Activity
From 12/24/2013 to 01/22/2014
01/22/2014
- 10:52 PM Revision 2b0f8cbf (ceph): Merge branch 'wip-client-io-final'
- Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: David Zafman <david.zafman@inktank.com> - 10:39 PM Revision 167f8523 (ceph): RadosModel: RollBack dirties the object
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 7e3ccfef (ceph): ReplicatedPG: take ReplicatedPGRef in BlessedContext
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 96d4a0c2 (ceph): PG::calc_acting: consider newest_update_osd when choosing backfill peers
- We must include newest_update_osd->second.log_tail when considering backfill
peers because in GetLog we will request ... - 10:39 PM Revision e73a9f7b (ceph): ReplicatedPG: make STARTSYNC a noop
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 4fc4573f (ceph): PG: drop messages from down peers
- This overlaps with the existing old_peering_msg() mechanism
except in one case: pulls from a replica not in the actin... - 10:39 PM Revision ecbdeb19 (ceph): PGBackend: clarify rollback_stash name and method comments
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 0bba79b7 (ceph): ReplicatedPG: support async reads on ec pools
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision a382ab82 (ceph): ReplicatedPG: requeue in_progress_async_reads on_change
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision b3c2ef2d (ceph): ReplicatedPG: make COPY_GET async read friendly
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision f2384412 (ceph): ReplicatedPG: EOPNOTSUPP on ec_pool if there are omaps in the COPY_GET
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision eb8f66d5 (ceph): ReplicatedPG: handle removing the old object in finish_copy_op
- do_osd_ops will need to either copy the old version out of the
way or simply delete it depending on mod_desc. Thus, ... - 10:39 PM Revision e8a08ffa (ceph): ReplicatedPG::finish_copyfrom: check new_obs.exists, not obs.exists
- If the transaction previously created the object, the stat update
will have already happened.
Signed-off-by: Samuel ... - 10:39 PM Revision 7c9425c3 (ceph): ReplicatedPG: fill in mod_desc in finish_copyfrom
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 9fd7a800 (ceph): ReplicatedPG: use append for write and writefull
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 183f41c0 (ceph): ReplicatedPG: fill in mod_desc for hitset repops
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 9273177a (ceph): ReplicatedPG::_process_copy_chunk: use append
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision d0eda792 (ceph): common/hobject: print shard as unsigned, not as char
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision ca37d08e (ceph): ReplicatedPG::trim_object: remove unused int r; declaration
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 6e74123c (ceph): RadosModel: undirty with update_object_undirty
- Otherwise it doesn't behave properly with snapshots since
update_object_version doesn't add another snapshot entry.
... - 10:39 PM Revision 155b4b06 (ceph): PGBackend,ReplicatedBackend: add support for rolling back log entries
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision ecae37c3 (ceph): osd/: introduce xattr caching for ec pools
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision c98cb1ce (ceph): ReplicatedPG: Fill in rollback info for log events
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 25eced59 (ceph): PGLog: create interface allowing interface user to cleanup/rollback
- We need to be able to allow the PGLog interface user to provide
logic for rolling back and trimming log entries. To ... - 10:39 PM Revision 347d968a (ceph): PGLog: we cannot rollback past a recovered object entry
- We don't recover the older versions, so we can't do a rollback.
In general, this is ok since we had enough copies to ... - 10:39 PM Revision a3e4a844 (ceph): PGLog,PG: reset can_rollback_to when we get a log entry without a trans...
- We cannot rollback past an entry which we didn't actually
apply locally.
Signed-off-by: Samuel Just <sam.just@inktan... - 10:39 PM Revision 66bc91cf (ceph): ObjectStore: _collection_move_rename needs to increment ops
- Bugfix.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:39 PM Revision 2fe5e1f2 (ceph): osd/: add sync/async read methods to PGBackend
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 3ef6282c (ceph): ReplicatedPG: use pgbackend->objects_read_sync() as appropriate
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 647e75ae (ceph): src/: add CEPH_OSD_OP_READ_SYNC
- EC pools won't support SYNC reads. This means that TMAP and friends
along with reads from an object class will not b... - 10:39 PM Revision c60185c0 (ceph): PGBackend: add PGTransaction
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision bf9390f3 (ceph): hobject_t/ReplicatedPG: tempness is now an hobject thing
- PGBackend implmentations will have complete control over the temp
collection. Rather than specifying the collection ... - 10:39 PM Revision 5732f0c9 (ceph): ReplicatedBackend: implement RPGTransaction
- RPGTransaction is essentially a wrapped ObjectStore::Transaction.
The coll_t argument is elided, tempness is instead ... - 10:39 PM Revision 09717e6d (ceph): PG/ReplicatedPG: move snap map management into append_log
- Both the replica and the primary should use the same append_log
mechanism to update the snap map.
Signed-off-by: Sam... - 10:39 PM Revision 26b804f2 (ceph): ReplicatedPG: split out op_applied_replica
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 1a4694fd (ceph): ReplicatedPG: factor out should_send_op
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision c58c8990 (ceph): PGBackend: allow queue_transaction to specify an op
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 53c26bc3 (ceph): osd/: move client op handling into ReplicatedBackend
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision a7a6a2f1 (ceph): osd_types: ObjectModDesc
- Contains information needed to roll back a pg_log_entry_t.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:39 PM Revision 63f64de1 (ceph): osd_types: add ObjectModDesc into pg_log_entry_t
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 05289745 (ceph): osd_types: add a can_rollback_to line to pg_log_t
- This allows us to determine how far back the log can be rolled.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:39 PM Revision 1b5992b0 (ceph): FileStore::lfn_link: handle link within a collection
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 1b0335ed (ceph): common: add option to disable filestore sharded object check
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 03c5344f (ceph): PGLog: allow for empty pg log in update_range if log_tail == eversion_t()
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 8494aa84 (ceph): src/test/osd: remove tmap test
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 69ad880e (ceph): test/osd/Object: create encoders for ContDesc
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 7fe22da2 (ceph): test/osd: restructure Object/RadosModel in prep for append
- Attribute handling no longer has special support in ContentsGenerator.
The most recent operation information is now s... - 10:39 PM Revision ca91fb4c (ceph): test/osd: adapt RadosModel to not do omap ops on ec_pools
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision d0933eef (ceph): test/osd: add append support to WriteOp
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 0885880b (ceph): test/osd/TestRados.cc: skip 0 weight ops
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 4d56e3c8 (ceph): test/osd/: verify length of buffer returned in read
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision d3c8e85a (ceph): ReplicatedBackend: update the collection_list mechanisms to skip stashe...
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:39 PM Revision 393a035b (ceph): OSDMonitor: add debug_fake_ec_pool
- This flag will cause ReplicatedPG to act as though the
pool were actually an EC pool in that operations will
be restr... - 10:39 PM Revision 6d05fa23 (ceph): PG: prefer older last_update for ec_pool
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision a1ee1ee7 (ceph): PGLog::rewind_divergent_log: we can rewind to tail, fix assert
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision c03d027c (ceph): hobject: admit that gen_t is actually version_t
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision 518774dc (ceph): ObjectStore: improve name of the queue_transaction which cleans up the ...
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision 06f7a98a (ceph): ReplicatedBackend.h: don't need to be active for pushes
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision b7d100b6 (ceph): FileStore::_collection_move_rename: remove source before closing guard
- Fixes a bug in _collection_move_rename replays.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 10:38 PM Revision 1f9457c5 (ceph): ReplicatedBackend: remove pushing entry for each object as it completes
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:38 PM Revision a2006507 (ceph): PGLog: don't move up log.tail
- Moving up log.tail unnecessarily risks backfilling
a replica after a split. Also, it disrupts the
property that repl... - 09:44 PM Revision d6b592c0 (ceph): Merge pull request #1079 from ceph/wip-tier-bits
- osd: a few tiering fixes
Reviewed-by: Greg Farnum <greg@inktank.com> - 09:10 PM Bug #7208 (Resolved): CEPH_FEATURE_CRUSH_V2 feature mismatch
- In attempting to test kernel rbd with 0.74-553-g54d51d8-1saucy on kernel 3.11, I'm seeing:
feature set mismatch, m... - 07:39 PM Bug #7207 (Resolved): Lock contention at filestore I/O (FileStore::lfn_open) during filestore fol...
- There is a lock contention happening at filestore I/O (FileStore::lfn_open, https://github.com/guangyy/ceph/blob/mast...
- 06:48 PM CephFS Bug #7206: Ceph MDS Hang on hadoop workloads
- just enable admin socket (without enabling client debug and mds debug). When the hang occurs, dump client request and...
- 06:18 PM CephFS Bug #7206: Ceph MDS Hang on hadoop workloads
- Greg Farnum already knows this, but for reference
I spent a large part of today with debug logging on to try to ca... - 02:01 PM CephFS Bug #7206: Ceph MDS Hang on hadoop workloads
- That should be enough to either diagnose it or realize we need to reproduce it locally.
- 01:58 PM CephFS Bug #7206: Ceph MDS Hang on hadoop workloads
- Greg Farnum wrote:
> This could be an issue with time sync on the nodes; check your clock drift. (That's the only is... - 01:26 PM CephFS Bug #7206: Ceph MDS Hang on hadoop workloads
- This could be an issue with time sync on the nodes; check your clock drift. (That's the only issue I know of that we'...
- 12:30 PM CephFS Bug #7206 (Can't reproduce): Ceph MDS Hang on hadoop workloads
- With blatant disregard for posted advice about the stability of ceph-fs, I attempted to use it for testing out some h...
- 05:40 PM Revision 022e531b (ceph): added roles for bucket quota tests
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 04:33 PM Tasks #6185: expand upgrade tests to be able to test downgrade
- I don't think we want to do this until we decide that supporting downgrade is a worthwhile investment.
- 04:33 PM RADOS Feature #6436: erasure-code: Downgrade testing
- We don't support daemon downgrade in general, so let's not try and make that a requirement for testing here!
- 04:20 PM Revision 8eaef414 (ceph): Merge pull request #1126 from ceph/wip-osdmapenc-fix
- MOSDMap: reencode maps if target doesn't have OSDMAP_ENC
Reviewed-by: Sage Weil <sage@inktank.com> - 03:41 PM Bug #7186: ceph-rest-api: /tell/<osdid-or-pgid>/query doesn't work
- The help information is a little confusing here, "query" only works for placement group while "version" only works fo...
- 03:34 PM Revision 8060afd7 (ceph): MOSDMap: reencode maps if target doesn't have OSDMAP_ENC
- Reencode both full and incremental maps if target doesn't know how to
decode OSDMAP_ENC maps (CEPH_FEATURE_OSDMAP_ENC... - 02:44 PM Feature #5996 (Resolved): EC: [link] PG::calc_acting and friends should always choose the shortes...
- 02:44 PM Feature #5990 (Resolved): EC: [link] Factor out the ReplciatedPG object replication and client wr...
- 02:44 PM Feature #6832 (Resolved): EC: Adapt pg log to include information necessary for rollback
- 02:44 PM Feature #6831 (Resolved): EC: Adapt ReplicatedPG read path to handle async reads
- 12:50 PM Bug #7185: ceph-rest-api inconsistent routing with show_human_help
- /tell is a command itself:...
