): client: include ll_ref when printing inode
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 08:35 AM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- It doesn't need to be an absolute offset that gets fed to the standby-replay MDS, as long as it can use the informati...
- 06:55 AM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- So rewriting the truncate_finish part isn't too hard if we want to do that:
https://github.com/ceph/ceph/commit/4ae6... - 03:40 AM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
The history:...- 07:40 AM Feature #9431: mon: audit read-only commands on a lower level than default
- Maybe the readonly-ness could be inferred from the "r" or "rw" in the COMMAND() definition from MonCommands.h, so the...
- 07:38 AM Feature #9431 (New): mon: audit read-only commands on a lower level than default
- ...
- 07:16 AM Documentation #6142 (Resolved): Ceph needs mor than 32k pids
- merged commit 73fdc7b into ceph:master
- 07:10 AM CephFS Bug #9341: MDS: very slow rejoin
- I re-built and re-deployed ceph with fuse patch; re-configured all kernel clients to use fuse client; re-mounted Ceph...
- 07:08 AM Linux kernel client Bug #9426: kcephfs: soft lockup in handle mds map
- maybe this can be fixed by: "ceph: protect kick_requests() with mdsc->mutex" and "ceph: make sure request isn't in an...
- 06:56 AM CephFS Feature #9375 (Fix Under Review): Send single 'many clients' health warning instead of N warnings...
- 06:56 AM CephFS Feature #9189 (Fix Under Review): Expose client identifying metadata to MDS, e.g. hostname
- 04:14 AM Documentation #9430 (Closed): dev documentation about incompat features
- turn the following dialog into a documentation for developers...
- 03:44 AM Revision 2235aa99 (ceph): Merge pull request #2444 from wonzhq/read-recency
- osd: set min_read_recency_for_promote to default 1 when doing upgrade
Reviewed-by: Sage Weil <sage@redhat.com> - 03:37 AM Revision 59f1a49b (ceph): Merge pull request #2449 from majianpeng/fix3
- fix two bugs about perfcounter
Reviewed-by: Sage Weil <sage@redhat.com> - 02:19 AM Revision cf70b906 (ceph): test: Fix ceph_test_rados_watch_notify to delete the pools it creates
- Fixes: #7934
Signed-off-by: David Zafman <dzafman@redhat.com> - 12:32 AM Revision 2e4e98bd (ceph): ReplicatedPG: Make perfcounter record the read-size for
- async-read.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com> - 12:09 AM Revision cf34e00d (ceph): ReplicatedPG: record correctly subop for perfcounter.
- In log_subop_stats, it omit to record the counter of subop.
Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
09/10/2014
- 10:45 PM Revision 45a9d10e (ceph): Merge pull request #132 from ceph/revert-131-wip-9357-wusui
- Revert "Use same upgrade-test as firefly."
- 10:42 PM Revision fd12e0d2 (ceph): Revert "Use same upgrade-test as firefly."
- 10:09 PM CephFS Bug #9428 (Resolved): mds: tight mon reconnect loop
- ...
- 10:08 PM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- wip-mds has hacky workaround
- 09:38 PM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- ESubtreeMap has an expire_pos field, and we set it in ESubtreeMap::replay() if it is > the current expire pos. I thi...
- 09:30 PM CephFS Bug #9427: osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_written.expire_pos)
- ...
- 08:52 PM CephFS Bug #9427 (Resolved): osdc/Journaler.cc: 405: FAILED assert(last_written.write_pos >= last_writte...
- ...
- 10:03 PM Revision 07653938 (ceph): Merge pull request #131 from ceph/wip-9357-wusui
- Use same upgrade-test as firefly.
- 09:49 PM Revision 5be4aaf9 (ceph): Use same upgrade-test as firefly.
- The changes for firefly upgrade tests apply to giant as well.
- 07:29 PM Revision b99c85da (ceph): Merge pull request #130 from ceph/wip-9220
- tasks: add watch_notify_same_primary
Reviewed-by: Sage Weil <sage@redhat.com> - 07:21 PM Linux kernel client Bug #9426: kcephfs: soft lockup in handle mds map
- looks like it was waiting for i_ceph_lock, did the soft lockup recover?
- 06:03 PM Linux kernel client Bug #9426 (Resolved): kcephfs: soft lockup in handle mds map
- ...
- 07:19 PM Bug #7934 (Fix Under Review): ceph_test_rados_watch_notify doesn't clean-up all pools it creates
- 07:15 PM Revision e50c426b (ceph): Merge pull request #2454 from athanatos/wip-9269
- FileStore: report l_os_j_lat as commit latency
Reviewed-by: Sage Weil <sage@redhat.com> - 07:09 PM Revision 6846811e (ceph): Merge pull request #2453 from athanatos/wip-9220
- Objecter::_recalc_linger_op: resend for any acting set change
Reviewed-by: Sage Weil <sage@redhat.com> - 07:09 PM Revision dc5e98d3 (ceph): Merge pull request #2443 from ceph/wip-9241
- osdc/Objecter: drop bad session nref assert
Reviewed-by: Samuel Just <sam.just@inktank.com> - 07:02 PM Revision d0feeaa1 (ceph): Merge pull request #2293 from ceph/wip-hitset-bytes
- osd: improve agent calculation by factoring out hit_set bytes used properly
Reviewed-by: Samuel Just <sam.just@inkta... - 07:02 PM CephFS Bug #9341: MDS: very slow rejoin
- that patch is for kernel client. here is the patch for ceph-fuse
- 05:23 PM CephFS Bug #9341: MDS: very slow rejoin
- Zheng Yan wrote:
> are you using kernel client? If you are, please try the attached patch. I hope it will improve re... - 06:19 AM CephFS Bug #9341: MDS: very slow rejoin
- are you using kernel client? If you are, please try the attached patch. I hope it will improve rejoin speed.
- 06:56 PM rgw Bug #8111 (Resolved): /etc/init.d/ceph-radosgw for RHEL needs QA
- I believe this issue cropped up because of using other than short host name. Ran through this with Tamil, and the not...
- 06:49 PM Revision 117de007 (ceph): erasure-code: mon, osd etc. depend on the plugins
- Since the erasure code plugin version check has been introduced,
whenever a library/binary that can load plugin needs... - 06:31 PM Revision c6678eb5 (ceph): Merge pull request #2451 from ceph/wip-osdc-leak
- osdc/Objecter: fix leak of MStatfsReply
Reviewed-by: John Spray <john.spray@redhat.com> - 06:11 PM Feature #9031 (Fix Under Review): List RADOS namespaces and list all objects in all namespaces
- 06:11 PM Feature #9262 (Fix Under Review): Additional namespace issues
- 05:13 PM Bug #9326: osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- 04:59 PM Revision bead44cd (ceph): Merge pull request #2447 from reclosedev/s3_colon_in_access_key
- [rgw][s3] Allow colon ':' in access key
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 04:45 PM Revision 5a05e6b8 (ceph): [rgw][s3] Allow colon ':' in access key
- When access key contains ':', e.g. `some_info:for_user',
authorization header looks like:
"AWS some_info:for_user:re... - 03:56 PM CephFS Bug #9178: samba: ENOTEMPTY on "rm -rf"
- the fix https://github.com/ceph/ceph/pull/2431 hasn't been merged yet
- 01:54 PM CephFS Bug #9178: samba: ENOTEMPTY on "rm -rf"
- /a/teuthology-2014-09-08_23:14:02-samba-master-testing-basic-multi/474551/
- 03:15 PM CephFS Bug #9423: failure in client_recovery task
- Had seen this previously at http://pulpito.ceph.com/teuthology-2014-09-05_23:04:02-fs-master-testing-basic-multi/4701...
- 01:27 PM CephFS Bug #9423: failure in client_recovery task
- http://pulpito.ceph.com/teuthology-2014-09-08_23:04:01-fs-master-testing-basic-multi/474441/
- 01:27 PM CephFS Bug #9423 (Resolved): failure in client_recovery task
- ...
- 03:07 PM Revision f05c977b (ceph): mon: add 'osd pool ls [detail]' command
- This is much friendlier than
ceph osd dump | grep ^pool
Signed-off-by: Sage Weil <sage@redhat.com> - 03:00 PM Revision d3e880af (ceph): test/cli-integration/rbd: fix trailing space
- Newer versions of json.tool remove the trailing ' ' after the comma. Add
it back in with sed so that the .t works on... - 02:45 PM Revision 53363986 (ceph): os/FileStore: fix mount/remount force_sync race
- Consider:
- mount
- sync_entry is doing some work
- umount
- set force_sync = true
- set done = true
- syn... - 02:12 PM devops Bug #9332: Placeholder for "SSH connection was lost" errors
- On today's run with new VM images, there are more SSH errors:
http://pulpito.front.sepia.ceph.com/teuthology-2014-... - 02:01 PM CephFS Bug #8427: ceph-fuse: Dumpling "cache still has 0+1 items, waiting (for caps to release?)" on shu...
- /a/teuthology-2014-09-09_19:06:01-fs-dumpling-testing-basic-multi/475752
I copied the server logs to it - 01:57 PM Revision 46bbe30e (ceph): osdc/Objecter: fix leak of MStatfsReply
- Signed-off-by: Sage Weil <sage@redhat.com>
- 01:56 PM Revision 136c5e11 (ceph): Merge pull request #2448 from ceph/wip-replay-locking
- mds: fix replay locking
Reviewed-by: John Spray <john.spray@redhat.com> - 01:54 PM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
- /teuthology-2014-09-08_19:06:01-fs-dumpling-testing-basic-multi/473897/
/teuthology-2014-09-08_19:06:01-fs-dumpling-... - 01:54 PM CephFS Bug #9280: valgrind failures in ceph-fuse
- /teuthology-2014-09-08_23:04:01-fs-master-testing-basic-multi/474458/
/teuthology-2014-09-08_23:04:01-fs-master-test... - 01:46 PM CephFS Bug #8576: teuthology: nfs tests failing on umount
- ...
- 12:28 PM Bug #9422: librados: client.admin authentication error (110) Connection timed out
- /a/samuelj-2014-09-09_19:17:59-rados-wip-sam-testing-testing-basic-multi/475898
- 12:28 PM Bug #9422 (Can't reproduce): librados: client.admin authentication error (110) Connection timed out
- 2014-09-10T01:12:01.088 INFO:tasks.ceph.mon.b:Restarting daemon
2014-09-10T01:12:01.088 INFO:teuthology.orchestra.ru... - 12:21 PM Bug #9241 (Resolved): osdc/Objecter.cc: 1320: FAILED assert(s->get_nref() == 1)
- 12:15 PM Bug #9220 (Pending Backport): objecter doesn't reconnect watch on interval change w/ same primary
- 12:05 PM Bug #9220 (Fix Under Review): objecter doesn't reconnect watch on interval change w/ same primary
- 12:15 PM Bug #9269 (Pending Backport): ceph osd perf outputs the commit cycle latency for fs_commit_latency
- 12:05 PM Bug #9269 (Fix Under Review): ceph osd perf outputs the commit cycle latency for fs_commit_latency
- 11:40 AM Documentation #6142: Ceph needs mor than 32k pids
- PR opened https://github.com/ceph/ceph-deploy/pull/238
- 10:35 AM Feature #9420 (Resolved): erasure-code: tools and archive to check for non regression of encoding
- to ensure the encoding of the data does not change over time. This must be made to check that a given plugin with a g...
- 10:28 AM Bug #9419: dumpling->firefly upgrade, sending setallochint?
- client rbd (firefly) --with setallochint--> primary (firefly) --with setallochint--> replica (dumpling) crash
- 10:14 AM Bug #9419 (Resolved): dumpling->firefly upgrade, sending setallochint?
- Crash on dumpling osds with bad op 39 when the first osd is upgraded to firefly, setallochint.
https://github.com/... - 09:39 AM rgw Feature #8945: rgw: support swift /info api
- This info just looks like some internal configurables. If it doesn't make sense as an api, I'm not sure we want to su...
- 06:21 AM rgw Feature #8945: rgw: support swift /info api
- After some investigation it seems there is no real standard because OpenStack Swift is not an API. Therefore, client...
- 09:25 AM Bug #9418 (Resolved): mon: drop internal-purpose messages from clients without proper caps
- 09:25 AM Revision 5b41dacf (ceph): EC-ISA: add intelligent table cache
- 09:25 AM Revision 1ff94ccf (ceph): EC-ISA: avoid usage of table cache lock outside the class implementatio...
- 09:05 AM Bug #9417 (Duplicate): "Segmentation fault" in upgrade:dumpling-giant-x-master-distro-basic-vps run
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-09_17:00:02-upgrade:dumpling-giant-x-master-distro...