- 12:26 PM Revision 9792500a (ceph): doc: add-or-rm-mons.rst: better explanation on number of monitors
- Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
- 10:39 AM RADOS Feature #7205 (New): erasure-code: update HITSET_GET to somehow use a CEPH_OSD_OP_READ op
- Or, re-express using async read/sync read.
- 09:45 AM Revision 9eac5e37 (ceph): Added update_metadata, clear_metadata, and usage description code
- 09:38 AM Bug #7203 (Can't reproduce): osd/ReplicatedPG.cc: 9551: FAILED assert(r >= 0) in scan_range
- on next:...
- 09:36 AM Subtask #7158: EC: flesh out how the ceph tool should be used to manage ec pools and create sub t...
- i think pool creation is all there. the create-pyramid rule we can treat separately, i think!
but... should we ha... - 02:46 AM Subtask #7158: EC: flesh out how the ceph tool should be used to manage ec pools and create sub t...
- Loic pointed out that there's already support to create pools via the ceph tool
https://github.com/ceph/ceph/blob/... - 09:32 AM rgw Bug #7169 (In Progress): rgw: list multipart parts broken (> 1000 parts)
- 09:31 AM Bug #7184 (Resolved): osd: scrub complains about num_objects_dirty for old pools
- 03:08 AM Revision e4abe40c (ceph): added roles to rest yaml
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 02:51 AM Revision dcca4137 (ceph): added perl script for rgw bucket quota tests
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 02:50 AM Revision 28c75f81 (ceph): removing rgw_tests.sh
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
(cherry picked from commit 54caa0192b02b03549fe4ca5d062495e6e429... - 02:44 AM Revision 54caa019 (ceph): removing rgw_tests.sh
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 02:35 AM Revision 98c78736 (ceph): modified rgw suite for bucket quota tests
- Signed-off-by: tamil <tamil@tamil-VirtualBox.(none)>
- 01:50 AM Revision 4550fb5f (ceph): Merge branch 'master' of github.com:ceph/ceph
- 01:50 AM Revision 8b8ede7a (ceph): modified the port to 7280 in the script instead of the default 80
- Signed-off-by: tamil <tamil@tamil-VirtualBox.(none)>
- 01:30 AM Revision 81ff12ec (ceph): Merge pull request #1124 from ceph/wip-cache-mode-fix
- mon/MonCommands: 'invalidate+forward' -> 'forward'
Reviewed-by: Sage Weil <sage@inktank.com>
01/21/2014
- 11:56 PM Revision 850b4f42 (ceph): script for rgw bucket quota testing
- Signed-off-by: tamil <tamil@tamil-VirtualBox.(none)>
- 10:27 PM Revision 339bed1b (ceph): mon/MonCommands: 'invalidate+forward' -> 'forward'
- Commit 4e439857a694 introduced invalidate+forward cache mode, commit
81279e3bb6e0 renamed it to forward, but missed t... - 10:18 PM Revision b88c8eae (ceph): doc: Added Documenting Ceph (draft) document.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:18 PM Revision 90aea596 (ceph): doc: Added documenting Ceph to the toc tree.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:05 PM devops Feature #6986 (In Progress): Build packages containing RBD and CephFS kernel modules for installa...
- I'm building against kernel-3.10.0-54.0.1.el7.x86_64, and I get the following error when using df5f7b6abd0dacf6fa60ac...
- 12:39 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- 08:16 PM Revision 9fa8d2d1 (ceph): mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
- There is one path where a mds that is not sending its beacon (e.g.,
because it is not running at all) will lead to pr... - 08:16 PM Revision e8298515 (ceph): mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
- There is one path where a mds that is not sending its beacon (e.g.,
because it is not running at all) will lead to pr... - 08:12 PM Revision 3d88dfff (ceph): mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
- There is one path where a mds that is not sending its beacon (e.g.,
because it is not running at all) will lead to pr... - 08:04 PM Revision 291407bc (ceph): Merge pull request #1123 from ceph/wip-stray-mdsmaps
- mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
Reviewed-by: Greg Farnum <greg@inktank.com> - 07:51 PM Revision ed2decb1 (ceph): doc: Fixed end string from '' to `` to stop warnings. Enforced 80 char ...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:29 PM Revision 584c2dd6 (ceph): mon/MDSMonitor: do not generate mdsmaps from already-laggy mds
- There is one path where a mds that is not sending its beacon (e.g.,
because it is not running at all) will lead to pr... - 06:34 PM Revision cd217ec0 (ceph): Merge pull request #1119 from ceph/wip-7184
- osd: ignore num_objects_dirty for old pools
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 03:44 PM devops Bug #7157: ceph-disk list fails in encrypted disk setup
- I'm having this problem, with a fresh install of Ubuntu 12.04 and the latest (as of yesterday) Ceph release (ceph ver...
- 03:40 PM Revision 4c968534 (ceph): packaging: ship libdir/ceph
- Automake puts ceph_common.sh into libdir/ceph, but the Red Hat packaging
was not capturing this file.
Add the libdir... - 01:41 PM Feature #5999 (In Progress): EC: [link] OSD internals must work in terms of cpg_t
- 01:38 PM Feature #6437 (In Progress): EC: implement ECBackend recovery
- 01:37 PM Feature #7167 (Resolved): Add op_process_latency in perf counters
- 01:31 PM Feature #5990 (Fix Under Review): EC: [link] Factor out the ReplciatedPG object replication and c...
- 01:28 PM Feature #6832 (Fix Under Review): EC: Adapt pg log to include information necessary for rollback
- 01:28 PM Feature #6831 (Fix Under Review): EC: Adapt ReplicatedPG read path to handle async reads
- 01:27 PM Feature #5996 (Fix Under Review): EC: [link] PG::calc_acting and friends should always choose the...
- 01:27 PM Feature #6438 (Fix Under Review): EC: implement EC client IO
- 12:39 PM devops Feature #7047: rhel7: build process for rbd.ko, ceph.ko kernel modules
- 12:39 PM devops Feature #7046: rhel7: gitbuilder for rbd.ko, ceph.ko kernel modules
- 12:20 PM Feature #7196 (New): qa: test encoding semantics, not just being able to decode/encode different ...
- I have a branch with at least slightly better detection by comparing the json dumps. Need to generate new object corp...
- 12:16 PM Feature #7195 (Resolved): librados: complete C write object operation bindings
- They're missing individual return values, exec, omap (covered by #7194), and possibly a few other features from the c...
- 12:14 PM Feature #7194 (Resolved): librados: omap in the c api
- It doesn't exist, but should. Add it to the read/write object operations bindings.
- 12:13 PM Feature #7193 (Resolved): librados: object operation bindings for reads
- Writes were added recently. It would be useful to have these available for binding to other languages.
- 11:50 AM Revision 6bee1885 (ceph): Docs: Add backfill_toofull to list of PG states
- Signed-off-by: John Spray <john.spray@inktank.com>
- 11:21 AM devops Documentation #2974 (Resolved): doc: update chef docs for mon key distribution
- 10:57 AM Revision 199e614b (ceph): Initial version of the source code. Statistics extraction is done
- 09:45 AM Feature #7192 (Resolved): An easier-to-process health report
Currently, the "ceph health" is great for human consumption, but a bit awkward to feed into a monitoring app, becau...- 05:34 AM Revision 7a2d7f13 (ceph): Merge pull request #1120 from xylv/wip-op-process-latency-squashed
- osd: Add op_process_latency in perf counter
Reviewed-by: Sage Weil <sage@inktank.com> - 03:11 AM Revision 78150bd9 (ceph): osd: Add op_process_latency in perf counter
- Fixes: #7167
Signed-off-by: Ray Lv <xiangyulv@gmail.com>
01/20/2014
- 10:50 PM Revision 67d8b72e (ceph): common: fix bufferlist::append(istream) test
- bufferlist::append(istream) now filters out empty lines; reflect this in
the test
Signed-off-by: Loic Dachary <loic@... - 10:20 PM Feature #7191 (Closed): Replace Mutex to RWLock with fdcache_lock in FileStore
- done by https://github.com/somnathr/ceph/commits/wip-queueing-sd1.
- 10:09 PM Feature #7191 (Closed): Replace Mutex to RWLock with fdcache_lock in FileStore
- Now FileStore can't make full use of SSD because of poor parallelism, 4 OSD in the same SSD are recommended to achiev...
- 09:45 PM Bug #7184 (Fix Under Review): osd: scrub complains about num_objects_dirty for old pools
- 08:11 PM Revision 403be003 (ceph): Fix #7187: Include all summary items in JSON health output
- Signed-off-by: John Spray <john.spray@inktank.com>
(cherry picked from commit fdf3b5520d150f14d90bdfc569b70c07b0579b38) - 08:11 PM Revision a2e34b8d (ceph): Fix #7187: Include all summary items in JSON health output
- Signed-off-by: John Spray <john.spray@inktank.com>
(cherry picked from commit fdf3b5520d150f14d90bdfc569b70c07b0579b38) - 08:09 PM Feature #7167: Add op_process_latency in perf counters
- Squashed commits to a new pull request: https://github.com/ceph/ceph/pull/1120
- 04:53 AM Feature #7167: Add op_process_latency in perf counters
- Seems I'm spamming the backlog of pull requests. Here is a new pull request including changes after your comments: h...
- 08:07 PM Revision 368f1736 (ceph): Merge pull request #1117 from jcsp/wip-7187
- Fix #7187: Include all summary items in JSON health output
Backport: emperor, dumpling
Reviewed-by: Sage Weil <sage@... - 12:11 PM Fix #7187 (Resolved): In JSON "ceph health" output, 'summary' list only contains one item
- 03:42 AM Fix #7187: In JSON "ceph health" output, 'summary' list only contains one item
- https://github.com/ceph/ceph/pull/1117
- 03:07 AM Fix #7187 (Resolved): In JSON "ceph health" output, 'summary' list only contains one item
The human readable output gives you a list of items, but the JSON output for some reason just picks the first one o...- 11:10 AM Revision fdf3b552 (ceph): Fix #7187: Include all summary items in JSON health output
- Signed-off-by: John Spray <john.spray@inktank.com>
- 11:07 AM Bug #7190 (Resolved): osd/OSD.cc: 2338: FAILED assert(oldmap->have_pg_pool(pgid.pool()))
- ...
- 08:51 AM Bug #7122 (Resolved): push 0/hit_set_... v 0'0 failed because local copy is 818'7131
- 08:51 AM Bug #7117 (Resolved): centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_comm...
- 08:24 AM Bug #7188 (Won't Fix): Admin socket files are lost on log rotation calling initctl reload (ubuntu...
- h3. Workaround
Using *restart* instead of *reload* restarts the daemons instead of sending them a signal that grac... - 07:15 AM rgw Bug #6951: rgw: bucket re-create fails when cache is disabled
- I have the same problem (kind of) as described in the mailing list.
But it's not stopped by the cache, AFAICT, the c...
01/19/2014
- 08:00 PM Revision 63a77f46 (ceph): Merge pull request #1112 from kazhang/wip-7180
- Fix #7180 in ceph_rest_api
Reviewed-by: Sage Weil <sage@inktank.com> - 07:50 PM Bug #7180 (Resolved): 500 error when get mds help via ceph-rest-api
- Landed to master 63a77f465654261e13cf7fbd5ffc6ed0c9f2eca0
- 06:03 PM Revision c78c3002 (ceph): Remove some almost-duplicate COMMAND definitions
- These seem to have been added twice in 8985e1c9e and
then in 7b42deef3. The second version looks more
correct.