- 08:48 AM Revision 83bd3430 (ceph): mds: fix replay locking
- When replaying EImportFinish/EFragment event, the replay thread may call
MDS::queue_waiters. MDS::queue_waiters() req... - 08:30 AM Bug #9416 (Duplicate): ods crash in upgrade:dumpling-dumpling-distro-basic-vps run
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-09_18:30:01-upgrade:dumpling-dumpling-distro-basic...
- 07:47 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- radosgw side log before hang.
- 07:18 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- Another hang we observed with debug ms = 1 at radosgw side, it is confirmed that the osd_op does not have its peer os...
- 07:17 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- While in the progress of testing with debug ms = 20, I have a couple of questions:
1> What is the risk of adding ti... - 06:54 AM Fix #9413 (Fix Under Review): erasure-code: disable erasure-code plugin version check when runnin...
- 05:45 AM Fix #9413: erasure-code: disable erasure-code plugin version check when running from sources
- it's a pain for the developer, deal with this quickly
- 05:40 AM Fix #9413 (Resolved): erasure-code: disable erasure-code plugin version check when running from s...
- because dependencies are taken care of by the makefiles and it's easy to create false positives
- 06:53 AM Revision 3448d852 (ceph): Cache tiering: do not allow flush/evict when there are requests pending
- Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
- 06:02 AM Bug #9128: Newly-restarted OSD may suicide itself after hitting suicide time out value because it...
- Guang Yang wrote:
> Wei's patch - https://github.com/ceph/ceph/pull/2371
looks good, just needs a signed-off-by line - 05:58 AM CephFS Feature #7316: improve mds state dumps (memory usage, completeness)
- NB as follow up to our new health checks (9282, 9284) we should ensure we add admin socket commands for dumping the s...
- 05:08 AM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.
- Still fails on master e11f061
- 03:58 AM Revision b40cce7a (ceph): osd: set min_read_recency_for_promote to default 1 when doing upgrade
- When upgrading from a build without the promotion on 2nd read feature,
should set min_read_recency_for_promote to the... - 02:58 AM rbd Bug #8329: qemu-img rpm provided breaks snapshooting functionality on centos
- As far as I know, there is still no support for RBD in qemu-kvm/qemu-img packages...
Also I don't know if cloudsta... - 01:39 AM devops Feature #9411: remove qemu symlink for librbd on rhel7.1 (and later)
- Some people complained about this when I synced up the fedora's spec file with the upstream one (especially the fact ...
- 01:08 AM Revision d734600f (ceph): Change CrushWrapper::crush to private
- Currently in CrushWrapper, the member "struct crush_map *crush" is a public member,
so people can break the encapsul... - 12:48 AM rgw Feature #9359 (Fix Under Review): rgw: Export user stats in get-user-info Adminops API
- Please help to review the pull request: https://github.com/ceph/ceph/pull/2446
- 12:45 AM rgw Feature #9359: rgw: Export user stats in get-user-info Adminops API
- Changed the sample response as to
Sample Response {
"user_id":"foo_user",
"display_name":"foo_user",
"email":"... - 12:28 AM Revision 5a784cd3 (ceph): osdc/Objecter: drop bad session nref assert
- This is a bad assert. Specifically, handle_osd_op_reply may still be
holding the session ref while it is calling the...
09/09/2014
- 11:40 PM Revision e11f0617 (ceph): Merge pull request #2433 from ceph/wip-rbd-force-write-back
- rbd should use write-back when caching is enabled
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 10:31 PM Revision f6582f89 (ceph): tasks: add watch_notify_same_primary
- Reproduces: #9220
Signed-off-by: Samuel Just <sam.just@inktank.com> - 09:45 PM Revision 3c6e8884 (ceph): osd/ClassHandler: fix build
- Broken by 70ce400a8b4e0f5a20e6ea9877784998cdbb9a2d.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:38 PM Bug #8589: scrub error with leveldb backend
- Haomai, I hope you could have a look at another scrub bug mis-detecting inconsistent PGs on caching pools: #8752. Tha...
- 01:33 PM Bug #8589 (Resolved): scrub error with leveldb backend
- 02:55 AM Bug #8589 (Fix Under Review): scrub error with leveldb backend
- 09:33 PM Revision f7498c47 (ceph): osdc/Objecter: revoke rx_buffer on op_cancel
- If we cancel a read, revoke the rx buffers to avoid a use-after-free and/or
other undefined badness by using user buf... - 09:30 PM Revision 3abf9545 (ceph): osdc/Objecter: revoke rx_buffer on op_cancel
- If we cancel a read, revoke the rx buffers to avoid a use-after-free and/or
other undefined badness by using user buf... - 09:24 PM Revision bc410844 (ceph): ceph_test_rados_api_io: add read timeout test
- Verify we don't receive data after a timeout.
Based on reproducer for #9362 written by
Matthias Kiefer <matthias.kie... - 09:24 PM Revision 19e76065 (ceph): ceph_test_rados_api_*: expose nspace
- Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 977d289055d69ab8a7baaf7ef68c013019225833) - 09:07 PM Revision d165238b (ceph): FileStore: report l_os_j_lat as commit latency
- l_os_commit_lat is actually the commit cycle latency.
Fixes: #9269
Backport: firefly
Signed-off-by: Samuel Just <sam... - 08:53 PM Revision 1bce730b (ceph): Merge pull request #2441 from ceph/wip-9365
- osd/ClassHandler: improve error logging
Reviewed-by: Samuel Just <sam.just@inktank.com> - 08:52 PM devops Feature #9411 (New): remove qemu symlink for librbd on rhel7.1 (and later)
- rhel 7.1's qemu will no longer need the goofy runtime linking or this symlink.
this should be done for 7.1 and lat... - 08:39 PM Bug #9340: ceph mon crashes when tried with rocksdb on precise
- i can not reproduce this error on my precise setup (upgrade gcc to 4.7.3), can you give more detail info?
- 08:38 PM Revision 70ce400a (ceph): osd/ClassHandler: improve error logging
- Fixes: #9365
Signed-off-by: Sage Weil <sage@redhat.com> - 08:36 PM Revision 1388f35b (ceph): Merge pull request #2412 from dachary/wip-9370-flush-logs
- tests: flush logs before grepping them
- 08:14 PM Revision 0c24bff6 (ceph): Merge pull request #2434 from dachary/wip-9381-erasure-code-rpm
- packaging: add to RPM packages isa and lrc
Reviewed-by: Sage Weil <sage@redhat.com> - 08:11 PM Revision 0d66922c (ceph): Merge pull request #2427 from ceph/wip-9362
- librados: do not write to user buffer after timeout
Reviewed-by: Greg Farnum <greg@inktank.com> - 08:10 PM Revision 4597401e (ceph): Merge pull request #2437 from athanatos/wip-9339
- ReplicatedPG: create max hitset size
Reviewed-by: Sage Weil <sage@redhat.com> - 08:01 PM Revision 977d2890 (ceph): ceph_test_rados_api_*: expose nspace
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:01 PM Revision f295c1fe (ceph): ceph_test_rados_api_io: add read timeout test
- Verify we don't receive data after a timeout.
Based on reproducer for #9362 written by
Matthias Kiefer <matthias.kie... - 08:01 PM Revision 2305b289 (ceph): osdc/Objecter: revoke rx_buffer on op_cancel
- If we cancel a read, revoke the rx buffers to avoid a use-after-free and/or
other undefined badness by using user buf... - 07:58 PM Revision 1349383a (ceph): Objecter::_recalc_linger_op: resend for any acting set change
- Fixes: #9220
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com> - 07:56 PM Revision 9e2d04f7 (ceph): erasure-code: fix BlaumRoth sanity check on w
- For the BlaumRoth technique, w+1 must be a prime. This sanity check is
present in the sample code from the jerasure l... - 07:43 PM Revision b477abbf (ceph): Merge pull request #2439 from athanatos/wip-9054
- Wip 9054
Reviewed-by: Sage Weil <sage@redhat.com> - 07:40 PM Revision 881c267f (ceph): Revert "PG: mark_log_for_rewrite on resurrection"
- Actually, we don't want to backport this one without the fix
for #9293.
This reverts commit 7ddf0a252bb887553b29fd93... - 07:38 PM Revision f9171661 (ceph): erasure-code: refactor jerasure Liberation parameters checks
- Create a virtual function to check for each parameter for Liberation
and its derivations Liber8tion and BlaumRoth. Th... - 07:37 PM Revision 4843fd51 (ceph): ReplicatedPG:start_flush send a second delete
- Suppose we start with the following in the cache pool:
30:[29,21,20,15,10,4]:[22(21), 15(15,10), 4(4)]+head
The obj... - 07:36 PM Revision ead61950 (ceph): Merge pull request #2256 from ceph/wip-9005
- doc: add calamari to the api/index section.
Reviewed-by: John Wilkins <jowilkin@redhat.com> - 07:36 PM Revision 78d7499d (ceph): Revert "osd/ReplicatedPG: avoid dereferencing iterator at end()"
- This reverts commit ded1cf4a93d15ca3be312d296c3ab956109587e3.
- 07:36 PM Revision 937d2b0e (ceph): Revert "ReplicatedPG:start_flush send a second delete"
- This reverts commit a525bf3190a712fdceb018267a0aa08e90c44e62.
Conflicts:
src/osd/ReplicatedPG.cc - 07:26 PM Revision 3578b119 (ceph): ReplicatedPG: create max hitset size
- Otherwise, hit_set_create could create an unbounded size hitset
object.
Fixes: #9339
Backport: firefly
Signed-off-by... - 07:11 PM Revision 18176b7a (ceph): Merge pull request #2436 from athanatos/wip-9293
- PG::init: clear rollback info for backfill as well
- 07:10 PM Revision 8d9da05a (ceph): Merge pull request #2435 from athanatos/wip-9179
- PG: wait until we've build the missing set to discover_all_missing
Reviewed-by: Sage Weil <sage@redhat.com> - 06:35 PM Revision 0f564aa0 (ceph): Merge pull request #116 from dachary/wip-9298-rgw-ec-starvation
- rgw: increase the number of OSDs for erasure coded pools
- 06:35 PM Bug #9128: Newly-restarted OSD may suicide itself after hitting suicide time out value because it...
- Wei's patch - https://github.com/ceph/ceph/pull/2371
- 01:26 PM Bug #9128: Newly-restarted OSD may suicide itself after hitting suicide time out value because it...
- any progress on this?
- 06:25 PM Feature #9410 (New): Crush optimization for unbalanced data/pg distribution
- Hi all,
? Several months ago we met an issue of read performance issues (17% degradation) when working on ceph obje... - 06:19 PM Revision f00ff5f3 (ceph): Merge pull request #2430 from yuyuyu101/kvstore-scrub-inconsistence
- Kvstore scrub inconsistence
Reviewed-by: Sage Weil <sage@redhat.com> - 06:17 PM Revision 245a62c8 (ceph): Merge pull request #2429 from yuyuyu101/kvstore-fix-superblock
- KeyValueStore: Fix upgrade from non-exist superblock OSD version error
Reviewed-by: Sage Weil <sage@redhat.com> - 05:44 PM Revision d73eb895 (ceph): PG::can_discard_op: do discard old subopreplies
- Otherwise, a sub_op_reply from a previous interval can stick around
until we either one day go active again and get r... - 05:40 PM Revision 7ddf0a25 (ceph): PG: mark_log_for_rewrite on resurrection
- Fixes: #8777
Backport: firefly
Signed-off-by: Samuel Just <sam.just@inktank.com>
(cherry picked from commit 8346e1075... - 05:30 PM Bug #9241: osdc/Objecter.cc: 1320: FAILED assert(s->get_nref() == 1)
- this is just a bad assert. one thread hits the assert:...
- 05:29 PM Bug #9241 (Fix Under Review): osdc/Objecter.cc: 1320: FAILED assert(s->get_nref() == 1)
- https://github.com/ceph/ceph/pull/2443
- 05:14 PM Bug #9241 (In Progress): osdc/Objecter.cc: 1320: FAILED assert(s->get_nref() == 1)
- sage-bug-9241-a
- 05:24 PM Revision 49a08eb4 (ceph): debian: only B-R yasm on amd64
- Make yasm dependency amd64 only, it isn?t used elsewhere
but breaks x32 (which is mis-detected as amd64)
Signed-off-... - 05:23 PM Revision 0cd989a7 (ceph): Merge pull request #2428 from ceph/wip-yasm
- fix yasm checks for x32
Reviewed-by: Daniel Schepler <dschepler@gmail.com> - 04:59 PM Bug #9326 (In Progress): osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- Note to sam: DOBjectMap xattr lookup/header lookup race due to no longer having per-collection locks?
- 04:45 PM Bug #9408: erasure-code: misalignment
- Local hack that fixes the problem:...