Sign... - 08:22 AM Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap
- Just noticed there is a rest-api category (I was searching for ceph-rest-api..). Could someone flag this properly ? I...
- 08:21 AM Bug #7186: ceph-rest-api: /tell/<osdid-or-pgid>/query doesn't work
- Just noticed there is a rest-api category (I was searching for ceph-rest-api..). Could someone flag this properly ? I...
- 08:18 AM Bug #7186 (Won't Fix): ceph-rest-api: /tell/<osdid-or-pgid>/query doesn't work
- tell/<osdid-or-pgid>/query is shown twice in show_human_help (probably part of the problem?)
For instance, /tell/1... - 08:21 AM Bug #7185: ceph-rest-api inconsistent routing with show_human_help
- Just noticed there is a rest-api category (I was searching for ceph-rest-api..). Could someone flag this properly ? I...
- 07:58 AM Bug #7185 (Won't Fix): ceph-rest-api inconsistent routing with show_human_help
- This works:
http://apiserver:5000/api/v0.1/osd <-- Returns help for /osd calls
http://apiserver:5000/api/v0.1/osd/... - 07:58 AM Revision 20cf41d6 (ceph): Missing a key for perm 'w' in permmap (src/pybin/ceph_rest_api.py:277)
- It leads to a 500 error when getting mds help info via rest api.
Changed "w" to "rw" in MonCommands.h
Fixes: #7180
S... - 05:48 AM Revision 26a534e6 (ceph): Merge branch 'bclibcoop/next-cors' of https://github.com/BCLibCoop/ceph...
- Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
- 05:30 AM Revision 6dcf462c (ceph): rgw: Document fields for access/secret key
- The field name mapping for access vs secret key is not clear, this
helped in debugging.
Signed-off-by: Robin H. John... - 05:30 AM Revision 52a91540 (ceph): rgw: Fix signature variable naming/failure print
- The signature variables for expected vs got are poorly named, and this
lead them being swapped in the signature valid... - 05:30 AM Revision 586ad1f6 (ceph): rgw: Use correct secret key for POST authn
- The POST authentication by signature validation looked up a user based
on the access key, then used the first secret ... - 05:19 AM Revision 7314cde1 (ceph): osd: ignore num_objects_dirty for old pools
- Way back in a0ed9c20048750fd4b2c7ce0339fa8b20ef08ca3 we introduced the
dirty flag, but we did not track it in the sta... - 05:06 AM Revision 8ef7268a (ceph): Merge pull request #1111 from ceph/wip-osdmap-7177
- OSDMap: Populate primary_temp values a little more carefully
Reviewed-by: Sage Weil <sage@inktank.com>
01/18/2014
- 10:03 PM Revision 1fcbddc8 (ceph): libc++: avoid hash re-definitions
- The definitions of hash<> for int64_t/uint64_t that were not available
on i386 in the __gnu_cxx namespace are availab... - 10:03 PM Revision 1b308b65 (ceph): libc++: fix null pointer comparison
- This error is thrown when comparing a shared_ptr to NULL. To resolve
this we just use shared_ptr::operator bool that ... - 10:03 PM Revision aae47009 (ceph): libc++: use ceph::shared_ptr in installed header
- librados.hpp uses std::tr1::shared_ptr which may not be available such
as in libc++. This switches the use to ceph::s... - 10:03 PM Revision 4c4e1d0d (ceph): libc++: use ceph:: namespaced data types
- Switches the implemetnation of smart pointers and unordered map/set to
use the ceph:: versions.
Signed-off-by: Noah ... - 10:03 PM Revision 8e867204 (ceph): libc++: create portable smart ptr / hash_map/set
- Creates ceph:: namespaced versions of smart pointers and unordered map
and sets.
Signed-off-by: Noah Watkins <noahwa... - 09:51 PM rgw Bug #7183 (Pending Backport): RGW CORS POST fix for subusers and cleanups
- 09:31 PM rgw Bug #7183: RGW CORS POST fix for subusers and cleanups
- Just to be clear, the main issue that is fixed is POST request might not authenticate when user has multiple access k...
- 09:08 PM rgw Bug #7183 (Resolved): RGW CORS POST fix for subusers and cleanups
- Patches already emailed to ceph-devel list at vger.
Patch #3 is the important one: looking up the correct secret k... - 09:22 PM Bug #7184 (Resolved): osd: scrub complains about num_objects_dirty for old pools
- 09:21 PM Bug #7177 (Resolved): osd crashed at OSD::advance_pg on master branch
- 12:52 PM Bug #7180: 500 error when get mds help via ceph-rest-api
- Need review: https://github.com/ceph/ceph/pull/1112
- 12:13 PM Bug #7180 (Resolved): 500 error when get mds help via ceph-rest-api
- I was playing with ceph-rest-api, when I try to access http://localhost:5000/api/v0.1/mds, it's supposed to return th...
- 12:51 PM devops Documentation #7182 (Resolved): documentation miscategorizes debian as upstart based. debian use...
- The ceph documentation miscategorizes debian as an upstart based distro. Debian uses systemvinit. An example of suc...
- 12:47 PM devops Bug #7181 (Rejected): debian 7 wheezy init.d script will not start OSDs not corresponding to a mo...
- I'm in the midst of testing a ceph deployment, and am not using individual disks for ceph. They're on my main OS dri...
- 05:04 AM Revision 7e7eda47 (ceph): OSDMap: Populate primary_temp values a little more carefully
- In _get_temp_osds(), we populate temp_pg from the list in the OSDMap,
but we also skip anybody in the list who's down... - 12:58 AM Revision 35e5b128 (ceph): suite for rest
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 12:08 AM CephFS Documentation #3672: doc: how to mount ceph-fuse from fstab
- On "Debian":http://debian.org only the following (sample) works with *ceph-fuse*-0.72.2-1:
/etc/fstab:
moun...
01/17/2014
- 11:57 PM Bug #7178 (Resolved): sample.ceph.conf needs update
- At the moment "sample.ceph.conf" is incomplete (some essential options are missing);
it uses old deprecated options ... - 11:56 PM Revision 47bc71a7 (ceph): fixed the syntax for test maxmds=2
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 11:41 PM Revision a13ebd37 (ceph): fix for the test maxmds=2
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 09:33 PM Revision ad203d59 (ceph): doc: Fixed ruleset typo.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:17 PM Revision 6a19daa2 (ceph): Merge pull request #1106 from ceph/wip-rgw-contention
- Wip rgw contention
Reviewed-by: Sage Weil <sage@inktank.com> - 08:52 PM Revision 6b7f27c0 (ceph): librados: use rwlock for lookup pool, cache results
- Use rwlock to read cahed results, only use regular lock if
results are not cached.
Invalidate cache on new osdmap.
S... - 08:34 PM Revision 4b31456b (ceph): rgw: add .h file to Makefile rule
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 07:59 PM Revision 7f61d249 (ceph): Merge pull request #1082 from dachary/wip-ceph-disk
- ceph-disk: larger timeout in the test script
Reviewed-by: Sage Weil <sage@inktank.com> - 07:47 PM Revision be5ac2b7 (ceph): Merge pull request #1088 from christian-marie/wip-c-atomic-writes
- librados: Add C API coverage for atomic write operations
Reviewed-by: Sage Weil <sage@inktank.com> - 07:23 PM Revision d1de32cf (ceph): doc: Added default-placement to list of placement targets. Added SSL co...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:22 PM Revision 80212eaa (ceph): doc: Added additional comment about port number.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:24 PM Revision 184e64d0 (ceph): Merge pull request #1092 from ceph/wip-rgw-loadgen
- Wip rgw loadgen
Reviewed-by: Sage Weil <sage@inktank.com> - 06:16 PM Revision 02153421 (ceph): rgw: loadgen, configurable num of objs, buckets
- also fix request draining
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 06:14 PM Revision 7c867645 (ceph): rgw: loadgen shutdown, error out on failures
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 06:14 PM Revision e8a4b307 (ceph): rgw: loadgen frontend read uid, init access key
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 06:14 PM Revision 85267cf6 (ceph): rgw: sign loadgen requests
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 06:14 PM Revision 57137cb4 (ceph): rgw: add a load generation frontend
- Still missing some pieces, but already generates requests.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 05:58 PM Revision b04d26ae (ceph): Merge pull request #1107 from xylv/wip-osd-crash-with-leveldb-cache-set
- osd: OSD failed to start with osd_leveldb_cache_size > 0
Reviewed-by: Sage Weil <sage@inktank.com> - 05:36 PM Revision 65d33484 (ceph): Merge pull request #1109 from ceph/wip-rbd-mount_timeout
- rbd: expose mount_timeout map option
Reviewed-by: Sage Weil <sage@inktank.com> - 05:31 PM Bug #7177: osd crashed at OSD::advance_pg on master branch
- Unfortunately I can't make much sense of the core dump.
Fortunately just by looking at the code involved I'm pretty ... - 04:50 PM Bug #7177 (Resolved): osd crashed at OSD::advance_pg on master branch
- logs: ubuntu@mira057.front.sepia.ceph.com:/home/ubuntu/bug_osd_crash
hit this issue when trying to run the rest/te... - 04:05 PM Revision e8bb1654 (ceph): call wait() on the teuthology-results Popen object
- This ought to fix the issue where zombie teuthology-results processes
stick around.
Signed-off-by: Zack Cerza <zack.... - 12:42 PM Revision f01202de (ceph): Fixes: #7172
- To ensure LevelDBStore::db is destroyed before dependent LevelDBStore::db_cache and LevelDBStore::filterpolicy.
Sign... - 09:53 AM Revision b1a853e4 (ceph): rbd: expose mount_timeout map option
- Expose mount_timeout map option. (I missed it in commit 9b7364d2450c,
which added -o / --options option and among ot... - 09:31 AM Feature #7173 (Resolved): Pre-populate exceptions
- Pre-populate exceptions on the monitor after a large map change to avoid message storms in large clusters.
- 08:49 AM Subtask #7146: implement osd crush rule create-erasure
- Hi Sage
What if **osd crush rule create-pyramid** was something like:
* get the properties of a designated eras... - 02:58 AM Subtask #7146: implement osd crush rule create-erasure
- ...
- 08:37 AM rgw Feature #6747: PowerDNS backend for RGW bucket directing
- You can fetch the region map with GET /admin/config
- 01:08 AM rgw Feature #6747: PowerDNS backend for RGW bucket directing
- This is now in development in the wip-rgw-powerdns branch.
Currently lacking support from the RGW Admin API to fet... - 08:23 AM Feature #6835 (Resolved): EC: ec pgs will need to be able to specify temp primaries other than ac...
- Merged in commit:3b696f28d6cee1fd3a2b78911cc8377445158ba1
- 08:22 AM Linux kernel client Feature #6991: support CEPH_FEATURE_OSDMAP_ENC
- We did not coalesce the feature flags.
- 06:44 AM Revision 8ec7fa85 (ceph): PendingReleaseNotes: note ceph -s fix
- See 790dda9c0b75fa5a59ce6187b6131dad54d6d3fe.