- 04:04 PM Bug #9408 (Resolved): erasure-code: misalignment
- commit:fdeeeb1b6be7a95f473ad33d3344a499a61106a1...
- 04:38 PM Bug #9274: "AssertionError: failed to recover before timeout expired" in upgrade:dumpling-x:stres...
- 04:33 PM Bug #9274: "AssertionError: failed to recover before timeout expired" in upgrade:dumpling-x:stres...
- This seems to have been due to a hung ceph command, I think.
- 03:49 PM Bug #9274 (In Progress): "AssertionError: failed to recover before timeout expired" in upgrade:du...
- 04:37 PM Bug #9272: Test failed on wait_until_healthy in upgrade:dumpling-firefly-x-master-distro-basic-vp...
- probably need to backport the timeout updates to the dumpling suite?
- 04:36 PM Bug #9272: Test failed on wait_until_healthy in upgrade:dumpling-firefly-x-master-distro-basic-vp...
- Slow recovery, it was still actively recovering.
- 03:56 PM Revision 66efcef5 (ceph): Merge pull request #122 from ceph/wip-9357-wusui
- Create dumpling-giant-x stress-split upgrade suite
- 03:54 PM Bug #9372: injectarg boolean option is discarded
- I have no pull request for this ;-) But I can try and make one if that's what you're suggesting.
- 01:18 PM Bug #9372: injectarg boolean option is discarded
- Looks right. Pull request?
- 03:48 PM Bug #9202: Performance degradation during recovering and backfilling
- There are two issues here. First is that we stubbornly wait on a contended pg lock instead of moving to the next ite...
- 03:31 PM Fix #6754 (Fix Under Review): erasure-code: jerasure plugin does not check parameters properly
- 03:05 PM Revision e9b09e8c (ceph): packaging: add all erasure code plugins to RPM packages
- It means distributing a few plugins that are only used for unit testing
but it does not use much disk space and this ... - 02:35 PM Bug #9362 (Resolved): librados, rados_read corrupts memory on timeout
- 01:11 PM Bug #9362 (Pending Backport): librados, rados_read corrupts memory on timeout
- Merged to master in commit:0d66922cfbe5ea6c0c8dc57e2ecf9af58e6926de.
- 01:00 PM Bug #9362: librados, rados_read corrupts memory on timeout
- Thanks, Matthias. Those look like unrelated problems with the master branch.
- 12:41 AM Bug #9362: librados, rados_read corrupts memory on timeout
- Thanks for your investigations! I have checked your patch and it indeed seem to solve the problem with the memory acc...
- 02:14 PM devops Bug #9267: "Gem::DependencyError" in upgrade:dumpling-dumpling-distro-basic-vps
- Note for testing re-run - upgrade/dumpling (ubuntu) on vps'es
- 02:10 PM Bug #9220: objecter doesn't reconnect watch on interval change w/ same primary
- 02:10 PM Bug #9269: ceph osd perf outputs the commit cycle latency for fs_commit_latency
- 02:00 PM devops Bug #9239 (Resolved): Process from starting locking machines till tests start running take long t...
- sending a sighup to dhcpd daemon insures lease file stays fresh without tons of duplicates which prevents DNS from ge...
- 01:58 PM Bug #9365 (Resolved): Logging on class load failure could be better
- 01:51 PM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- Can't reproduce it with ...
- 01:46 PM Bug #9170 (Resolved): erasure-code: preload erasure code plugins
- 01:46 PM Bug #9111 (Won't Fix): PG stuck with 'active+remapped' forever with cluster wide change (add/remo...
- 01:45 PM Bug #9369: init: ceph-osd (...) main process (...) killed by ABRT signal
- Mixture of ext3, ext4, and xfs. The rbd volume that completely lost its partition table was GPT partitioned with an ...
- 01:35 PM Bug #9369: init: ceph-osd (...) main process (...) killed by ABRT signal
- This looks like some kind of local fs corruption. What os/filesystem ar you using?
- 01:45 PM Bug #9277 (Rejected): gitbuilder URL not found in rgw:verify-dumpling-testing-basic-vps
- not enough info here?
- 01:45 PM Bug #9384: OSD is crashing while io is running and querying withadmin socket
- Following pull request has the fix.
https://github.com/ceph/ceph/pull/2440 - 01:44 PM Bug #9303 (Rejected): file deps.deb.txt doesn't exist in github repository
- https://github.com/ceph/ceph/blob/master/deps.deb.txt
...unless i'm missing something?
- 01:38 PM Bug #9364 (Duplicate): "Assertion: osd/Watch.cc: 290: FAILED assert(!cb)" in upgrade:dumpling-dum...
- 01:36 PM Bug #9370 (Resolved): gitbuilder: mon/osd-erasure-code-profile.sh often fail on trusty-i386-basic
- 01:32 PM Bug #9392 (Duplicate): "Segmentation fault" in upgrade:firefly-firefly-testing-basic-vps run
- same as #9040
need to make the test not upgrade the client code while the test workload is running - 09:22 AM Bug #9392 (Duplicate): "Segmentation fault" in upgrade:firefly-firefly-testing-basic-vps run
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-08_19:00:01-upgrade:firefly-firefly-testing-basic-...
- 01:29 PM Bug #8641 (Can't reproduce): Cache tiering agent cannot flush or evict objects during the benchmark
- 01:28 PM Bug #8974 (Can't reproduce): osd crashed with merge_log assert due to removal of isds
- 01:24 PM Bug #9079 (Resolved): osd: bad learned_addr during send_boot
- not worth backporting
- 01:23 PM Bug #9077 (Can't reproduce): Cluster is up in MON node even if Ceph is uninstalled in OSD node
- 01:16 PM Bug #9342: Different implementation for PGTransaction::get_bytes_written between RPGTransaction a...
- Hmm, this doesn't include stuff written to omap.
- 01:14 PM Bug #9381 (Pending Backport): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" e...
- merged to master
- 07:54 AM Bug #9381 (Fix Under Review): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" e...
- "proposed fix":https://github.com/ceph/ceph/pull/2434
- 07:52 AM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- ceph-mon and the plugins are in the "ceph package":https://github.com/ceph/ceph/blob/7d8fe2d994a673f2187bf99ac8e20df6...
- 07:18 AM Bug #9381 (In Progress): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error ...
- 07:17 AM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- It does not look like a duplicate after all. It fails when preloading the lrc erasure code plugin.
- 01:13 PM Bug #9386 (Rejected): Remote transport is ignoring rekey requests
This is just doing the equivalent of an "scp" operation, so it must be an infrastructure issue.- 01:12 PM Bug #9203: ceph_test_rados: ObjectDesc::iterator::advance(bool): Assertion `pos < limit' failed.
- can we retest now that #9054 fix is merged?
- 01:11 PM Bug #9339 (Pending Backport): ReplicatedPG crash in hitset_create
- 01:10 PM Bug #9296 (Duplicate): timeout during recovery, slow requests
- probably dup of #9179
- 01:09 PM Bug #9300 (Can't reproduce): os/DBObjectMap.cc: 399: FAILED assert(!valid || cur_iter->valid())
- looks like a bad disk
- 01:08 PM Bug #9294 (Rejected): invalid read of size 8 in ReplicatedPG::start_flush()
- closed, this patch got reverted
- 12:44 PM Bug #9054 (Pending Backport): ceph_test_rados: FAILED assert(!old_value.deleted())
- 12:42 PM Bug #8777 (Pending Backport): osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == lo...
- 10:45 AM Bug #8777 (Resolved): osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to_riter == log.rbegin())
- 12:41 PM Bug #9179 (Pending Backport): unfound objects, recovery timeout
- 12:40 PM Bug #9293 (Pending Backport): _collection_move_rename EEXIST
- 12:35 PM devops Bug #9407 (Won't Fix): sysvinit: "-a" option and "osd crush update on start" don't work together
- If you use "-a" to restart an OSD on a different node, its crush location gets set to the host executing the restart ...
- 11:54 AM Revision df666549 (ceph): Merge pull request #2305 from majianpeng/fix5
- Make ceph_erasure_code_benchmark more robust.
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 11:49 AM rbd Bug #9405 (Resolved): rbd/test_librbd.sh outputs garbage
- It trips up teuthology and there isn't a reasonable way to protect against it....
- 11:26 AM Feature #9262: Additional namespace issues
- For now we will disable rados export/import (see feature #9403 to re-enable).
- 11:26 AM devops Bug #9404 (Won't Fix): rpm: mount.fuse.ceph is installed in /usr/sbin instead of /sbin
- 11:24 AM Feature #9403 (Duplicate): Make rados import/export fully functional and re-enable
The rados import/export features have been disabled. They don't support namespace. They don't support xattr/omap ...- 10:45 AM Bug #9259 (Resolved): persistent slow request warnings on an osd/pg combo which has not been for ...
- 10:44 AM Bug #9205 (Resolved): osd: notify ops reordered
- Does not need to be backported.
- 10:15 AM Revision c5bafae6 (ceph): erasure-code: Avoid k/m that we set are not equal in erasure-plugin.
- Because erasure-plugin has default k/m and can autotune if k or m
invalid. Check k/m they are the same as we want.
S... - 10:06 AM Feature #9401 (Resolved): RBD: Provide a way to know how much space is used by a clone
- There's currently no easy way to know how much space is used by a RBD clone, the current procedure involves running a...
- 10:02 AM rgw Feature #9400 (In Progress): object versioning: sync agent support
- 10:01 AM rgw Feature #9399 (Resolved): rgw: object versioning, bucket index log support for olh
- olh operations should go to bucket index log.
- 09:58 AM Linux kernel client Bug #9192: krbd: poor read (about 10%) vs write performance
- If it's the queueing change that is causing this, 3.16.x with the fix will show the same results as 3.17, but it woul...
- 06:46 AM Linux kernel client Bug #9192: krbd: poor read (about 10%) vs write performance
- I plan to test on the Ubuntu PPA 3.16.x once the fix gets into that branch. Will that be good enough?
- 03:26 AM Linux kernel client Bug #9192: krbd: poor read (about 10%) vs write performance
- Eric,
To see if it's the new queueing regression, it'd be best to compare wip-request-fn and wip-request-fn~1. Th... - 09:54 AM Revision 9606232b (ceph): KeyValueStore: Fix scrub causing inconsistence state
- Fix #8589
Backport #Firefly
Signed-off-by: Haomai Wang <haomaiwang@gmail.com> - 09:48 AM Revision ae05edda (ceph): ObjectStore Test: Add test for simulating scrub behavior
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 09:45 AM rgw Feature #9397 (Resolved): rgw: object versioning, tie to REST api
- object listing
get object version
remove object version - 09:42 AM rbd Bug #8859 (Closed): krbd crash while serving linux-lio iscsi: rbd_assert(img_request != NULL);
- ...
- 09:41 AM rgw Feature #9396 (Resolved): rgw: object versioning, objclass support for olh
- link olh to an object instance / deletion marker
remove object instance / deletion marker
maintain olh ops log
rea... - 09:39 AM rgw Feature #9395 (Resolved): rgw: object versioning, olh functionality
- Create a new type of object that serves as a soft link to other objects.
- 09:37 AM rgw Feature #9394 (Resolved): rgw: object versioning, support for multiple object instances
- 09:33 AM Bug #9156 (Resolved): SWIFT tests failed in upgrade:dumpling:rgw-dumpling-distro-basic-vps suite
- Looked good on this run - http://pulpito.front.sepia.ceph.com/teuthology-2014-09-04_15:40:01-upgrade:dumpling-dumplin...
- 09:18 AM rbd Bug #9391 (Won't Fix): fio rbd driver rewrites same blocks
- 09:18 AM rbd Bug #8709 (Resolved): stale size reported by ioctl(BLKGETSIZE64) after librbd_resize() returns
- 09:16 AM rgw Bug #9169: 100-continue broken for centos/rhel
- Same in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-08_19:00:01-upgrade:firefly-firefly-testing-basic-vps ...
- 09:15 AM rbd Bug #9078: Removing an RBD is very slow whenever there is write's in other RBD which also belongs...
- Yes, smaller order means more objects to remove...
- 09:14 AM rgw Bug #9254: rgw: civetweb requires explicit \r\n for http headers
- There's an open pull request for it (2337).
- 09:14 AM rbd Bug #8329: qemu-img rpm provided breaks snapshooting functionality on centos
- do we still need to provide these packages? :(
- 09:12 AM rbd Bug #8385 (Need More Info): RBD / QEMU Crash: Invalid fastbin entry (free)
- 09:11 AM rbd Bug #6494 (Pending Backport): High memory consumption of qemu/librbd with enabled cache
- 09:08 AM rgw Bug #9353 (Pending Backport): Log files created under /var/log/radosgw/ do not have the .log exte...
- 09:05 AM rgw Bug #9160 (Closed): rgw failures with 'NoneType' object has no attribute 'get_contents_as_string'
- this appears to be fixed now
- 09:04 AM rgw Bug #9279 (Duplicate): test_s3.test_atomic_* tests failed in rgw:multifs-dumpling-testing-basic-v...
- 09:04 AM rgw Bug #9310 (Duplicate): Rados gateway swift subuser creation problem on FIREFLY
- there is a pending pull request to fix.. #9155
- 08:38 AM rgw Feature #8945: rgw: support swift /info api
- Here is a sample output retrieved from OpenStack Swift:...
- 08:27 AM Revision c6641791 (ceph): KeyValueStore: Fix upgrade from non-exist superblock OSD version error
- If user upgrade OSD from Firefly or lower than 0.84, it won't get valid
keyvaluestore backend.
Signed-off-by: Haomai... - 07:18 AM CephFS Bug #8055 (Can't reproduce): knfs: NFS: nfs4_discover_server_trunking unhandled error -5. Exiting...
- 07:17 AM CephFS Bug #7613 (Can't reproduce): mds/MDCache.cc: 216: FAILED assert(inode_map.count(in->vino()) == 0)
- 07:12 AM CephFS Bug #8757 (Won't Fix): no need to hold write lock on hardlink's dir while creating anchortable entry
- the anchor table is no more, yay!
- 07:08 AM CephFS Bug #8576 (Need More Info): teuthology: nfs tests failing on umount
- 07:07 AM CephFS Bug #9280: valgrind failures in ceph-fuse
- 07:07 AM CephFS Bug #9341 (Need More Info): MDS: very slow rejoin
- 07:04 AM CephFS Bug #5382 (Can't reproduce): mds: failed objecter assert on shutdown
- 06:58 AM Revision cb41c549 (ceph): Merge pull request #2378 from swamireddy/wip-orgnaizationmap
- mailmap updates
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 06:35 AM Revision 6681008b (ceph): mailmap: M Ranga Swami Reddy affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
- 06:12 AM Revision 448342ec (ceph): mailmap: John Spray name normalization
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
Reviewed-by: John Spray <jspray@redhat.com> - 06:12 AM Revision 2f2c37ff (ceph): mailmap: Marco Garcês affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
Reviewed-by: Marco Garcês <marco@garces.cc> - 06:12 AM Revision 02049987 (ceph): mailmap: Erik Logtenberg affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
Reviewed-by: Erik Logtenberg <erik@logtenberg.eu> - 06:00 AM Revision 284fb49d (ceph): mailmap: François Lafont affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
- 03:56 AM Revision c729372a (ceph): Create dumpling-giant-x stress-split upgrade suite
- Fixes: 9357
Signed-off-by: Warren Usui <warren.usui@inktank.com> - 03:30 AM CephFS Bug #9178: samba: ENOTEMPTY on "rm -rf"
- ...
- 02:22 AM Revision 35b3f461 (ceph): Update giant.yaml
- 12:48 AM Revision 8d3f1354 (ceph): Merge pull request #2422 from ceph/wip-6494
- Enforce cache size on read requests
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 12:41 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- The patch is in the queue, just haven't been applied yet: https://lkml.org/lkml/2014/9/3/918.
09/08/2014
- 11:38 PM Revision 8755466c (ceph): dumpling-giant-x: use giant version of workunits
- The firefly version of test_librbd_python at least does
not work on giant.
Signed-off-by: Josh Durgin <josh.durgin@i... - 11:15 PM Bug #9342 (In Progress): Different implementation for PGTransaction::get_bytes_written between RP...
- Here you go :-)
- 08:03 PM Bug #9342: Different implementation for PGTransaction::get_bytes_written between RPGTransaction a...
- Loic, i think i have no permit to change stats.
The pull request:
https://github.com/ceph/ceph/pull/2400 - 11:11 PM Bug #9381 (Duplicate): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in...
- It is a duplicate of http://tracker.ceph.com/issues/9343
- 06:58 PM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- Looks the same on giant (centos and rhel specific (?)) - http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-08_17...
- 03:00 PM Bug #9381: "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in upgrade:dum...
- More from Josh:...
- 02:49 PM Bug #9381 (Resolved): "jerasure load dlopen(/usr/lib64/ceph/erasure-code/libec_lrc.so)" error in ...
- Per Josh analysis: ...
- 10:51 PM Revision 63075361 (ceph): configure: do not try yasm on x32
- Signed-off-by: Sage Weil <sage@redhat.com>
- 09:45 PM Bug #9390 (Duplicate): EEXIST on split due to import/export
- ...
- 09:38 PM Bug #9389 (Duplicate): ec pg stuck peering, did not send query for one shard
- ...
- 09:33 PM Revision f19162ad (ceph): osd: fix osd_tp shutdown
- We need to clear the queue, not just drain the currently executing jobs.
Fixes: #9218
Backport: firefly
Signed-off-b... - 09:04 PM Revision d5777c42 (ceph): doc: Fixed broken links and clarified terms for new authentication docs.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:59 PM Bug #9388 (Duplicate): osd/PG.cc: 2945: FAILED assert(r == 0) in update_snap_map
- ...
- 08:53 PM Bug #9241: osdc/Objecter.cc: 1320: FAILED assert(s->get_nref() == 1)
- ubuntu@teuthology:/a/teuthology-2014-09-08_02:32:01-rados-master-testing-basic-multi/472170...
- 08:51 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- ubuntu@teuthology:/a/teuthology-2014-09-08_02:32:01-rados-master-testing-basic-multi/472355
- 08:49 PM Bug #9386 (Rejected): Remote transport is ignoring rekey requests
- ubuntu@teuthology:/a/teuthology-2014-09-08_02:32:01-rados-master-testing-basic-multi/472416...
- 08:48 PM Bug #9385 (Duplicate): ceph_test_rados: incorrect buffer at pos ...
- ...
- 08:24 PM Revision e2de11a8 (ceph): doc: Minor cleanup.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:13 PM Revision 06e69be0 (ceph): Merge pull request #2424 from ceph/wip-9366
- lttng: Fix 'make tag' when configured --without-lttng
- 08:04 PM Bug #7995: osd shutdown: ./common/shared_cache.hpp: 93: FAILED assert(weak_refs.empty())
- ubuntu@teuthology:/a/teuthology-2014-09-08_02:32:01-rados-master-testing-basic-multi/472466
(most recent osdmap) - 07:49 PM Revision 9ab46dc5 (ceph): debian: only B-R yasm on amd64
- Make yasm dependency amd64 only, it isn?t used elsewhere
but breaks x32 (which is mis-detected as amd64)
Signed-off-... - 07:47 PM Revision aa041f88 (ceph): Merge pull request #2390 from ceph/wip-9311
- Parallelize RBD import/export
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 07:22 PM Revision 0e823720 (ceph): Merge pull request #2426 from ceph/wip-doc-qemu
- doc: Added more robust usage example for uid and conf for QEMU.
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 07:19 PM Revision c050afdb (ceph): Merge remote-tracking branch 'gh/next'
- 07:19 PM Revision 836a6701 (ceph): doc/release-notes: v0.85
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:58 PM Revision 591a1795 (ceph): doc: Added more robust usage example for uid and conf for QEMU.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:50 PM CephFS Bug #9358 (Resolved): mds: standby-replay mds crashes
- 06:26 PM Revision b178e974 (ceph): doc: Clean up syntax to suppress warnings.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:24 PM Revision 4ad1106e (ceph): doc: Clean up syntax.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:48 PM Bug #9384: OSD is crashing while io is running and querying withadmin socket
- I think I have root caused it..
1. OpTracker::RemoveOnDelete::operator() is calling op->_unregistered() which clea... - 05:47 PM Bug #9384 (Resolved): OSD is crashing while io is running and querying withadmin socket
- I faced a crash in OSD with latest Ceph master. Here is the log trace for the same.
ceph version 0.85-677-gd5777c4... - 05:31 PM Revision 598cde42 (ceph): osd/PG: fix crash from second backfill reservation rejection
- If we get more than one reservation rejection we should ignore them; when
we got the first we already sent out cancel... - 05:13 PM Revision 4dbb5010 (ceph): mon/Paxos: don't spam log with is_readable at dout level 1
- Backport: firefly, dumpling
Reported-by: Aanchal Agrawal <Aanchal.Agrawal@sandisk.com>
Signed-off-by: Sage Weil <sage... - 05:12 PM Revision 0a9d49e5 (ceph): mon/Paxos: don't spam log with is_readable at dout level 1
- Backport: firefly, dumpling
Reported-by: Aanchal Agrawal <Aanchal.Agrawal@sandisk.com>
Signed-off-by: Sage Weil <sage... - 05:09 PM Feature #9383 (New): admin socket: allow multiple pending requests
- Right now we have a mutex that is held while processing an admin socket interaction (and possibly a bunch of other st...
- 05:06 PM Revision 312d7401 (ceph): Merge pull request #2421 from ceph/wip-paxos-dout
- mon/Paxos: don't spam log with is_readable at dout level 1
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com> - 04:16 PM Revision 7531b280 (ceph): lttng: Fix 'make tag' when configured --without-lttng
- Signed-off-by: Adam Crume <adamcrume@gmail.com>
- 03:52 PM Bug #9356 (In Progress): ceph_test_rados_striper_api_aio Segmentation faults
- It improves the situation but there still are issues. I believe you can reproduce this on a ubuntu 14.04. Here are tr...
- 05:47 AM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- I've just submitted a fix via pull request 2419. I copy here the comment of the pool request :
"This definitely fixe... - 02:52 AM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- There is indeed a missing lock in RadosStriperImpl, that should serialize rados_req_read_safe and rados_req_read_comp...
- 12:14 AM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- I'm having a look at it. From what I can see so far, we are somehow reusing a deallocated object.
- 03:42 PM Bug #9220: objecter doesn't reconnect watch on interval change w/ same primary
- ceph-qa-suite:wip-notify
- 03:40 PM Bug #9220: objecter doesn't reconnect watch on interval change w/ same primary
- have test
- 03:39 PM Revision 1c346017 (ceph): mon/Paxos: make is_readable debug output show result
- Signed-off-by: Sage Weil <sage@redhat.com>
- 03:34 PM Revision 62ca27d0 (ceph): mon/Paxos: don't spam log with is_readable at dout level 1
- Backport: firefly, dumpling
Reported-by: Aanchal Agrawal <Aanchal.Agrawal@sandisk.com>
Signed-off-by: Sage Weil <sage... - 03:31 PM Bug #9295: osd/OSD.cc: 5501: FAILED assert(session) in ms_fast_dispatch
- teuthology-2014-09-07_02:30:03-rados-next-testing-basic-multi/470990
- 03:12 PM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- Could be the same as here:
http://pulpito.ceph.com/teuthology-2014-09-07_17:08:02-upgrade:dumpling-firefly-x-maste... - 03:03 PM Bug #9219: lost_unfound test got ENOENT: i don't have pgid 1.e
- ubuntu@teuthology:/a/teuthology-2014-09-07_02:30:03-rados-next-testing-basic-multi/470719
same pg even (1.e) - 02:42 PM Bug #9285: osd: promoted object can get evicted before promotion completes
- 02:35 PM Bug #9366: make tags broken by tracing/ files
- awesome. Thanks.
- 01:13 PM Bug #9366 (Resolved): make tags broken by tracing/ files
- merged. and fixed you permissions, adam!
- 01:11 PM Bug #9366: make tags broken by tracing/ files
- There's a wip-9366 branch that fixes this. I issued a pull request, but it's not merged to master, yet. I apparentl...
- 01:05 PM Bug #9366: make tags broken by tracing/ files
- It seems that lttng is configured off by default with my standard practice of "do_autogen.sh -d 1". Regardless, I th...
- 02:33 PM Bug #9218 (Resolved): FAILED assert(peering_queue.empty()) during OSD shutdown
- 01:46 PM Bug #9362 (Fix Under Review): librados, rados_read corrupts memory on timeout
- https://github.com/ceph/ceph/pull/2427
- 10:29 AM Bug #9362: librados, rados_read corrupts memory on timeout
- Sage pointed out elsewhere (and I'm with him) that it looks like the actual response is coming in and then the messen...
- 09:34 AM Bug #9362: librados, rados_read corrupts memory on timeout
- which version of librados is this? Thanks!
- 01:31 PM Revision a0c22842 (ceph): 0.85
- 01:22 PM Bug #9193 (Pending Backport): notify does not return an error code on timeout
- 01:11 PM CephFS Bug #9178: samba: ENOTEMPTY on "rm -rf"
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-07_23:14:01-samba-next-testing-basic-multi/471653/
- 01:10 PM CephFS Bug #6613: samba is crashing in teuthology
- Still popping up:
http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-07_23:14:01-samba-next-testing-basic-multi/... - 01:01 PM CephFS Bug #9341: MDS: very slow rejoin
- If you set the "mds dump cache after rejoin" config option to true it will dump out the cache to a file which we can ...