Signed-off-by: Sage Weil <sage@inktank.com> - 06:42 AM Revision d6b2f87e (ceph): Merge pull request #1104 from concubidated/master
- osd: OSDMap: fix output from ceph status --format=json for num_in_osds
Reviewed-by: Sage Weil <sage@inktank.com> - 06:33 AM Revision f29ffa28 (ceph): Merge pull request #1097 from ceph/fix-configure-rerunning
- avoid re-running configure after initial 'make' invocation
This appears to pass the gitbuilders (and the multiple ru... - 05:30 AM Feature #7167: Add op_process_latency in perf counters
- I have sent another clean pull request: https://github.com/ceph/ceph/pull/1108
Please help to check and merge it. - 05:26 AM Linux kernel client Bug #7139 (Fix Under Review): dead tcp connections during connection negotiation may lead to hangs
- wip-7139; https://github.com/ceph/ceph-client/commits/wip-7139
- 05:24 AM Linux kernel client Bug #7139: dead tcp connections during connection negotiation may lead to hangs
- Reproduced it a couple days ago, kernel client had a half-open socket:...
- 05:09 AM Bug #5804: mon: binds to 0.0.0.0:6800something port
- I saw this happening multiple times during a training class.
During the CEPH 100 training we use the machine "dais... - 04:48 AM Bug #7172: osd: OSD failed to start with osd_leveldb_cache_size > 0
- I have sent a pull request to fix this bug: https://github.com/ceph/ceph/pull/1107
Please help to review it. - 04:40 AM Bug #7172: osd: OSD failed to start with osd_leveldb_cache_size > 0
- As shown in the following stack trace, the OSD process crashed while trying to
umount filestore after upgrading file... - 04:38 AM Bug #7172 (Resolved): osd: OSD failed to start with osd_leveldb_cache_size > 0
- After applying osd_leveldb_cache_size = 52428800 on an OSD host, 2 ~ 3 OSDs on
the host failed to start due to the f... - 04:35 AM Revision f6149105 (ceph): Merge pull request #1103 from ceph/wip-7073
- rgw: use configurable prime number for gc hash
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 01:02 AM rgw Feature #6748 (Resolved): Return bucket name in response header
- Merged into master as of yesterday, resolved.
- 01:00 AM Revision 6e931326 (ceph): doc: Removed Calxeda example.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:57 AM Revision dd4a9e18 (ceph): doc: Removed saucy salamander reference.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:54 AM Revision e7d28a72 (ceph): doc: Added host resolution commentary.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:51 AM Revision 22e9a756 (ceph): doc: Added comments from #6682.
- fixes: #6682
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 12:48 AM Revision 6e37b7a8 (ceph): doc: Added modification to quick start.
- fixes: #6682
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 12:11 AM Revision 742d425f (ceph): doc: Changed file name.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:11 AM Revision f7f9bedf (ceph): doc: Added source file for graphic.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:10 AM Revision 26aa6021 (ceph): doc: Added source file for graphic.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:10 AM Revision c30b1387 (ceph): doc: Added third monitor icon.
- fixes: #6959
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 12:09 AM Revision add59b83 (ceph): doc: Added additional monitor icon.
- fixes: #6959
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 12:09 AM Revision 4b03326b (ceph): doc: Removed old file. Changed file name.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
01/16/2014
- 10:30 PM Revision 86c15480 (ceph): rgw: handle racing object puts when object doesn't exist
- If the object didn't exist before and now we have multiple puts coming
in concurrently, we need to make sure that we ... - 10:30 PM Revision 5c24a7ea (ceph): rgw: don't return -ENOENT in put_obj_meta()
- Fixes: #7168
An object put may race with the same object's delete. In this case just
ignore the error, same behavior... - 09:32 PM Revision a84cf15f (ceph): rgw: use rwlock for cache
- Use a window for cache lru updates, so that we don't need to get the
writer lock every read.
Signed-off-by: Yehuda S... - 09:00 PM Revision 1ddc3fdb (ceph): Merge pull request #176 from ceph/wip-7164
- lock.py: request only rsa keys from ssh-keyscan
- 08:51 PM Revision 381d4aa8 (ceph): lock.py: request only rsa keys from ssh-keyscan
- New versions of ssh-keyscan return two keytypes if possible; this
breaks the comparison of "number of lines of output... - 08:36 PM rgw Bug #7073 (Resolved): "rgw gc max objs" should have a prime number as default value
- commit:5ed7865ae8880f47c4715d7db4f069f60544128f
- 08:12 PM Revision 790dda9c (ceph): osd: OSDMap: fix output from ceph status --format=json for num_in_osds
- num_up_osds returns as an int value, while num_in_osds returns as a string.
Since only an int can be returned from ge... - 07:58 PM Revision 27590c7a (ceph): Merge pull request #915 from ceph/wip-rgw-expose-bucket
- 6748: rgw: Optionally return the bucket name in a response header.
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com> - 07:36 PM Revision 3194d66d (ceph): doc: Fixed keyring command and updated for current conventions.
- fixes: #6692
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 05:57 PM Revision 66312f7e (ceph): ceph_manager: in test_map_discontinuity, delay killing osd for 20s
- This should give enough time for the replica to get the message
marking them clean.
Signed-off-by: Samuel Just <sam.... - 05:23 PM Bug #7160 (Duplicate): rados tests failed on next branch with scrub mismatch failures
- 05:17 PM Revision ec5f7a5e (ceph): client: ceph-fuse use fuse_session_loop_mt to allow multithreaded opera...
- Signed-off-by: Moritz Moeller mm@mxs.de
- 04:57 PM Documentation #6682 (Resolved): Adjustments to the Ceph Quick installation guide
- 04:53 PM Revision 53fc2d93 (ceph): Log a warning when killing long-running jobs.
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:46 PM devops Feature #7171 (Resolved): rbdmap should be part of ceph-common
- Currently, the rbdmap sysvinit/upstart script and /etc/ceph/rbdmap are included only in the ceph package.
Machines... - 04:38 PM Revision 769ef8a9 (ceph): Kill jobs that run for over 3 days (configurable)
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:23 PM Revision d4160fde (ceph): Merge pull request #1094 from ceph/wip-7117-redhat
- packaging: ship libdir/ceph
Reviewed-by: Sage Weil <sage@inktank.com> - 04:20 PM Bug #7170 (Can't reproduce): rados api test LibRadosMisc.CopyPP failed on next branch
- logs are copied to mira057.front.sepia.ceph.com:/home/ubuntu/testing_jan16_2...
- 04:17 PM rgw Bug #7169 (Resolved): rgw: list multipart parts broken (> 1000 parts)
- 04:11 PM Documentation #6959 (Resolved): Diagrams showing even numbers of Monitors
- 12:35 PM Documentation #6959 (In Progress): Diagrams showing even numbers of Monitors
- 02:26 PM rgw Bug #7168 (Fix Under Review): 404 Errors When save immediately follows a delete
- 08:45 AM rgw Bug #7168: 404 Errors When save immediately follows a delete
- 05:14 AM rgw Bug #7168 (Resolved): 404 Errors When save immediately follows a delete
- If the delete operation clashed with the save, and the save returned an 404. Is there an option to make save successf...
- 02:20 PM Revision e1fd0e81 (ceph): first commit
- 01:32 PM Revision 50808af9 (ceph): ceph-disk: larger timeout in the test script
- Otherwise slow machines may timeout just because they are slow and not
because the process is blocked by an actual pr... - 01:32 PM Revision ac5a9fec (ceph): mon: larger timeout for mon mkfs.sh test
- Otherwise slow machines may timeout just because they are slow and not
because the process is blocked by an actual pr... - 01:21 PM Revision d27731fe (ceph): Merge pull request #1095 from ceph/wip-7166
- osd: OSDMap: build reverse name->pool map upon decoding
Reviewed-by: Loic Dachary <loic@dachary.org> - 12:51 PM Revision 1aa26018 (ceph): osd: OSDMap: build reverse name->pool map upon decoding
- Commit 3d7c69fb09 introduced a new OSDMap encoding/decoding scheme.
However, while the classic decoding function stil... - 12:41 PM Bug #7071 (Resolved): osd/ReplicatedPG.cc: 5953: FAILED assert(attrs->count("_"))
- This was probably fixed by 0c81849c4f1661ce89b8623edaf2873bb6611f62.
- 11:36 AM Bug #6692 (Resolved): Documentation: Command line incorrect.
- http://ceph.com/docs/master/rados/operations/authentication/
- 10:45 AM Bug #6692 (In Progress): Documentation: Command line incorrect.
- 10:47 AM Subtask #7146 (In Progress): implement osd crush rule create-erasure
- ...
- 10:39 AM devops Documentation #5265 (Closed): node-name is confusing. hostname is probably more accurate
- looks good
- 10:31 AM devops Documentation #5265 (Resolved): node-name is confusing. hostname is probably more accurate
- http://ceph.com/docs/master/start/quick-ceph-deploy/
- 10:35 AM Documentation #3674 (Resolved): Deployment documentation is confusing
- Documentation on deployment has been completely rewritten, and mkcephfs is no longer part of it.
- 09:04 AM Feature #7154: Allow reweight of crush subtree via command line tools
- Greg Farnum wrote:
> In what scenarios would this actually be a good idea? You trying to do incremental bring-up or ... - 09:02 AM Revision f97264d4 (ceph): 6748: rgw: Optionally return the bucket name in a response header.
- This can be useful in situations where accounting of traffic is done externally
when for example HTTP traffic is cach... - 05:23 AM Bug #7166 (Resolved): mon: Error ENOENT: unrecognized pool 'data' on a new cluster
- 04:53 AM Bug #7166 (Fix Under Review): mon: Error ENOENT: unrecognized pool 'data' on a new cluster
- wip-7166, pr https://github.com/ceph/ceph/pull/1095
Waiting for the gitbuilders to pass the make checks. - 02:16 AM Bug #7166 (In Progress): mon: Error ENOENT: unrecognized pool 'data' on a new cluster
- This breaks current master (3b696f28d6cee1fd3a2b78911cc8377445158ba1). Might have something to do with the changes in...
- 02:02 AM Bug #7166 (Resolved): mon: Error ENOENT: unrecognized pool 'data' on a new cluster
- After compiling todays "master from sources":https://github.com/ceph/ceph/commit/3b696f28d6cee1fd3a2b78911cc837744515...
- 05:00 AM Feature #7167: Add op_process_latency in perf counters
- The code change has been done in pull request: https://github.com/ceph/ceph/pull/1096
Please help to review it. Thanks. - 04:59 AM Feature #7167 (Resolved): Add op_process_latency in perf counters
- We found that there is a need to get latency of op threads because op threads get slow during accessing LevelDB or ge...
- 03:49 AM Feature #6507: librados shouldn't block indefinitely when cluster doesn't respond
- I try to write perl bindings for librados, so it would be great to have that
managed inside librados. - 03:04 AM Revision 3b696f28 (ceph): Merge pull request #1086 from ceph/wip-temp-primary
- Add a "primary temp" to go with "pg temp"
Reviewed-by: Sage Weil <sage@inktank.com> - 03:02 AM Revision 2d0d48b8 (ceph): packaging: ship libdir/ceph
- Automake puts ceph_common.sh into libdir/ceph, but the Red Hat packaging
was not capturing this file.