- 12:48 PM rbd Feature #9311 (Resolved): rbd: parallelize import/export
- commit:4f35714f12843bc726307a615803dccc0828fbda
- 12:11 PM rbd Bug #9380 (Resolved): rbd cache sizing is per image
- Performing IO on a cloned image will result in a unique cache for each layer of the image hierarchy (image, parent im...
- 12:04 PM rbd Bug #9379 (Won't Fix): rbd cache sizing does not account for wasted space due to heap alignment
- The rbd cache uses the data extent size when determining the space utilized by the cache. When writing data in chunk...
- 11:43 AM Revision 27340b1d (ceph): Merge pull request #2395 from ceph/wip-9282
- mds: warn clients which aren't revoking caps
- 11:29 AM CephFS Bug #9177: ceph-fuse: failing MPI mdtest runs
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-05_23:04:02-fs-master-testing-basic-multi/470162/
- 11:29 AM CephFS Bug #8576: teuthology: nfs tests failing on umount
- This is still showing up frequently. One example: http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-07_23:10:01-...
- 11:18 AM Revision d27ac441 (ceph): Merge pull request #2391 from ceph/wip-mds-lock
- Wip mds lock
Reviewed-by: John Spray <john.spray@redhat.com> - 11:17 AM Revision 5561d7eb (ceph): Merge pull request #2404 from ceph/wip-9358
- osdc/Journaler: fix try_read_entry()/wait_for_readable() locking
Reviewed-by: John Spray <john.spray@redhat.com> - 10:35 AM Bug #9369: init: ceph-osd (...) main process (...) killed by ABRT signal
- I eventually decided to take the problematic OSDs down and try reweighting them to 0 in an attempt to get the cluster...
- 10:31 AM Bug #8863 (Resolved): osd: second reservation rejection -> crash
- 10:20 AM Feature #9161: Cache warmup and ejection
- There are (ir)regularly scheduled backlog audits with product and engineering that go through the new tickets and pri...
- 08:34 AM Feature #9161: Cache warmup and ejection
- Thanks Greg, that part is clear enough :) My question is about the life cycle of a feature request - when dopes this...
- 10:18 AM CephFS Feature #9283: mds: limit inodes with caps to <100% of cache
- I don't think we specified any policies when discussing this. If you're thinking something like requesting each clien...
- 06:59 AM CephFS Feature #9283: mds: limit inodes with caps to <100% of cache
- Can't remember if we discussed policies for this at all -- could this be as simple as applying a some fraction to mds...
- 10:10 AM Subtask #9378 (Closed): mon: audit helpers that put() messages but do not get() them
- I noticed at least one helper function that put()s a message without get()ing it. For instance, OSDMonitor::send_late...
- 07:40 AM Linux kernel client Bug #8806: libceph: must use new tid when watch is resent
- My tests confirmed that wip-watch-tid-8806 fixes this particular krbd bug. However, with thrashosds thrown into the ...
- 07:02 AM devops Bug #9376 (Resolved): ceph-release package has '.centos' in its name, breaks installers
- On http://ceph.com/rpm-firefly/el7/noarch/ :...
- 06:45 AM CephFS Feature #9375 (Resolved): Send single 'many clients' health warning instead of N warnings for N c...
Recently added per-client health warnings for 9282, 9284, these should (for some configurable threshold N of client...- 06:43 AM CephFS Feature #9284 (Fix Under Review): mds: warn when clients are not responding to cache pressure
- 06:43 AM CephFS Feature #9282 (Resolved): mds: warn (and kill?) sessions to clients which aren't revoking caps
- Merged the patch that adds clog warnings, addition "ceph status" warnings added in the 9284 branch.
- 06:23 AM rbd Feature #9374 (Resolved): rbd: use a rolling average for bench-write
- bench-write currently calculates a cumulative average, which makes it difficult to see throughput variations. Using a...
- 04:08 AM CephFS Bug #9280: valgrind failures in ceph-fuse
- More:
http://pulpito.ceph.com/teuthology-2014-09-05_23:04:02-fs-master-testing-basic-multi/470180/
http://pulpito... - 03:23 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Guillaume Subiron wrote:
> Xavier Trilla wrote:
>
> > Ok, well try 3.16.2 once it's released and we will let you ... - 03:20 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Xavier Trilla wrote:
> Ok, well try 3.16.2 once it's released and we will let you know.
Sadly, your patch has n...
09/07/2014
- 11:31 PM Fix #6754 (In Progress): erasure-code: jerasure plugin does not check parameters properly
- 11:02 PM CephFS Bug #9341: MDS: very slow rejoin
- To the best of my knowledge this has nothing to do with #9264 and #8962.
Here is MDS restart log:... - 11:02 PM Linux kernel client Feature #9345: look into using MSG_MORE and MSG_SENDPAGE_NOTLAST more
- If I understand correctly, MSG_SENDPAGE_NOLAST is an internal flag and is set by pipe_to_sendpage. Should user be set...
- 08:54 PM Revision 1250c31e (ceph): Merge pull request #2365 from infernix/centos5
- Include types.h after stdint.h to accommodate for older distributions.
Reviewed-by: Loic Dachary <loic-201408@dachar... - 08:50 PM Revision f0e0c315 (ceph): Include types.h after stdint.h to accomodate for older distributions. T...
- Signed-off-by: Gerben Meijer <infernix@gmail.com>
- 05:25 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- It would probably be a good thing to add these tests to make check so they are caught by the gitbuilder ?
- 05:23 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- Something else is still crashing the test but not with valgrind. A race condition of some kind ?...
- 05:16 PM Bug #9356 (Fix Under Review): ceph_test_rados_striper_api_aio Segmentation faults
- "pull request":https://github.com/ceph/ceph/pull/2418
- 05:15 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- ...
- 02:35 PM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
- ...
- 02:18 PM Bug #9342: Different implementation for PGTransaction::get_bytes_written between RPGTransaction a...
- Ma Jianpeng, if you are working on it would you be so kind as to change from *New* to *In Progress* ?
- 08:59 AM Revision 92547926 (ceph): rbd should use write-back when caching is enabled
- librbd now defaults to write-through until the first flush
is received. For rbd, force the use of write-through.
Si... - 07:22 AM Revision 21e84d67 (ceph): Merge pull request #2415 from johnkeepmoving/fix-off-error-in-rbd.cc
- correct error of repeatedly adding io_size to off in do_bench_write
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 06:48 AM Revision 0fb3e52e (ceph): correct error of repeatedly adding io_size to off in do_bench_write
- Signed-off-by: jiangheng <jiangheng0511@gmail.com>
- 03:04 AM Revision 4fc9fffc (ceph): Enforce cache size on read requests
- In-flight cache reads were not previously counted against
new cache read requests, which could result in very large
c...
09/06/2014
- 08:51 PM Revision 7d8fe2d9 (ceph): Merge pull request #2396 from ceph/wip-SCA-20140902
- SCA and Coverity fixes
Reviewed-by: Loic Dachary <loic@dachary.org> - 05:18 PM Bug #9369: init: ceph-osd (...) main process (...) killed by ABRT signal
- If it would be of any use, I have a paired log file and core dump for one of the occurrences.
- 11:50 AM Bug #9369: init: ceph-osd (...) main process (...) killed by ABRT signal
- I've reverted the main pool "size" to 2 in an attempt to get the storage cluster back to a completely healthy state, ...
- 10:40 AM Revision fd8d1fcb (ceph): Merge pull request #2398 from xiaoxichen/refactor_lrc_crush
- Refactor ErasureCodeLrc::create_ruleset
Reviewed-by: Loic Dachary <loic@dachary.org> - 10:26 AM Revision f0a4a2fc (ceph): tests: flush logs before grepping them
- Otherwise the test races with the daemon writing the logs and can
sometime fail.
Signed-off-by: Loic Dachary <loic-2... - 07:25 AM Fix #9245 (Fix Under Review): remove Monitor::osdmonitor_prepare_command
- "proposed fix":https://github.com/ceph/ceph/pull/2413
- 05:54 AM Bug #9372 (Resolved): injectarg boolean option is discarded
- ...
- 03:28 AM Bug #9370 (Fix Under Review): gitbuilder: mon/osd-erasure-code-profile.sh often fail on trusty-i3...
- force rebuild fixed the issue. It is likely to be a race when grepping the logs.
"the proposed fix":https://github.c... - 02:17 AM Bug #9370: gitbuilder: mon/osd-erasure-code-profile.sh often fail on trusty-i386-basic
- forced rebuild to assert it is a transient error. Running ...
- 12:53 AM Bug #9370: gitbuilder: mon/osd-erasure-code-profile.sh often fail on trusty-i386-basic
- compiling manually on the trusty i386 machine to find out why ...
- 12:51 AM Bug #9370 (Resolved): gitbuilder: mon/osd-erasure-code-profile.sh often fail on trusty-i386-basic
- http://gitbuilder.sepia.ceph.com/gitbuilder-ceph-tarball-trusty-i386-basic/log.cgi?log=98b18ed3f109e3c69e219e6262922e...
- 03:11 AM Fix #9371 (New): ceph should fail when an argument is of the wrong type
- Instead it drops the argument for which a type error has been discovered and proceeds with the command. The outcome m...
- 02:14 AM Bug #9304 (Fix Under Review): pool create with invalid crush rule name succeeds
- "proposed patch":https://github.com/ceph/ceph/pull/2411
- 01:01 AM Revision 98b18ed3 (ceph): Merge pull request #2410 from dachary/wip-9368
- erasure-code: init function is extern "C"
- 12:55 AM Revision 03aa113b (ceph): erasure-code: init function is extern "C"
- Otherwise it is not found and the real cause obscured by the test
setup. The thread returns immediately and the timeo...
09/05/2014
- 10:41 PM Bug #9369 (Can't reproduce): init: ceph-osd (...) main process (...) killed by ABRT signal
- All storage nodes are running the same (firefly) version:
$ ceph --version
ceph version 0.80.5 (38b73c67d375a2552d8... - 09:17 PM Revision 484fb857 (ceph): doc: Update logging path to reflect log rotate script default path.
- Fixes: #9351
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 09:09 PM Revision 5a882673 (ceph): Merge pull request #2407 from ceph/wip-formatter-va
- formatter: clean up dump_format()
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 06:59 PM Revision fcf80597 (ceph): Merge pull request #2408 from theanalyst/doc/contributing
- doc: Add a contributing file at the root of repo
Reviewed-by: Loic Dachary <loic@dachary.org> - 06:53 PM Revision f0a1d281 (ceph): doc: Add a contributing file at the root of repo
- Github shows up a `guidelines for contributing` while opening a new pull
request for any repository with a CONTRIBUTI... - 06:02 PM Bug #9368 (Resolved): ErasureCodePluginRegistryTest intermitent failure
- 06:00 PM Bug #9368 (Fix Under Review): ErasureCodePluginRegistryTest intermitent failure
- "need review":https://github.com/ceph/ceph/pull/2410
- 03:26 PM Bug #9368: ErasureCodePluginRegistryTest intermitent failure
- ...
- 03:20 PM Bug #9368 (Resolved): ErasureCodePluginRegistryTest intermitent failure
- Happens on a regular basis at gitbuilder-ceph-tarball-precise-amd64-basic ...
- 05:48 PM Revision 4c4131c6 (ceph): Merge pull request #2406 from alram/master
- rgw: add .log to default log path
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 05:36 PM Revision 46732420 (ceph): rgw: add .log to default log path
- Fixes: #9353
Signed-off-by: Alexandre Marangone <alexandre.marangone@inktank.com> - 05:29 PM Revision 6fac8152 (ceph): formatter: clean up dump_format()
- Create a common dump_format_va() function, and make all the different
variants call it.
Signed-off-by: Yehuda Sadeh ... - 04:40 PM CephFS Bug #9341: MDS: very slow rejoin
- how may data files do you have? I guess the MDS is doing backpointer lookup, we can make the client first trim its ca...
- 02:21 PM CephFS Bug #9341: MDS: very slow rejoin
- I wonder if this is the same as some of our own recent tickets (see #9264 and #8962). What size journal does the MDS ...
- 04:06 PM Revision 8c602868 (ceph): Merge pull request #2329 from theanalyst/f/s3_bucket_loc
- WIP: S3: Implementation for S3 Get Bucket Location
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 03:56 PM Revision 6cac39e8 (ceph): Merge pull request #2405 from dachary/wip-workunits
- qa: scripts to help test ceph-qa-suite
Reviewed-by: Cheng Cheng <ccheng.leo@gmail.com> - 03:49 PM Revision df3c70a9 (ceph): qa: scripts to help test ceph-qa-suite
- This workunit will be used by tests as a placeholder that always return
true. This is helpful in tests when a script ... - 02:36 PM Bug #9367 (In Progress): erasure-code: use LRU cache for decoding in ISA
- ...