Add the libdir... - 03:02 AM Revision 2b36761b (ceph): mon: favor 'mds set max_mds' over 'mds set_max_mds'
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:02 AM Revision 8b09a436 (ceph): doc/release-notes: v0.76 draft notes
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:01 AM Revision f3021efe (ceph): Merge pull request #1093 from kdreyer-inktank/redhat-missing-tests
- packaging: add missing test files
Reviewed-by: Sage Weil <sage@inktank.com> - 02:40 AM Revision e60dcfa8 (ceph): packaging: add missing test files
- Two tests were inadvertantly missing from the RPM packaging:
ceph_test_c_headers was added to ceph in
ffdd30efc870ea... - 12:45 AM Revision 5ed7865a (ceph): rgw: use configurable prime number for gc hash
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 12:33 AM Revision 0c305093 (ceph): OSDMap: add primary out params to internal _pg_to_up_acting_osds function
- And use pointers instead of references for out params.
Now pg_to_up_acting_osds and pg_to_acting_osds can plug in to... - 12:33 AM Revision 12122b11 (ceph): OSDMap: add primary out params to _pg_to_osds and _raw_to_up_osds
- Switch to use pointers for the out parameters instead of references.
These functions are still just pointing at the ... - 12:33 AM Revision fafc8e93 (ceph): OSDMap: pay attention to the temp_primary in _get_temp_osds
- Switch _get_temp_osds to use pointers instead of references, and force callers
to check the out params instead of rel... - 12:33 AM Revision b1831157 (ceph): OSDMap: move temp manipulation functions out of OSDMonitor
- In doing so, consolidate remove_down_pg_temp() and remove_down_primary_temp().
Signed-off-by: Greg Farnum <greg@inkt... - 12:33 AM Revision 28e0b761 (ceph): test: add an OSDMap unittest
- This is not super-sophisticated, but it does basic mapping function
consistency checks and looks at the [pg|primary]_... - 12:33 AM Revision e7b32367 (ceph): OSDMonitor: make sure we don't send out maps with a primary_temp mapping
- Making sure a cluster supports primary_temp is complicated and we don't
have any of the machinery in place right now ... - 12:33 AM Revision 2646d5ed (ceph): OSDMap: add primary out param to pg_to_raw_up, and use pointers instead...
- The only user is in the OSDMonitor, and it's going to want that
information anyway.
Signed-off-by: Greg Farnum <greg... - 12:33 AM Revision 3d7c69fb (ceph): OSDMap: add a CEPH_FEATURE_OSDMAP_ENC feature, and use new encoding
- Bring our OSDMap encoding into the modern Ceph world! :) This is
fairly straightforward, but has a few rough edges:
P... - 12:33 AM Revision b55c45e8 (ceph): OSDMap: update Incremental encode/decode to match the full map's
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision 74bdcb69 (ceph): OSDMap: add [new_]primary_temp to the map and Incremental
- It's not used actively yet, but there it is.
Signed-off-by: Greg Farnum <greg@inktank.com> - 12:33 AM Revision e9e615cb (ceph): OSDMap: add primary_temp to apply_incremental()
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision 1f81fda5 (ceph): OSDMap: dedup the primary_temp
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision a2460395 (ceph): OSDMap: expose the primary_temp in print()
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision e2db4aea (ceph): OSDMonitor: remove primary_temp entries when you remove their pool
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision 412afea9 (ceph): OSDMonitor: make remove_redundant_pg_temp clear primary, too
- So that this works with future CRUSH changes, we copy the map and clear
out the primary_temp, then compare its output... - 12:33 AM Revision 1c750c65 (ceph): OSDMonitor: implement remove_down_primary_temp()
- Same as remove_down_pg_temp()
Signed-off-by: Greg Farnum <greg@inktank.com> - 12:33 AM Revision 268ae82a (ceph): osd: do not misuse calc_pg_role
- We've been using the role returned from this to determine if we're
the primary or not. Don't.
This is mostly about re... - 12:33 AM Revision 7a9c1712 (ceph): OSDMap: doc the different pg->OSD mapping functions
- Some of these look like what you should use for mapping and they absolutely
are not suitable for that. Make it cleare... - 12:33 AM Revision c1a95f83 (ceph): OSDMap: remove get_pg_primary() function
- This was used only by SyntheticClient, and that wants get_pg_acting_primary()
anyway. Delete the easily-misused get_p... - 12:33 AM Revision 69a2ec27 (ceph): OSDMap: unify the pg_to_acting_osds and pg_to_up_acting_osds implementa...
- These were the same except for a call to _raw_to_up_osds(). Move the
existing pg_to_up_acting_osds into a private fun... - 12:33 AM Revision 5367d92e (ceph): OSDMap: rename _raw_to_temp_osds() -> _get_temp_osds()
- This function does not (and never has!) used the raw vector, so remove it
and don't use a name which implies it is do... - 12:33 AM Revision 5b699782 (ceph): OSDMap: switch pg_to_osds to have an explicit primary param
- Use pointers instead of references for the out params, too!
Signed-off-by: Greg Farnum <greg@inktank.com> - 12:33 AM Revision 9749f30c (ceph): OSDMap: implement pg_to_up_acting_osds with primary interface
- Use our pointer calling conventions instead of a reference for the
new version of the function.
Right now we're just... - 12:33 AM Revision 93d481a5 (ceph): mon, osdmaptool: switch to primary-specifying pg_to_up_acting_osds
- Signed-off-by: Greg Farnum <greg@inktank.com>
- 12:33 AM Revision 045e1d75 (ceph): OSDMap: add primary-specifying pg_to_acting_osds
- This works the same as pg_to_up_acting_osds
Signed-off-by: Greg Farnum <greg@inktank.com> - 12:33 AM Revision a09d4f17 (ceph): PG: do not use role == 0 as a determinant of primacy
- We already have an is_primary() function to use instead.
Signed-off-by: Greg Farnum <greg@inktank.com>
01/15/2014
- 11:28 PM Revision c60ae09b (ceph): Merge pull request #978 from ceph/wip-3454
- Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
- 11:12 PM Revision 644afd67 (ceph): radosgw-admin: add temp url params to usage
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 11:05 PM rbd Bug #7165 (Resolved): high ping latency in vm using rbd writeback cache in qemu 1.7 with async flush
- See thread on qemu-devel: http://lists.nongnu.org/archive/html/qemu-devel/2014-01/msg01993.html
It happens only wh... - 10:51 PM rbd Bug #6851 (Resolved): rbd snap unprotect test failing
- commit:e91fb910653a672560867d4a81aa30f9d5dc0af8
- 10:50 PM rbd Bug #6693 (Resolved): "rbd ls" returns error if the pool empty
- commit:ac547a5b7dc94282f079aef78e66348d99d9d5e9
- 07:06 PM Bug #7117 (Fix Under Review): centOS: init script silently dies due to missing '/usr/lib64/ceph/c...
- I've verified that with this patch, the RPM packaging properly captures the @ceph_common.sh@ file in @/usr/lib64/ceph...
- 05:17 AM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- This is now affecting every test that runs on a system with that init script.
For example, all OSDs deployed in th... - 06:43 PM Revision fdf1a410 (ceph): autogen: re-run aclocal after libtoolize
- This seems to prevent configure from re-running after the first
invocation of 'make' following ./autogen.sh. The auto... - 06:25 PM Revision 980ef0e8 (ceph): Merge pull request #1089 from dachary/wip-mailmap
- mailmap: add athanatos <sam.just@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com> - 06:22 PM Revision 73e469c9 (ceph): Merge pull request #963 from dachary/wip-erasure-code-api
- erasure code interface helpers
Reviewed-by: Samuel Just <sam.just@inktank.com> - 06:08 PM Revision 970f9387 (ceph): doc: Updated paths for OSDs using the OS disk.
- fixes: #6682
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 04:49 PM rgw Bug #7054 (Fix Under Review): rgw: replace mongoose with civetweb
- 04:47 PM rgw Bug #7073 (Fix Under Review): "rgw gc max objs" should have a prime number as default value
- 04:33 PM Bug #7161 (Can't reproduce): rados api test LibRadosMisc.Exec failed on next branch
- logs: ubuntu@teuthology:/a/teuthology-2014-01-14_23:00:02-rados-next-distro-basic-plana/39381...
- 04:29 PM Feature #6835 (Fix Under Review): EC: ec pgs will need to be able to specify temp primaries other...
- https://github.com/ceph/ceph/pull/1086
- 04:26 PM Bug #7160 (Duplicate): rados tests failed on next branch with scrub mismatch failures
- logs: ubuntu@teuthology:/a/teuthology-2014-01-14_23:00:02-rados-next-distro-basic-plana/39382...
- 04:14 PM Bug #7159 (Resolved): ceph status --format=json num_in_osds and num_up_osds formatting not consis...
- From support ticket #932
root@burnupi02:/home/ubuntu# ceph status --format=json-pretty
"osdmap": { "osdmap": ... - 04:00 PM rgw Feature #3454 (Resolved): Support temp URLs for Swift API
- merged upstream, commit:c60ae09b389c808735415553e31c871cfa42f776. I opened a new task (#7156) for creating a test for...
- 03:07 PM Subtask #7158 (Resolved): EC: flesh out how the ceph tool should be used to manage ec pools and c...
- Not sure how much of this has been handled already.
- 02:45 PM devops Bug #7157 (Duplicate): ceph-disk list fails in encrypted disk setup
- In a setup (Dumpling 0.67.5) that is using dm-crypt "ceph-disk" list fails with:
# ceph-disk list
Traceback (most... - 02:37 PM rgw Feature #7156 (New): rgw: create tests for temp url functionality
- 02:33 PM rgw Feature #6874 (Fix Under Review): rgw: user level quota: rest api
- 02:33 PM rgw Feature #6873 (Fix Under Review): rgw: user level quota: radosgw-admin interface
- 02:33 PM rgw Feature #6872 (Fix Under Review): rgw: user level quota: async update of bucket info in user objc...
- 02:33 PM rgw Feature #6871 (Fix Under Review): rgw: user level quota: core functionality
- 02:19 PM Feature #7154: Allow reweight of crush subtree via command line tools
- In what scenarios would this actually be a good idea? You trying to do incremental bring-up or something?
- 02:08 PM Feature #7154 (Resolved): Allow reweight of crush subtree via command line tools
- Instead of having to individual adjust weights for all osds in a rack or downloading the crushmap, modifying and enti...
- 11:27 AM Bug #7116: mon: pg_temp left behind if pool deleted while pg_temp exist
The first cut at fixing this problem is in wip-7116. Sage had comments on the pull request for this branch https:/...- 10:58 AM Revision 4425f9ed (ceph): librados: Add C API coverage for atomic write operations
- Signed-off-by: Christian Marie <christian@ponies.io>
- 10:12 AM Documentation #4117 (In Progress): Document various language bindings better
- http://ceph.com/docs/wip-doc-librados-intro/rados/api/librados-intro/
I've covered the basics for C, C++, Java and... - 08:23 AM Revision 1ffe4226 (ceph): mailmap: add athanatos <sam.just@inktank.com>
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 05:57 AM Revision 4050eae3 (ceph): Merge pull request #1084 from dachary/wip-cephtool-test
- qa: cleanup cephtool/test.sh tmp files
Reviewed-by: Sage Weil <sage@inktank.com> - 12:17 AM Feature #7150 (Duplicate): mon: hunt for multiple mons in parallel
- I use a few helper scripts to automate things, and those scripts
call ‘ceph’ command multiple times, like
------...