- 02:10 PM Bug #9367: erasure-code: use LRU cache for decoding in ISA
- "implementation for ISA":https://github.com/ceph/ceph/pull/2409 being tested with thrashosd and "isa workload":https:...
- 02:10 PM Bug #9367 (Resolved): erasure-code: use LRU cache for decoding in ISA
- because it is expensive to compute
- 02:21 PM Revision 2280736c (ceph): osdc/Journaler: fix try_read_entry()/wait_for_readable() locking
- These two functions should be protected by the Journal::mutex. Furthermore,
wait_for_readable() is racy. The journal ... - 02:18 PM Documentation #9351 (Resolved): Change RGW log file in config documentation
- Changed path as suggested.
- 02:11 PM Documentation #9351 (In Progress): Change RGW log file in config documentation
- 02:17 PM Feature #9161: Cache warmup and ejection
- This is a feature request ticket; the code implementing it doesn't exist yet. ;)
- 02:11 PM Bug #9366: make tags broken by tracing/ files
- This only happens when configured without lttng, whether explicitly or implicitly.
- 02:01 PM Bug #9366 (Resolved): make tags broken by tracing/ files
- Something's going wrong in tracing/Makefile.am with make tags:...
- 01:04 PM Bug #9365 (Resolved): Logging on class load failure could be better
- Florian points out that if you're missing cls_rbd on your system, the debug is just horrible at figuring that out. I...
- 12:03 PM Bug #9363 (Resolved): locking vps results in libvirt error
- Machine was down. SOL logging disconnected so didn't get a lot of info on what happened.
Powercycled the machine. - 07:06 AM Bug #9363 (Resolved): locking vps results in libvirt error
- ...
- 10:39 AM Bug #9356: ceph_test_rados_striper_api_aio Segmentation faults
Same issue with the ceph_test_rados_striper_api_io test.
$ gdb .libs/lt-ceph_test_rados_striper_api_io
GNU gdb ...- 08:57 AM Bug #9364 (Duplicate): "Assertion: osd/Watch.cc: 290: FAILED assert(!cb)" in upgrade:dumpling-dum...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-04_15:40:01-upgrade:dumpling-dumpling-distro-basic...
- 08:21 AM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- I was able to reproduce this on manual run.
- 06:09 AM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- I'll try to reproduce German's fio problem.
To those who are stuck on 3.15 for some reason but can rebuild their k... - 05:36 AM Linux kernel client Bug #8806 (Fix Under Review): libceph: must use new tid when watch is resent
- wip-watch-tid-8806
- 05:31 AM Bug #9362: librados, rados_read corrupts memory on timeout
- I have also tried to reproduce this problem with a firewall dropping incoming packages from the primary osd the objec...
- 05:08 AM Bug #9362 (Resolved): librados, rados_read corrupts memory on timeout
- If you configure librados with rados_osd_op_timeout, timeouts on rados_read will result in memory corruptions and seg...
- 05:07 AM Revision 94173a37 (ceph): mds: warn clients which aren't revoking caps
- Add a list to Locker to track revoking caps. print a warning message if client does not
release caps within the given... - 04:53 AM Bug #9361 (Resolved): "ceph -w" sometimes segfaults on Ctrl+C
- # ceph -w
cluster d4fba9f9-3fa8-4f5f-a81a-acc312cb0152
health HEALTH_OK
monmap e2: 3 mons at {devenv... - 04:23 AM Revision 06fca35e (ceph): Merge pull request #119 from ceph/wip_9352
- Wip 9352
- 03:36 AM Revision d8e672f3 (ceph): rgw: Implementation for S3 Get Bucket Location
- S3 API supports getting the location for a bucket, which gives out one
of those geographic zones (US-WEST-1, EU for e... - 03:34 AM Revision 2fb51b2b (ceph): Formatter: add a dump_format_ns to base Formatter
- `dump_format_ns` is a generic formatter to dump a simple format along
with a namespace. It is `dump_format` with an o... - 03:34 AM Revision eaa2bb0a (ceph): rgw_formats: dump_format_ns implementation
- Since the base formatter was extended with a dump_format_ns class,
implementing this here. For now, this is exactly s... - 03:34 AM Revision 9e503b56 (ceph): test/formatter: Add tests for dump_format_ns
- Adding basic unit test to test the new formatter class'
dump_format_ns. Since the functionality only affects XML (and... - 01:35 AM rgw Feature #9359 (Resolved): rgw: Export user stats in get-user-info Adminops API
- The user stats is useful information to view the space usage and total number of objects aggreagated at user level. I...
- 01:21 AM Revision 14d87bd5 (ceph): Refactor ErasureCodeLrc::create_ruleset
- Refactor ErasureCodeLrc::create_ruleset
1.Use CrushWrapper instead of crush C-api.
2.Make ruleset_id == rule_id, cons... - 01:07 AM CephFS Bug #9358 (Resolved): mds: standby-replay mds crashes
- ...
- 12:56 AM Revision b32712f1 (ceph): Merge pull request #2375 from ceph/wip-8648
- mds: clear objects' dirty flags after log segment is expired
- 12:33 AM Revision eae88dad (ceph): mds: clear objects' dirty flags after log segment is expired
- When standby-replay MDS detects a log segment is expired, it should check
the expired segment's dirty lists and clear... - 12:33 AM Revision 677bb45b (ceph): Merge pull request #2393 from dachary/wip-warning
- erasure-code: warning fix
Reviewed-by: David Zafman <dzafman@redhat.com>
09/04/2014
- 11:13 PM CephFS Bug #4746 (Resolved): client: invalidate callback can deadlock
- client does async invalidate now
- 10:49 PM CephFS Bug #8648 (Resolved): Standby MDS leaks memory over time
- fixed by commit eae88dad4c32e4bb5fb255ec4bf1be18b09d498e
- 10:21 PM Revision 5864369a (ceph): Merge pull request #118 from dachary/wip-more-osds
- add osds to firefly-x upgrade suite
- 09:32 PM Revision 80242da4 (ceph): Updated README
- Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
- 09:24 PM Revision 8c43f475 (ceph): First draft for a new suite dumpling-giant-x
- Signed-off-by: Yuri Weinstein <yuri.weinstein@inktank.com>
- 09:03 PM Bug #9356 (Closed): ceph_test_rados_striper_api_aio Segmentation faults
- "mail thread":https://www.mail-archive.com/ceph-devel@vger.kernel.org/msg20440.html...
- 06:27 PM Revision 5315cf0a (ceph): doc: add note on soft JS dependency for navigating docs
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 657be818375bea2d8b5998ea1e5505eedc2... - 06:27 PM Revision 707076e3 (ceph): doc: fix missing bracket
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 69638dfaeb0dcd96dac4b5f5c00ed080424... - 06:27 PM Revision 2ebdcfd5 (ceph): doc: attempt to get the ayni JS into all head tags
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 35663fa55ac1579a3b0c8b67028a3a8dfea... - 06:27 PM Revision 1660503f (ceph): doc: add note on soft JS dependency for navigating docs
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 657be818375bea2d8b5998ea1e5505eedc2... - 06:26 PM Revision 2f075d44 (ceph): doc: fix missing bracket
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 69638dfaeb0dcd96dac4b5f5c00ed080424... - 06:26 PM Revision 4dacb593 (ceph): doc: attempt to get the ayni JS into all head tags
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 35663fa55ac1579a3b0c8b67028a3a8dfea... - 05:51 PM Revision bbb77006 (ceph): Merge pull request #2399 from ceph/wip-8366
- doc: adds a navigational menu for docs
Reviewed-by: John Wilkins <john.wilkins@inktank.com> - 05:44 PM Revision 69638dfa (ceph): doc: fix missing bracket
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 05:44 PM Revision 35663fa5 (ceph): doc: attempt to get the ayni JS into all head tags
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 05:28 PM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- >Sage Weil wrote:
> Hmm, and this is reproducible? It may be the throttling in the msgr layer.. debug ms = 20 might... - 04:37 PM Linux kernel client Bug #9355 (Closed): rbd: map fails with EINVAL inside a container
- https://lists.linuxcontainers.org/pipermail/lxc-users/2013-October/005795.html
carmstrong on irc had the same issu... - 04:36 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- We're seeing this on CoreOS, kernel 3.15.8. Upgrading the kernel ourselves isn't an option, so this'll block RBD usag...
- 03:47 PM rgw Bug #9169: 100-continue broken for centos/rhel
- Per Yehuda - it's a timeout issue
- 03:30 PM rgw Bug #9307: "s3.test_multipart_upload_multiple_sizes ... ERROR" in upgrade:dumpling-firefly-x-mast...
- Here is similar failure in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-04_11:08:01-upgrade:dumpling-firefl...
- 02:16 PM rgw Bug #9353 (Resolved): Log files created under /var/log/radosgw/ do not have the .log extension
- The radosgw creates log files in /var/log/radosgw/ when no "log file" option is set.
The log filename is <client_... - 02:09 PM Revision 409c9556 (ceph): OSDMonitor.cc: fix potential division by zero
- Fix for:
CID 1232602 (#1 of 1): Division or modulo by zero (DIVIDE_BY_ZERO)
divide_by_zero: In expression num_pg_co... - 01:14 PM Documentation #9351 (Resolved): Change RGW log file in config documentation
- http://ceph.com/docs/master/radosgw/config/#add-a-gateway-configuration-to-ceph
The current documentation show the... - 12:58 PM Revision 7b772107 (ceph): objectstore/store_test.cc: fix unintentional integer overflow
- CID 1232603 (#1 of 1): Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)
overflow_before_widen: Potentially ove... - 12:36 PM Revision 6b704832 (ceph): os/HashIndex.cc: fix unintentional integer overflow
- CID 1232604 (#1 of 1): Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)
overflow_before_widen: Potentially ove... - 11:57 AM Revision f29e5b33 (ceph): mount.ceph.c: free memory before return
- CID 1219610 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable buf going out of scope leaks the
sto... - 11:14 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- "running upgrade tests":http://pulpito.ceph.com/loic-2014-09-05_00:02:14-upgrade:firefly-x:stress-split-wip-9343-eras...
- 08:50 AM Feature #9343: erasure-code: allow upgrades for lrc and isa plugins
- "pending pull request":https://github.com/ceph/ceph/pull/2397
- 08:44 AM Feature #9343 (Resolved): erasure-code: allow upgrades for lrc and isa plugins
- When upgrading from Firefly to Giant, an erasure coded pool using the two newly supported plugins (lrc & isa) must on...
- 11:01 AM Revision 903af9c4 (ceph): st_rados_create_pool.cc: cleanup before return from run()
- CID 739605 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable io_ctx going out of scope leaks the
st... - 10:49 AM Linux kernel client Feature #9345 (Resolved): look into using MSG_MORE and MSG_SENDPAGE_NOTLAST more
- Mike Christie pointed out the following:...
- 10:45 AM Linux kernel client Feature #9344 (Resolved): add a TCP_NODELAY option
- From Chaitanya Huilgol <Chaitanya.Huilgol@sandisk.com>...
- 10:20 AM Revision ece990dc (ceph): rados_list_parallel.cc: cleanup before return from run()
- Fix for:
CID 717112 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable io_ctx going out of scope lea... - 08:46 AM Feature #7291: EC: add mechanism for mon to detect and whitelist EC plugins which are globally av...
- Interim solution addressed by http://tracker.ceph.com/issues/9343
- 08:14 AM Revision 135ccad1 (ceph): erasure-code: warning fix
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 08:04 AM Revision 36554f90 (ceph): Merge pull request #2392 from majianpeng/fix1
- FileStore: Remove unused code.
Reviewed-by: Loic Dachary <loic@dachary.org> - 07:20 AM Bug #9342 (Resolved): Different implementation for PGTransaction::get_bytes_written between RPGTr...
- The l_osd_op_w_inb use this value. So the perf-dump display different value when write same object to erasure-pool/re...
- 07:02 AM Feature #9161: Cache warmup and ejection
- I'm not familiar with the process here, looks like this was opened 17 days ago though -- is there something else that...
- 06:01 AM Revision 7de8d17d (ceph): FileStore: Remove unused code.
- Signed-off-by: Ma Jianpeng <jianpeng.ma@intel.com>
- 05:25 AM CephFS Feature #9282 (Fix Under Review): mds: warn (and kill?) sessions to clients which aren't revoking...
- https://github.com/ceph/ceph/pull/2395
- 04:08 AM Revision 1bcf4045 (ceph): Merge pull request #2373 from ceph/wip-9026
- client: vxattr support for stat
Reviewed-by: Greg Farnum <greg@inktank.com> - 03:31 AM Revision 41df414b (ceph): mds: don't take mds_lock when doing journal operations
- Now journaler and objecter have internal locks to protect themself, there is
no need to take the mds_lock before usin... - 03:05 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- The following teuthology job passes:...
- 12:26 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- running rados thrasher on "the proposed fix branch":https://github.com/ceph/ceph/tree/wip-8914-unlink-fdcache
- 01:52 AM CephFS Bug #9341 (Resolved): MDS: very slow rejoin
- I had fiasco trying to use CephFS as network share: today restart of MDS (i.e. down time) took ~3 hours most of which...
09/03/2014
- 09:08 PM CephFS Feature #9026 (Resolved): client: vxattr support for rctime, rsize, etc.
- Merged in with commit:1bcf40459c48af5724c5eba672fdf2cc976c42ea
- 08:59 PM Revision 7ccabbf9 (ceph): librbd.cc: add check to prevent null deref in rbd_list()
- Fix for :
CID 1192467 (2 of 2): Explicit null dereferenced (FORWARD_NULL)
var_deref_model: Passing null pointer ibu... - 06:04 PM CephFS Feature #5486: kclient: make it work with selinux
- I think cephfs part is ready for selinux support. but ceph is not included in selinux policy.
- 05:40 PM Bug #9340 (Won't Fix): ceph mon crashes when tried with rocksdb on precise
- ceph branch: master [0.84-976-geadc352]
when tried testing rocksdb on ubuntu precise, ceph-mon crashed.... - 05:27 PM Revision 4f35714f (ceph): Parallelize RBD import/export
- Use librbd aio_read/_write calls where possible to improve speed.
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 03:54 PM Bug #9293: _collection_move_rename EEXIST
- 03:53 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- 03:52 PM Revision f53bf532 (ceph): PendingIO.cc: use static_cast instead of c-style cast
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 03:51 PM Revision cbd324da (ceph): FileStore.cc: use static_cast instead of c-style cast
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 03:48 PM Bug #9339 (Resolved): ReplicatedPG crash in hitset_create
ceph version 0.84-376-g970d983 (970d9830a3a6e8568337c660fb8b4c4a60a2b3bf)
1: ceph-osd() [0x9a676a]
2: (()+0xfc...- 02:18 PM Revision c7a05b91 (ceph): Merge pull request #1 from noahdesu/doc
- doc: update README with examples
- 02:16 PM Revision d509f9f5 (ceph): doc: update README with examples
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 02:15 PM Revision ae857db9 (ceph): travis: install ceph in ci environment
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 12:52 PM Revision eadc352d (ceph): Merge remote-tracking branch 'origin/next'
- 12:39 PM Revision 0f92657d (ceph): Merge pull request #2377 from ceph/wip-session-stats
- mds: cap and lease count in `session ls` asok
Reviewed-by: Yan, Zheng <ukernel@gmail.com> - 12:39 PM Revision ebd11529 (ceph): Merge pull request #2376 from ceph/wip-request-count
- mds: Fix Session::get_request_count
Reviewed-by: Yan, Zheng <ukernel@gmail.com> - 11:43 AM RADOS Fix #6109: pg <pgid> mark_unfound_lost fails if a completely-gone OSD still in map
- Workaround suggested by Craig Lewis : recreate the OSDs that Ceph wants to probe. It doesn't have to have anything o...
- 11:20 AM CephFS Feature #9338 (Resolved): A read timeout argument for cephfs mounts (fuse client especially)
- When a cluster's cephfs mount hangs, for reasons such as IO pausing due to a cluster going full or osds crashing, the...
- 10:20 AM devops Bug #9332: Placeholder for "SSH connection was lost" errors
- In http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_19:00:01-upgrade:firefly-firefly-testing-basic-vps/46682...
- 09:46 AM devops Bug #9332: Placeholder for "SSH connection was lost" errors
- In http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:25-rgw:singleton-dumpling-testing-basic-vps/466691...
- 09:00 AM devops Bug #9332: Placeholder for "SSH connection was lost" errors
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps/466743/teut...
- 08:56 AM devops Bug #9332: Placeholder for "SSH connection was lost" errors
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps/466734/teut...
- 08:54 AM devops Bug #9332 (Rejected): Placeholder for "SSH connection was lost" errors
- http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps/466738/teut...
- 10:16 AM Cleanup #9337 (Won't Fix): valgrind requires LIBOSD
- For some reason if unittest_sharedptr_registry is not linked with LIBOSD, it will not be valgrind clean. As if a glob...
- 10:04 AM devops Bug #9278 (Resolved): gitbuilder bad rpm URL in rgw:verify-dumpling-testing-basic-vps
- This should be resolved now that we have more notcmalloc gitbuilders.
- 10:02 AM devops Bug #9334 (Duplicate): "Could not find a valid gem 'mime-types' (< 2.0) in any repository" error ...
- duplicate of #9267
- 09:22 AM devops Bug #9334: "Could not find a valid gem 'mime-types' (< 2.0) in any repository" error in rgw:multi...
- Similar - http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps/4...
- 09:17 AM devops Bug #9334 (Duplicate): "Could not find a valid gem 'mime-types' (< 2.0) in any repository" error ...
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps...
- 10:02 AM devops Bug #9335 (Duplicate): "Error installing chef" in rgw:multifs-dumpling-testing-basic-vps suite
- duplicate of #9267
- 09:21 AM devops Bug #9335 (Duplicate): "Error installing chef" in rgw:multifs-dumpling-testing-basic-vps suite
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testing-basic-vps...
- 09:55 AM Bug #9333 (Won't Fix): SharedLRU lookup resurects cleared key
- Apparently it is by design. Lookup will look for the key in weak_refs and will assume that if it finds it there, it i...
- 09:08 AM Bug #9333 (Won't Fix): SharedLRU lookup resurects cleared key
- The "SharedLRU unittests":https://github.com/ceph/ceph/pull/2386 fail on...
- 09:34 AM rgw Bug #9279: test_s3.test_atomic_* tests failed in rgw:multifs-dumpling-testing-basic-vps suite
- Same issues in http://pulpito.front.sepia.ceph.com/teuthology-2014-09-02_16:17:37-rgw:verify-dumpling-testing-basic-v...
- 09:26 AM devops Bug #9267: "Gem::DependencyError" in upgrade:dumpling-dumpling-distro-basic-vps
- Also see the same in rgw tests.
Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_16:17:54-rgw:m... - 08:19 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- Guang Yang wrote:
> @radosgw
> $ ceph daemon /var/run/ceph/ceph-client.*.asok objecter_requests
> { "ops": [
> ... - 06:40 AM Bug #9008: Objecter: pg listing can deadlock when throttling is in use
- @radosgw
$ ceph daemon /var/run/ceph/ceph-client.*.asok objecter_requests
{ "ops": [
{ "tid": 15582485,
... - 08:13 AM Bug #9294 (Pending Backport): invalid read of size 8 in ReplicatedPG::start_flush()
- 08:12 AM Bug #9294 (Resolved): invalid read of size 8 in ReplicatedPG::start_flush()
- 08:11 AM Bug #9309 (Resolved): common/lockdep.cc: 127: FAILED assert(last_id < 1000)
- 08:09 AM CephFS Bug #9123 (Can't reproduce): kceph: had 130k+ inodes with write caps
- going with zheng's theory :)
- 08:08 AM CephFS Bug #9121 (Resolved): mds: inode stuck recovering after client restart
- 08:07 AM CephFS Bug #8962 (Resolved): kcephfs: client does not release revoked cap
- old bug; restarted those clients.
- 06:48 AM Revision c0650cf9 (ceph): added a test to use single disk for both data and journal.
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 06:42 AM Revision 1749ba91 (ceph): added dmcrypt and separate_journal_disk options
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 06:36 AM CephFS Bug #9212 (Won't Fix): mon election delays mds beacon
- ...
- 05:41 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- The proposed fix greatly reduces the window of opportunity for this problem to happen. It can however still happen an...
- 05:21 AM Fix #8914 (Fix Under Review): osd crashed at assert ReplicatedBackend::build_push_op
- "proposed change":https://github.com/ceph/ceph/pull/2384
- 05:31 AM RADOS Feature #9328: osd: generalize the scrub workflow
- ...
- 05:30 AM RADOS Feature #9328 (New): osd: generalize the scrub workflow
- The "scrub workflow":https://github.com/ceph/ceph/blob/firefly/src/osd/PG.cc#L3856 collects information and use them....
- 02:14 AM Bug #9327 (Rejected): Usability Issue: Ceph-deploy does not print all the commands which it is ex...
- Issue description: Noticed that during the osd prepare command ceph-deploy script is not printing all the command whi...
- 12:04 AM CephFS Feature #3289 (Resolved): ceph-fuse: somehow exert pressure on the VFS to remove dentries from th...
- resolved by commit 1d0c62facf client: try shrinking kernel inode cache when trimming session caps
09/02/2014
- 10:30 PM Revision ca6eb612 (ceph): client: vxattr support for rstat
- copy the kernel client code (which uses table to present vxattrs)
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 09:09 PM Revision 299f28b0 (ceph): Merge pull request #2363 from ceph/wip-9285
- osd/ReplicatedPG: do not evict blocked objects
Reviewed-by: Samuel Just <sam.just@inktank.com> - 09:02 PM Revision 8657022c (ceph): Merge pull request #2343 from ceph/wip-9151
- Wip 9151 + 9152
Reviewed-by: Sage Weil <sage@redhat.com> - 08:43 PM Revision 79989de8 (ceph): Merge pull request #112 from ceph/wip-8231-forreview
- Wip 8231 forreview
Reviewed-by: Samuel Just <sam.just@inktank.com> - 07:47 PM Revision d0e0eb47 (ceph): Merge pull request #2362 from ceph/wip-9294
- osd/ReplicatedPG: avoid dereferencing iterator at end()
Reviewed-by: Samuel Just <sam.just@inktank.com> - 07:24 PM rbd Bug #6494: High memory consumption of qemu/librbd with enabled cache
- We are experiencing the same issue, even if we are using quite different software versions.
We have our setup runn... - 06:40 PM Revision ed62ae58 (ceph): Merge pull request #2322 from ceph/wip-doc-authentication
- doc: Refactored authentication docs.
Reviewed-by: John Wilkins <john.wilkins@inktank.com> - 06:38 PM Revision c2443b95 (ceph): doc: Provided additional detail on daemon-specific caps syntax.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:37 PM Revision 8569b93c (ceph): doc: Updated Keyring settings from comments and ceph-deploy defaults.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:26 PM Revision 275df440 (ceph): Fix FTBFS on alpha due to incorrect check on BLKGETSIZE
- Ceph FTBFS on Alpha with:
~~~~
libtool: compile: g++ -DHAVE_CONFIG_H -I. -D__CEPH__ -D_FILE_OFFSET_BITS=64 -D_REENT... - 05:18 PM Revision 9c98d555 (ceph): Merge pull request #2372 from dachary/wip-disk-thread-ioprio
- documentation: document disk thread ioprio options
Reviewed-by: Volker Voigt <volker.voigt@1und1.de> - 05:08 PM Revision b6a97d83 (ceph): documentation: disk thread ioprio options
- Signed-off-by: Loic Dachary <loic-201408@dachary.org>
- 04:47 PM devops Bug #9267: "Gem::DependencyError" in upgrade:dumpling-dumpling-distro-basic-vps
- Several same errors in http://pulpito.front.sepia.ceph.com/teuthology-2014-09-02_16:17:54-rgw:multifs-dumpling-testin...
- 04:07 PM Bug #9326 (Resolved): osd crash in upgrade:dumpling-firefly-x-master-distro-basic-vps suite
- Logs are in http://qa-proxy.ceph.com/teuthology/teuthology-2014-09-02_10:22:50-upgrade:dumpling-firefly-x-master-dist...
- 04:03 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- 2014-08-31 01:24:18.568610 7f45febb1700 10 osd.2 pg_epoch: 192 pg[2.1( v 192'1575 (0'0,192'1575] local-les=105 n=49 e...
- 03:46 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- nvm, reopening
- 02:06 PM Bug #9054 (Resolved): ceph_test_rados: FAILED assert(!old_value.deleted())
- Opening new bug for that one.
- 02:05 PM Bug #9054: ceph_test_rados: FAILED assert(!old_value.deleted())
- 2014-08-31T01:36:27.057 INFO:tasks.rados.rados.0.plana80.stdout:update_object_version oid 348 v 3674 (ObjNum 1096 sna...
- 03:46 PM Bug #9324 (Rejected): FAILED assert(!old_value.deleted()), expect (ObjNum 34580816 snap 0 seq_num...