01/14/2014
- 11:12 PM Revision f19adc91 (ceph): Merge pull request #1085 from dachary/ceph-master
- Reviewed-by: Greg Farnum <greg@inktank.com>
- 11:08 PM Revision 4b5f2570 (ceph): common: fix bufferlist::append(istream) test
- bufferlist::append(istream) now filters out empty lines; reflect this in
the test
Signed-off-by: Loic Dachary <loic@... - 06:52 PM Feature #5997 (In Progress): EC: [link] Refactor scrub to use PGBackend methods
- 06:15 PM Bug #7116 (Fix Under Review): mon: pg_temp left behind if pool deleted while pg_temp exist
- 05:37 PM Revision e55a0896 (ceph): doc/release-notes: v0.75
- Signed-off-by: Sage Weil <sage@inktank.com>
- 05:03 PM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- I've pushed an experimental patch to https://github.com/ceph/ceph/tree/wip-7117-redhat
I need to test and see what... - 04:31 PM Revision 08c17b7c (ceph): qa: cleanup cephtool/test.sh tmp files
- When run in a shared environment ( as opposed as a machine created for
the purpose of running this test only ), it is... - 04:20 PM Feature #7149 (Resolved): EC: missing_loc should be maintained until the object is no longer degr...
- 04:16 PM Revision 03d7d97d (ceph): Merge branch 'next'
- 04:10 PM Revision a520026e (ceph): Merge pull request #1076 from dachary/wip-vector-op
- erasure-code: use uintptr_t instead of long long
Reviewed-by: Andreas Peters <andreas.joachim.peters@cern.ch> - 03:57 PM devops Bug #7144 (Resolved): Ceph build fails on fedora20 gitbuilder
- Ok this gitbuilder is working correctly now. rescheduled a bunch of the failed jobs to run again.
- 02:54 PM devops Bug #7144: Ceph build fails on fedora20 gitbuilder
- So it looks like it was going OOM. I didn't expect this as 12G is usually enough ram for our gitbuilders. I went ahea...
- 10:17 AM devops Bug #7144 (Resolved): Ceph build fails on fedora20 gitbuilder
- All builds are failing in the middle of the build when hitting mds:
http://gitbuilder-ceph-rpm-fedora20-amd64-basi... - 03:00 PM Subtask #7146 (Resolved): implement osd crush rule create-erasure
- Currently, what we need to do is make sure the EC plugin maps onto a
linear array of devices the same way that CRUS... - 02:56 PM Bug #4934: ceph-deploy: librbd1 missing as a dependency
- Interesting thing here. I think that this has been biting us recently. I will open another ticket on this because it...
- 11:23 AM Bug #7145: PG::update_snap_map assert
- ubuntu@teuthology:/a/teuthology-2014-01-13_23:00:02-rados-master-testing-basic-plana/37582/remote
- 11:23 AM Bug #7145 (Can't reproduce): PG::update_snap_map assert
- 0> 2014-01-14 01:35:32.305544 7f7bb3db0700 -1 osd/PG.cc: In function 'void PG::update_snap_map(std::vector<pg_lo...
- 11:21 AM Bug #7141 (Resolved): osd: segv in DBObjectMap::clear_keys_header()
- 09ae4bc2aa2530ea0d20c7487a4e1c11cb7eb488
- 10:15 AM Linux kernel client Bug #6809: 3.11 kernel panic: Workqueue: ceph-msgr con_work
- I am seeing a similar issue with Kernel 3.12.1 on Ubuntu 1310
Jan 14 03:09:27 gw2 kernel: [204578.112175] libceph:... - 08:38 AM devops Bug #6726: Official packages do not appear to be available for Saucy
- Since this has been brought to Mark and Peter's attention, is this going to be remedied any time soon? In the meantim...
- 06:48 AM Revision b67e9ef7 (ceph): ceph: Add inline state definition
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Signed-off-by: Li Wang <liwang@ubuntukylin.com> - 06:48 AM Revision caf79714 (ceph): mds: Add inline fields to inode_t
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Signed-off-by: Li Wang <liwang@ubuntukylin.com> - 06:48 AM Revision 5cf69089 (ceph): mds: Add inline encode/decode to inode_t
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Signed-off-by: Li Wang <liwang@ubuntukylin.com> - 06:48 AM Revision 978ec18b (ceph): ceph: Add inline data feature
- Signed-off-by: Yunchuan Wen <yunchuanwen@ubuntukylin.com>
Signed-off-by: Li Wang <liwang@ubuntukylin.com> - 06:38 AM Revision dc4e212d (ceph): Merge pull request #1078 from ceph/wip-mon-pgmap
- mon: make 'pg getmap' not include a trailing newline
Reviewed-by: Loic Dachary <loic@dachary.org> - 03:23 AM Revision e268e956 (ceph): updated "sample.ceph.conf":
- I've made the following changes to simplify initial cluster configuration:
* Removed deprecated option(s).
* Remov... - 02:00 AM rbd Bug #7125: Assertion failure in rbd_img_obj_callback()
- Thanks Eric, I'll try to reproduce it here on a smaller scale this week.
- 01:56 AM rbd Bug #5425 (Resolved): krbd: xfstest 89 hang, 'read_partial_message skipping long message'
- As noted in the ceph-devel post referenced in the above comment, this
is bug is sort of twofold. The first problem ... - 01:43 AM Revision 66a4f8a2 (ceph): Merge pull request #1071 from ceph/wip-max-file-size
- allow mds max file size to be adjusted
Reviewed-by: Yan, Zheng <zheng.z.yan@intel.com> - 12:55 AM Revision 413fc237 (ceph): osd/ReplicatedPG: no HitSet on non-primary
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:55 AM Revision 91b8c783 (ceph): mon/OSDMonitor: simplify and make 'osd pool set ...' output consistent
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:50 AM Revision c5cacf4e (ceph): Merge pull request #1058 from ceph/wip-cache-snap
- snap/clone promotion, flush, and other goodies
This is now passing the thrashing with both cache and snap ops:
sag... - 12:19 AM Revision be8db8c3 (ceph): osd/ReplicatedPG: use get_object_context in trim_object
- find_object_context() has all the logic to choose a particular clone given
a logical snap. In the trim case, we want... - 12:19 AM Revision f3ce2549 (ceph): osd/Replicated: mark obc !exists on snap trim
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision 8221a2a5 (ceph): osd/ReplicatedPG: cleanly abort flush if the object no longer exists
- If the object no longer exists (for example, because the snap trimmer just
killed it) clean up the flush state withou... - 12:19 AM Revision 8cab9e76 (ceph): ceph_test_rados_api_tier: partial test for promote vs snap trim race
- This reliably returns ENODEV due to the test at the finish of flush. Not
because we are actually racing with trim, t... - 12:19 AM Revision 10547e67 (ceph): osd/ReplicatedPG: always return ENOENT on deleted snap
- Previously, if a snap was deleted but the clone was there and we hadn't
trimmed it yet, we would still return the dat... - 12:19 AM Revision d41a1d3d (ceph): osd/ReplicatedPG: update ObjectContext's object_info_t for new hit_set ...
- We were fabricating an object_info_t correctly and writing it to disk, but
it was not reflected by the in-memory Obje... - 12:19 AM Revision 8b39719a (ceph): osd/OSDMonitor: fix 'osd tier add ...' pool mangling
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision b5ae76e8 (ceph): ceph_test_rados: do not delete in-use snaps
- There are a bunch of ops that read from snaps. Do not delete a snap
while they are in use.
Signed-off-by: Sage Weil... - 12:19 AM Revision 7daab5ac (ceph): osd/ReplicatedPG: adjust flow in process_copy_chunk
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision 1a7335d5 (ceph): ceph_test_rados_misc: test bad version for copy-from
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision cd42368e (ceph): osd/ReplicatedPG: simplify copy-from temp object handling
- Previously the caller was generating a temp object name and passing it
down in severaly different ways. Instead, gen... - 12:19 AM Revision 7e80fa06 (ceph): osd/ReplicatedPG: handle promote that races with snap deletion
- If we are promoting a clone and realize that the object is no longer
defined for any snaps, abort the copy and delete... - 12:19 AM Revision 6719d302 (ceph): osd: fix propagation of removed snaps to other tiers
- When we update removed_snaps we do not update snap_seq. Drop this broken
optimization.
Signed-off-by: Sage Weil <sa... - 12:19 AM Revision 48306e47 (ceph): mon: debug propagate_snaps_to_tiers
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision 21f3dcbd (ceph): osd: include snap_seq in copy-get results
- This is needed by the cache layer when reading a logical snap from a head
object on the backend in order to correctly... - 12:19 AM Revision 627bdead (ceph): osd/ReplicatedPG: infer snaps from head when promoting oldest clean clone
- Consider:
- base and cache have same object foo; marked clean in cache pool
- modify + clone foo in cache pool. f... - 12:19 AM Revision 27eb4c5e (ceph): ceph_test_rados: improve read debug output
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision ac446b5d (ceph): osd/ReplicatedPG: preserve clean/dirty state on clone
- If we have a clean object and clone it in make_writeable(), the clone
should also be clean (it does not need to be wr... - 12:19 AM Revision 7eede85f (ceph): osd/ReplicatedPG: handle promotion of rollback, src_oids, etc.
- Make other find_object_context() callers handle the case where the object
in question needs to be promoted. We add a... - 12:19 AM Revision d00116c6 (ceph): ceph_test_rados: flush can also fail due to snap trimming
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision 0b816c33 (ceph): osd/ReplicatedPG: make CopyResults inline in CopyOp
- No reason to put this on the heap. Make the lifetime match that of the
CopyOp.
Signed-off-by: Sage Weil <sage@inkta... - 12:19 AM Revision 934de77c (ceph): osd/ReplicatedPG: do not promote nonexistent clones
- Do not promote a clone for a snap that we know doesn't exist. If
find_object_context() didn't give us a missing_oid,... - 12:19 AM Revision c6b73eb4 (ceph): osd/ReplicatedPG: always set obc->ssc SnapSetContext for clones
- This can be useful!
Signed-off-by: Sage Weil <sage@inktank.com> - 12:19 AM Revision 55b83f16 (ceph): ceph_test_rados: is_dirty on non-flushing objects only
- This makes its results reliable. Otherwise, we can't mix the is_dirty
test with flush, which eliminates much of its ... - 12:19 AM Revision 9688642c (ceph): ceph_test_rados: don't update any state on successful cache-evict
- - we didn't touch the user_version
- we didn't change the clean/dirty state
Signed-off-by: Sage Weil <sage@inktank.com> - 12:19 AM Revision 6f4f6513 (ceph): ceph_test_rados: test flush/evict on snaps
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision ac635513 (ceph): ceph_test_rados: IsDirty on random snaps
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision b70c476a (ceph): ceph_test_rados: make flush clean correct snap in model
- 12:19 AM Revision af5a407c (ceph): ceph_test_rados: assert on read error
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision de8e8b5d (ceph): osd/ReplicatedPG: make cache-[try-]flush CACHE instead of WR ops
- This will allow us to send a flush op on a snap.
Signed-off-by: Sage Weil <sage@inktank.com> - 12:19 AM Revision 6bff648d (ceph): vstart.sh: allow MDS=0
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision c91166ec (ceph): osd/ReplicatedPG: refuse to flush when older dirty clones are present
- If the next oldest clone is dirty, we cannot flush. That is, we must
always flush starting with the oldest dirty clo... - 12:19 AM Revision 5b8d957b (ceph): osd: add pg_log_entry_t event type CLEAN
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision b2f752a9 (ceph): osd/ReplicatedPG: construct appropriate snapc for flush/writeback
- Construct a snap context that will trigger the appropriate cloning (if any)
on the base pool.
Signed-off-by: Sage We... - 12:19 AM Revision fc9f8ad5 (ceph): ceph_test_rados_api_tier: test flush on snaps/clones
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision c05765e8 (ceph): osd/ReplicatedPG: update snap_mapper for promoted clones
- A clone that comes into existence via promotion takes an entirely
different path than a typical clone (which comes in... - 12:19 AM Revision 1f4350e2 (ceph): ceph_test_rados_api_tier: test promotion of clones
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision 90e352ca (ceph): osd: add rados CACHE mode (different from RD and WR)
- It is useful to distinguish cache operations from read and modify
operations. Specifically, we will allow cache ops ... - 12:19 AM Revision 4e8259db (ceph): osd/ReplicatedPG: allow cache-evict on snaps
- We do three things here:
- make cache-evict a CACHE instead of WR op, allowing us to submit it
on snaps (not jus... - 12:19 AM Revision 5c94d530 (ceph): osd/ReplicatedPG: only encode SnapSet on head objects in finish_ctx
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision bfd45301 (ceph): osd/ReplicatedPG: mirror SnapSet info when promoting head
- When we promote the head for an object, get the list of snaps from the
backend pool and construct an appropriate Snap... - 12:19 AM Revision 38fe575d (ceph): osd/ReplicatedPG: always encode snaps in finish_ctx
- On promote we use finish_ctx to build the final log entries, and need to
encode the snaps vector in that case. (Norm... - 12:19 AM Revision c3c1541c (ceph): osd/ReplicatedPG: make find_object_context() pass missing_oid
- Prevoiusly we would return a snapid that we are blocked on if it is
missing. This is necessary because the missing c... - 12:19 AM Revision d22ecf3e (ceph): osd/ReplicatedPG: using missing_oid to decide which object to promote
- find_object_context() now tells us which object it could use if it
doesn't find it on disk. Promote that one.
Signe... - 12:19 AM Revision 6dd0a1f0 (ceph): osd/ReplicatedPG: include snaps in copy-get results
- When promoting a snapped object, we need to also get the set of snaps over
which the clone is defined. This is not s... - 12:19 AM Revision b840aae1 (ceph): osd/ReplicatedPG: adjust clone stats when promoting clones
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:19 AM Revision c70edf3e (ceph): osd/ReplicatedPG: add PROMOTE log entry type
- This is an alternative to MODIFY that indicates the object was just
promoted from another tier. Thanksfully, is_modi... - 12:19 AM Revision 05547358 (ceph): osd/osd_types: SnapSet::from_snap_set
- Signed-off-by: Sage Weil <sage@inktank.com>
01/13/2014
- 11:52 PM Revision 33b5ef40 (ceph): mon/PGMap: make decode version match encode version
- These should have been bumped way back in 091809b8.
Signed-off-by: Sage Weil <sage@inktank.com> - 11:52 PM Revision a5aaab3c (ceph): ceph-dencoder: include offset in 'stray data' error message
- Signed-off-by: Sage Weil <sage@inktank.com>
- 11:52 PM Revision 1308225c (ceph): buffer: do not append trailing newline when appending empty istream
- If we call
bl.append(some_istream);
do not include a \n if the istream is empty (which is apparently is not
the sa... - 10:25 PM Revision 196e3d62 (ceph): Merge pull request #931 from ceph/wip-5858-rebase
- Wip 5858 rebase
Reviewed-by: Samuel Just <sam.just@inktank.com> - 09:07 PM Revision 946d6036 (ceph): v0.75
- 08:57 PM Revision 90343708 (ceph): doc: Added comment and example for SSL enablement in rgw.conf
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:40 PM Revision 96156453 (ceph): doc: Added python example of handle, and closing session examples.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:39 PM Revision a657fad4 (ceph): osd: Interim backfill changes
- Make peer_backfill_info a map which holds a
BackfillInterval for all backfill targets.
Initially see if recover_backf... - 07:39 PM Revision c0d92b67 (ceph): osd: Implement multiple backfill target handling
- Fixes: #5858
Signed-off-by: David Zafman <david.zafman@inktank.com> - 07:34 PM devops Bug #6746: ceph-release rpm not playing well with yum-plugin-priorities
- Ken - hard to say...it's a fairly important repo so I'd say 2 or 3 on our 10 point scale, but I suspect this varies f...
- 07:22 PM Revision 09ae4bc2 (ceph): Merge pull request #1077 from ceph/wip-7141
- DBObjectMap::clear_keys_header: use generate_new_header, not _generate_n...
Reviewed-by: Sage Weil <sage@inktank.com> - 07:20 PM Revision 54e588c8 (ceph): doc: Modified doc examples to use rados_create2.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:02 PM Revision 4c92dc6f (ceph): DBObjectMap::clear_keys_header: use generate_new_header, not _generate_...
- We aren't holding the header_lock here, so we need the locked version.
Signed-off-by: Samuel Just <sam.just@inktank.... - 06:11 PM rbd Bug #7142 (Resolved): list_lockers() never returns after cluster restart and health_ok (librbdpy)
- The cluster was shut off and then all nodes were restarted. Mon's came up first, then OSDs all at once). While the OS...
- 05:31 PM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- I noticed the following warning today while building v0.75.
http://jenkins.ceph.com/job/ceph-build/Arch=x86_64,Dis... - 05:16 PM Revision 93a9b686 (ceph): erasure-code: use uintptr_t instead of long long
- Checking the pointer alignment using a cast to long long raises a
warning when --Wpointer-to-int-cast is given.
Sign... - 04:50 PM Feature #6452 (Resolved): cachepool: snapshots
- 04:46 PM Revision 5f165ed3 (ceph): Merge pull request #1075 from dachary/wip-crush
- improve crushtool --build useability and documentation
Reviewed-by: Sage Weil <sage@inktank.com> - 04:33 PM Revision 0fd6a244 (ceph): Merge pull request #1072 from ceph/wip-tier-snap
- Reviewed-by: Greg Farnum <greg@inktank.com>
- 04:23 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- Adding these links for reference.
http://people.redhat.com/jcm/el6/dup/docs/old/pre-release/whitepaper.pdf
http... - 03:47 PM Subtask #5858 (Resolved): Backfill should be able to handle multiple backfill peers
- c0d92b6744d793f85c1ce2f7c5c67b9e9c18a135
- 03:46 PM Feature #7039 (Resolved): Check other OSD CEPH_FEATURE_OSD_ERASURE_CODES to enable acting set cha...
- f9a11ef2f267591a38feea543a7903936702c76a
- 03:45 PM Feature #5994 (Resolved): EC: [link] Backfill should be able to handle multiple backfill peers
- c0d92b6744d793f85c1ce2f7c5c67b9e9c18a135
- 11:05 AM Bug #7141: osd: segv in DBObjectMap::clear_keys_header()
- It would help to be holding the lock when accessing that map. Testing fix now.
- 09:37 AM Bug #7141 (Resolved): osd: segv in DBObjectMap::clear_keys_header()
- ...
- 03:20 AM rbd Bug #5425 (Fix Under Review): krbd: xfstest 89 hang, 'read_partial_message skipping long message'
- For the record, this turned out to be not related to xfstest 89 at all,
it was just a coincidence. See ceph-devel (... - 01:20 AM Linux kernel client Bug #7139 (In Progress): dead tcp connections during connection negotiation may lead to hangs
- Nothing too suspicious in the dmesg:...
01/12/2014
- 10:39 PM rbd Bug #7125: Assertion failure in rbd_img_obj_callback()
- Hi Ilya,
So far I have not reproduced the problem.
Ceph cluster info:
ceph --version
ceph version 0.72.2 (a... - 05:06 PM Revision d3393e9d (ceph): crush: tests for crushtool --build
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 05:06 PM Revision 283793ac (ceph): doc: crushtool man page nroff format
- also includes a modification from a prior patch series that was not
formatted to nroff.
Signed-off-by: Loic Dachary ... - 05:06 PM Revision efbdd163 (ceph): doc: update the crushtool manual page
- * add information about CEPH_ARGS
* rework the --build documentation and example
* add an Author section
* replace vi... - 05:06 PM Revision b4054fcc (ceph): doc: copyright s/2013/2014/
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 05:06 PM Revision 0082d88c (ceph): doc: format man pages with s/2013/2014/
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 04:48 PM Revision 5b28405a (ceph): crush: crushtool --build uses OSDMap helpers for rulesets
- Instead of creating a ruleset from scratch, use the
OSDMap::build_simple_crush_rulesets helper. It is more likely to ... - 04:48 PM Revision d3d75a21 (ceph): crush: crushtool --build informative messages
- * dump the crush tree created by --build at debug level 1.
* display a warning at debug level 1 if there is more tha... - 04:48 PM Revision b705e523 (ceph): crush: crushtool emacs compile helper
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 04:48 PM Revision 26f7fa96 (ceph): crush: crushtool copyright notice update
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 04:48 PM Revision 76765503 (ceph): osd: ostream is enough for build_simple*
- There is no need to specialize the argument into stringstream. It is
replaced by a ostream which is convenient to dis... - 04:48 PM Revision cf9a7649 (ceph): osd: factorize build_simple_crush_map* rulesets creation
- Group the rulesets created by build_simple_crush_map* into a helper:
build_simple_crush_rulesets()
Signed-off-by: Lo... - 04:48 PM Revision 2765f81a (ceph): crush: parse CEPH_ARGS in crushtool
- The arguments are not given to global_init because the -c option would
conflict. Reading arguments from CEPH_ARGS the... - 04:48 PM Revision 5b95d183 (ceph): crush: display args on crushtool failure
- When the number of args provided to --build is not a multiple of 3,
display the arguments which do not comply.
For i... - 04:48 PM Revision 1368229e (ceph): crush: print --build debug information when verbose 2
- instead of verbose 0
Signed-off-by: Loic Dachary <loic@dachary.org>
01/11/2014
- 01:57 AM Feature #7114: Hinted recovery
- David Zafman is working on "multiple backfills":https://github.com/ceph/ceph/pull/931 to address the case where the p...