- 02:07 PM Bug #9324 (Rejected): FAILED assert(!old_value.deleted()), expect (ObjNum 34580816 snap 0 seq_num...
- ubuntu@teuthology:/a/sage-2014-08-30_20:39:25-rados-wip-sage-testing-testing-basic-multi/462661
probably a bug or ... - 03:24 PM CephFS Documentation #8258 (Resolved): 0.80~rc1: outdated MDS man page
Oops, forgot to close this at the time....- 03:23 PM CephFS Bug #9252 (Resolved): Worker thread to advance MDS in absence of messages
- ...
- 03:23 PM CephFS Bug #9152 (Resolved): mds: beacon needs to not take mds_lock
- ...
- 03:23 PM CephFS Bug #9151 (Resolved): mds should log/error/warn when segments are NOT getting trimmed
- ...
- 03:19 PM Revision 1b9e670e (ceph): systemd support in ceph-disk activate
- previous code just fails for systemd with an error
Signed-off-by: Owen Synge <osynge@suse.com> - 02:38 PM Bug #9285: osd: promoted object can get evicted before promotion completes
- This appears to be independent of promotion. Is the problem not simply that we don't guarantee that the object will ...
- 12:01 AM Bug #9285: osd: promoted object can get evicted before promotion completes
- Tested Sage's the pull request, checking the object is blocked or not doesn't work. Actually this check is already do...
- 02:05 PM devops Tasks #8366 (In Progress): Update ceph.com/docs to default to the latest major release (0.80)
- 01:58 PM devops Feature #7925: Feature: create new download.ceph.com site
- 01:55 PM Feature #9323 (Resolved): osdmonitor: allow users to set/clear cluster state flags (CEPH_OSDMAP_F...
- While investigating a cluster that was failing to remove the OSDMap's full flag, we wished we had a command that woul...
- 01:54 PM Bug #9321: pgmap updates from OSDMap can be delayed indefinitely
- I should also note that I *suspect* this condition might have been exacerbated by our full map handling. We probably ...
- 01:44 PM Bug #9321 (Resolved): pgmap updates from OSDMap can be delayed indefinitely
- We saw a customer cluster in which a full OSD had been removed from the OSDMap, but after almost two hours that chang...
- 01:48 PM Bug #9322 (Resolved): OSDMap updates from pgmap can be delayed indefinitely
- Like #9321, but in the other direction: we try to update the osdmap from the pgmap in OSDMonitor::tick(), but if pgmo...
- 01:47 PM rgw Feature #9013 (Fix Under Review): rgw: set civetweb as a default frontend
- 01:40 PM Feature #9320 (Rejected): "ceph osd dump" does not flag full OSDs
- Right now, when there are full OSDs in the cluster, the only way to find them is by running "ceph health detail". The...
- 01:15 PM Revision 6ec08e06 (ceph): client/Client.cc: fix null pointer checks for 'in'
- Cleanup checks for null pointer dereference of 'in' use a
wider code block the cover 'in'.
Signed-off-by: Danny Al-G... - 01:06 PM Revision f5e7a4be (ceph): mds: use HeartbeatMap to control beacons
- ...so that if something hogs mds_lock without
pinging the heartbeat map, we will stop sending
beacons to the mon and ... - 01:06 PM Revision 12ab937a (ceph): messages: add health fields to MMDSBeacon
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:06 PM Revision 0fc8a0da (ceph): mds: populate health metrics in beacon
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:06 PM Revision 1058a9e1 (ceph): mon: handle beacon health metrics in MDSMonitor
- Fixes: #9151
Signed-off-by: John Spray <john.spray@redhat.com> - 01:06 PM Revision 9ee80152 (ceph): mds: add thread to progress queues outside dispatch
- This speeds up processing of queued waiters.
Fixes: #9252
Signed-off-by: John Spray <john.spray@redhat.com> - 01:06 PM Revision c8d23880 (ceph): mon: add MDS metric metadata to health detail
- This is a bit wonky because the mon health structure expects
a string, and we have a key-val structure, but it's bett... - 01:06 PM Revision e69f3214 (ceph): include/CompatSet: make merge const()
- With usual shenanigans to deal with [] on map
not being const-safe.
Signed-off-by: John Spray <john.spray@redhat.com> - 01:06 PM Revision 17d547ed (ceph): mds: extract beacon state from MDS lock
- For clarity, give it its own class for the
state protected by the separate lock too.
Fixes: #9152
Signed-off-by: Jo... - 01:06 PM Revision 232757f4 (ceph): mds: promote Beacon to be a Dispatcher
- This allows it to handle its own MSG_MDS_BEACON
messages from the mon, outside of mds_lock.
This is less important t... - 01:06 PM Revision 124856ab (ceph): include/CompatSet: more const-ness
- Signed-off-by: John Spray <john.spray@redhat.com>
- 01:06 PM Revision 681f4f8b (ceph): mds: const getters in MDLog
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:56 PM Revision 46c94e95 (ceph): mds/CInode.h: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:56 PM Revision a5468abe (ceph): ceph_objectstore_tool.cc: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:18 PM Revision 4d9927e0 (ceph): mds: cap and lease count in `session ls` asok
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:18 PM Revision ea2a8d04 (ceph): mds: Fix Session::get_request_count
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:15 PM Revision 15df8ac4 (ceph): rbd_replay/Replayer.cc: pass const string parameter by reference
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:03 PM Revision 9d926f1e (ceph): ReplicatedPG.cc: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 12:03 PM Revision dae05938 (ceph): IndexManager.cc: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:57 AM Revision 40189a28 (ceph): PGMonitor.cc: prefer empty() over size() for emptiness check
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:53 AM Revision 99fc33f9 (ceph): ErasureCodeLrc.cc: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:49 AM Revision 2dcfe0fe (ceph): ErasureCodePlugin.cc: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:42 AM Revision a754ce51 (ceph): ErasureCodeLrc.cc: fix -Wmaybe-uninitialized compiler warning
- Init 'n' with 0 to prevent uninitialized integer.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de> - 11:35 AM Feature #9313: Add flag to ceph-deploy to start services manually (skip Upstart)
- Forgot to include - this is the error that ceph-deploy generates while it's waiting for the monitors to come up:
a... - 11:25 AM Feature #9313 (Closed): Add flag to ceph-deploy to start services manually (skip Upstart)
- I am trying to deploy Ceph entirely within Docker containers. I have no Upstart daemon running in my container - as i...
- 11:22 AM Revision 38e6c64a (ceph): test_rgw_admin_log.cc: remove unused variables
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:22 AM Revision 0d27bcc5 (ceph): osd/ReplicatedPG.cc: fix resource leak
- CID 1204294 (#1 of 1): Resource leak (RESOURCE_LEAK)
leaked_storage: Variable filter going out of scope leaks the
s... - 11:22 AM Revision 16cbabab (ceph): osd/PGLog.h: prefer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:22 AM Revision 473b33b0 (ceph): osdc/ObjectCacher.cc: refer ++operator for non-primitive iterators
- Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
- 11:06 AM CephFS Feature #9312 (Resolved): kclient: support signatures in kernel code
- While refactoring authentication docs, Sage pointed out in review comments that "somewhere here we should note that t...
- 10:29 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- "remove_snap_mapped_object":https://github.com/ceph/ceph/blob/master/src/osd/ReplicatedPG.cc#L199 removes the object ...
- 08:03 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- tentative solution to a problem related to what happens although it does not address the main issue https://github.co...
- 07:43 AM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- An idea, which is not a fix but may help. When the following sequence is done:...
- 09:56 AM Cleanup #2458 (Resolved): filestore: backend abstraction
- I believe this has been addressed by last year works on implementing alternatives to the FileStore
- 09:48 AM Cleanup #1960 (Resolved): You should be able to print daemon options without specifying a config ...
- Not sure when or how but I tested ceph-osd --help when no ceph.conf is available and the help shows as expected.
- 09:43 AM rbd Feature #9311 (Resolved): rbd: parallelize import/export
- Add AIO support when importing/exporting images to/from RBD.
- 09:41 AM Documentation #1819 (Resolved): document librados python api
- https://github.com/ceph/ceph/commit/8cb3dad5488874b2af3234ac3bf2c627ba19a5ae
- 09:32 AM RADOS Feature #628: crushtool: better error messages when parsing a crushmap.txt
- The crush compiler indeed has minimal error handling...
- 09:27 AM Feature #610 (Resolved): gui: make PG view prettier
- With http://tracker.ceph.com/projects/calamari :-)
- 09:24 AM RADOS Feature #1474: scrub: repair intra-node inconsistencies
- I believe this is no longer a problem. Is it ?
- 09:04 AM rgw Documentation #9237 (Closed): rgw: Feature comparison document
- 08:53 AM Cleanup #871 (Resolved): replace gnu_cxx::hash_map with tr1::unordered_set
- gnu_cxx::hash_map is not found in today's master
- 08:40 AM RADOS Fix #6109: pg <pgid> mark_unfound_lost fails if a completely-gone OSD still in map
- Is there a known workaround ?
- 06:29 AM CephFS Feature #9284 (In Progress): mds: warn when clients are not responding to cache pressure
- 02:54 AM CephFS Bug #8648 (Fix Under Review): Standby MDS leaks memory over time
- https://github.com/ceph/ceph/pull/2375
- 12:52 AM Revision 8f368c51 (ceph): mailmap: Ashish Chandra affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
- 12:34 AM Revision 5fd50c99 (ceph): mailmap: Boris Ranto affiliation
- Signed-off-by: M Ranga Swami Reddy <swamireddy@gmail.com>
09/01/2014
- 11:54 PM rbd Bug #9078: Removing an RBD is very slow whenever there is write's in other RBD which also belongs...
- Either rbd remove will get any impact on different orders(12-24) or it will be same behaviour ?, because for lesser o...
- 11:20 PM CephFS Feature #9026 (Fix Under Review): client: vxattr support for rctime, rsize, etc.
- https://github.com/ceph/ceph/pull/2373
- 05:18 PM CephFS Bug #9216: mds may regard active clients as stale due to slow pg recovery
- maybe it's the same as #9212
- 04:17 PM Fix #8914: osd crashed at assert ReplicatedBackend::build_push_op
- When scrub "scans objects":https://github.com/ceph/ceph/blob/3215c520e1306f50d0094b5646636c02456c9df4/src/osd/PGBacke...
- 03:38 PM Revision bbf569de (ceph): tasks: fix mount race in mds_client_recovery
- Signed-off-by: John Spray <john.spray@redhat.com>
- 12:25 PM Linux kernel client Bug #8818: IO Hang on raw rbd device - Workqueue: ceph-msgr con_work [libceph]
- Ilya Dryomov wrote:
> Well, the issue didn't get resolved, you simply found a kernel it doesn't show up on. I'm sti... - 09:24 AM Revision 3215c520 (ceph): Merge pull request #2366 from dachary/wip-mailmap
- mailmap: add .peoplemap
Reviewed-by: M Ranga Swami Reddy <swamireddy@gmail.com> - 08:14 AM Revision 5554c7e7 (ceph): Merge pull request #2370 from ceph/wip-9309
- lockdep: increase max locks (1000 -> 2000)
Reviewed-by: Loic Dachary <loic-201408@dachary.org> - 05:43 AM Bug #9224 (Need More Info): osd: segv in dlopen
- grep in the ceph-osd logs from runs august, 23,24,25 found no match for _dl_map_object_deps . I'm unable to find a le...
- 04:05 AM CephFS Bug #9297 (Duplicate): mds: null deref in _session_logged
- Dupe of #9173: the fix hadn't landed at the point these tests were run.
- 12:53 AM rgw Bug #9310 (Duplicate): Rados gateway swift subuser creation problem on FIREFLY
I have two problems both related to Rados gateway swift user creation on FIREFLY
Ceph version 0.80.5
Centos ...- 12:46 AM Revision a5b4c583 (ceph): lockdep: increase max locks (1000 -> 2000)
- We hit the 1000 lock limit on
ubuntu@teuthology:/a/teuthology-2014-08-31_02:30:01-rados-next-testing-basic-multi/46... - 12:32 AM Revision 41914ec3 (ceph): Merge pull request #2369 from dachary/wip-mark-unfound-lost
- documentation: add the mark_unfound_lost delete option
Reviewed-by: Sage Weil <sage@redhat.com> - 12:31 AM Revision 89ca4f6f (ceph): Merge pull request #2368 from dachary/wip-pg-const
- osd: MissingLoc::get_all_missing is const
Reviewed-by: Sage Weil <sage@redhat.com> - 12:30 AM Revision bf0a76ff (ceph): Merge pull request #2367 from dachary/wip-early-fail
- tests: qa/workunits/cephtool/test.sh early fail
Reviewed-by: Sage Weil <sage@redhat.com>
Also available in: Atom