01/10/2014
- 11:48 PM Revision dae248f2 (ceph): Merge pull request #998 from ceph/wip-omapdirfrag2
- use OMAP to store dirfrags
- 11:41 PM Revision cec8d858 (ceph): mds: require CEPH_FEATURE_OSD_TMAP2OMAP
- Require that all OSDs support TMAP2OMAP before starting the MDS. This
avoids doing some work and then crashing with ... - 11:40 PM Revision 0f0dd74c (ceph): osd: introduce TMAP->OMAP conversion operation
- Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
- 11:40 PM Revision b8dfcc1e (ceph): mds: use OMAP to store dirfrags
- MDS can fetch dirfrags from both TMAP and OMAP. When committing a
dirfrags that is stored in TMAP, MDS first uses OSD... - 11:40 PM Revision 1d8429de (ceph): osd/OSDMap: get_up_osd_features()
- Signed-off-by: Sage Weil <sage@inktank.com>
- 10:28 PM Revision 20f8bae9 (ceph): Merge pull request #1074 from ceph/wip-7135
- Fixes: #7135
Reviewed-by: Sage Weil <sage@inktank.com> - 10:02 PM Revision 617a50c0 (ceph): src/test/librados/misc.cc: verify that omap_clear doesn't blast xattr
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:02 PM Revision 90d6cec6 (ceph): src/test/ObjectMap: test clear_keys_header() call
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 10:02 PM Revision 0c81849c (ceph): os/DBObjectMap, FileStore: omap_clear should not remove xattrs
- Prevously, FileStore::_omap_clear() used ObjectMap::clear(), which
incorrectly also blasts any stored xattrs. Instea... - 09:58 PM Revision ecd50347 (ceph): Merge pull request #1073 from dachary/wip-mailmap
- mailmap updates
Reviewed-by: Sage Weil <sage@inktank.com> - 09:57 PM Revision 1ddf5893 (ceph): Merge remote-tracking branch 'gh/next'
- 07:50 PM Revision 6225f2cc (ceph): mailmap: fix typo in Alexandre Maragone
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 07:33 PM Revision 413afa82 (ceph): qa/workunits/cephtool/test.sh: hashpspool takes int in emperor
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:17 PM Revision d429ab54 (ceph): osd/OSDMonitor: fix 'osd tier add ...' pool mangling
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:17 PM Revision f49d9cde (ceph): osd: fix propagation of removed snaps to other tiers
- When we update removed_snaps we do not update snap_seq. Drop this broken
optimization.
Signed-off-by: Sage Weil <sa... - 07:15 PM Revision 3b3511c3 (ceph): mon: debug propagate_snaps_to_tiers
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:08 PM Revision 69d49627 (ceph): Merge pull request #174 from ceph/wip-7171
- thrashosds: change min_in from 2 -> 3
- 07:02 PM Revision 9a9ee610 (ceph): valgrind: ignore tcmalloc uninitialized memory
- This is the main source of noise when running valgrind +
tcmalloc. Apparently there are other issues, so I think we
... - 07:00 PM Revision 495f2163 (ceph): thrashosds: change min_in from 2 -> 3
- See #7171. In rare cases CRUSH can't handle it when only 2/6 of
the OSDs are marked in. Avoid those situations for n... - 06:44 PM Revision 631d0c71 (ceph): erasure-code: erasure code decode interface helper
- The decode_concat method is implemented, documented and tested. It is a
convenience method to concatenate chunks into... - 06:43 PM Revision d2b2f5de (ceph): erasure-code: jerasure implementation of chunk size helpers
- The encode methode uses the get_chunk_size method which is a translation
of the code it previously used for the same ... - 06:24 PM Revision 1f3c1bdd (ceph): erasure-code: add chunk size calculation helpers
- With the current erasure code interface, figuring out the size of a
chunk and the number of chunks for a given object... - 06:24 PM Revision eb2374ea (ceph): erasure-code: refactor the example to use chunk size helpers
- The get_chunk_size and get_chunk_count methods are implemented using the
code found in the encode() method. The encod... - 06:08 PM Revision 72168f71 (ceph): rgw: convert bucket info if needed
- Fixes: #7110
In dumpling, the bucket info was separated into bucket entry point and
bucket instance objects. When set... - 06:08 PM Revision ee9fc590 (ceph): rgw: convert bucket info if needed
- Fixes: #7110
In dumpling, the bucket info was separated into bucket entry point and
bucket instance objects. When set... - 06:07 PM Revision fa43d9ce (ceph): organizationmap: joe.buck@inktank.com is Inktank
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 05:51 PM Revision 74489532 (ceph): Merge pull request #1062 from ceph/wip-7110
- rgw: convert bucket info if needed
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 05:45 PM Revision a0e3f05d (ceph): Merge pull request #1070 from dachary/wip-organizationmap
- organizationmap: match authors with organizations
Reviewed-by: Sage Weil <sage@inktank.com> - 05:44 PM Revision 8604e762 (ceph): organizationmap: match authors with organizations
- Using the same format as .mailmap, match author names with the
organization sponsoring them, if any. It can be used f... - 05:26 PM Revision ce95cef3 (ceph): mongoose: fix warning
- Signed-off-by: Sage Weil <sage@inktank.com>
- 05:22 PM Revision 0c339fe4 (ceph): Merge pull request #169 from ceph/wip-utsrelease
- kernel: use utsrelease string for need_to_install() purposes
- 04:53 PM Revision 8f01eeb5 (ceph): Merge pull request #1069 from dachary/wip-mailmap
- mailmap updates
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Roald J. van Loon <roald@roaldvanloon.nl>
Rev... - 02:29 PM Bug #7065 (Resolved): omap_clear() also removes all of object's xatts
- 0c81849c4f1661ce89b8623edaf2873bb6611f62
- 02:29 PM Bug #7135 (Resolved): OMAP_CLEAR removes xattr entries as well
- 0c81849c4f1661ce89b8623edaf2873bb6611f62
- 01:26 PM Bug #7135: OMAP_CLEAR removes xattr entries as well
- wip-7135
- 10:39 AM Bug #7135: OMAP_CLEAR removes xattr entries as well
- Dup #7065?
- 10:34 AM Bug #7135 (Resolved): OMAP_CLEAR removes xattr entries as well
- The issue is in FileStore::_omap_clear, which calls ObjectMap::clear() rather than some form of ObjectMap::clear_omap...
- 02:15 PM Documentation #7007: Missing page for "OS Recommendations"
- Trying to update...
- 11:35 AM Feature #6452 (Fix Under Review): cachepool: snapshots
- 11:35 AM Feature #7057 (Resolved): crush location
- 11:21 AM Bug #7121 (Resolved): nightlies: failed to become clean before timeout expired due to degraded pg
- 10:59 AM Bug #7121: nightlies: failed to become clean before timeout expired due to degraded pg
- Yeah; our tests run a lot closer to the edges than users really do or should, so I'd just adjust the thrasher's min_in.
- 10:55 AM Bug #7121: nightlies: failed to become clean before timeout expired due to degraded pg
- 10:55 AM Bug #7121: nightlies: failed to become clean before timeout expired due to degraded pg
- the behavior is the same as on dumpling, so this isn't a recent regression.
old tunables work for this pg, but ten... - 10:32 AM Bug #7121: nightlies: failed to become clean before timeout expired due to degraded pg
- Sam suggests that maybe the retry behavior got changed erroneously with the recent refactor, and that that's what cau...
- 10:39 AM rgw Bug #7110 (Resolved): rgw: cannot set acls on pre-dumpling buckets
- Landed to dumpling - ee9fc590a556354b54a174a4790c4142949d7409
Landed to emperor - 72168f717b2632f2942fcabf69798ff59a... - 09:56 AM rgw Bug #7110: rgw: cannot set acls on pre-dumpling buckets
- Landed to next 744895326c718efca720bf71e164d65e4eb29d32
- 09:54 AM rgw Bug #7110 (Pending Backport): rgw: cannot set acls on pre-dumpling buckets
- 09:54 AM rgw Bug #7110: rgw: cannot set acls on pre-dumpling buckets
- still need to backport it to dumpling and emperor
- 09:51 AM rgw Bug #7110 (Resolved): rgw: cannot set acls on pre-dumpling buckets
- 10:36 AM rbd Bug #7125 (Need More Info): Assertion failure in rbd_img_obj_callback()
- Hi Eric,
Is it reproducible?
What kind of stress testing were you doing? Can you share a script or
at least d... - 10:26 AM Feature #6173 (Fix Under Review): Add LevelDB support to ceph cluster backend store
- https://github.com/ceph/ceph/pull/1049.patch
- 10:26 AM Revision bf481b91 (ceph): kernel: use utsrelease string for need_to_install() purposes
- Currently, to see if a node has rebooted into the right kernel,
need_to_install() compares a given 40-char commit has... - 09:57 AM rgw Bug #7084 (Can't reproduce): segv in RGWDataChangesLog::RGWDataChangesLog
- reopen if this ever comes up again... but looks like a bad build or something :/
- 09:51 AM Revision af0269de (ceph): mailmap: make roald@roaldvanloon.nl primary
- It seems more appropriate than @gmail.com
Signed-off-by: Loic Dachary <loic@dachary.org> - 09:46 AM Revision f18b310c (ceph): mailmap: remove company name from Pascal de Bruijn name
- It is in the email address anyway.
Signed-off-by: Loic Dachary <loic@dachary.org> - 09:44 AM Bug #6097 (Resolved): btrfs locking regression on async snap ioctl
- 09:44 AM rgw Bug #7083 (Resolved): rgw: public-read acl test failing on master
- Landed to master 6fac2acc5e6f77651ffcd7dc7aa833713517d8a6
- 09:43 AM Revision 39db90f7 (ceph): mailmap: add Andrew Leung, Carlos Maltzahn and Esteban Molina-Estolano
- Although the commit aliases are cryptic, the timeframe and account names
seem to match.
Signed-off-by: Loic Dachary ... - 09:42 AM Bug #7103 (Resolved): config: variable substitution happen only once
- 09:42 AM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- Something with the make file?
- 08:36 AM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- AFAICT, the reason is that the init-script is not even starting any daemon due to failing on...
- 09:40 AM Bug #7126 (Resolved): mon: returns wrong version from mon_commands
- 09:38 AM Bug #7126: mon: returns wrong version from mon_commands
- actually, le'ts not backport it.. i don't think it affects anything except the new tiering commands.
- 09:09 AM Revision 176cec82 (ceph): mailmap: add David Moreau Simard alias
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 09:05 AM Revision 440fbed2 (ceph): Merge pull request #1060 from dachary/wip-mailmap
- mailmap updates for reviewed-by
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
Reviewed-by: Greg Farnum <greg@inktan... - 03:15 AM Revision 29da2ec4 (ceph): FileStore: detect XFS properly
- We were only setting m_fs_type = FS_TYPE_XFS if
m_filestore_replica_fadvise was also set -- presumably
the bug fix ac...
01/09/2014
- 10:44 PM Revision 8c5c55c8 (ceph): mon: set next commit in mon command replies
- The mon command acks include a version that is used by the client to
determine which version of the map they need to ... - 10:28 PM Revision 0e43ac90 (ceph): rgw: add optional tenant name for swift urls
- In order to maintain compatibility with swift clients that expect the
url to also contain a reference to the tenant n... - 10:28 PM Revision 56e4015c (ceph): rgw: handle temp url key authorization
- Fixes: #3454
Implement auth handling code that checks temp url signature and
expiration. The key that is used for the... - 10:28 PM Revision 7ccb513c (ceph): rgw: support multiple temp url keys
- Apparently the swift api defines two temp url keys. Add internal support
for multiple keys, radosgw-admin can config ... - 10:28 PM Revision 9395e0d7 (ceph): rgw: fix temp url restful apis and support multiple keys
- the set temp url api didn't do anything, fix that. Also add support for
multiple temp url keys.
Signed-off-by: Yehud... - 10:28 PM Revision 2626101f (ceph): rgw: can set temp url key on user
- Either through new through new rest api call, or via
radosgw-admin
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:31 PM Revision 5ce1e2e9 (ceph): Merge pull request #1061 from kdreyer-inktank/rm-execute-bit-src
- remove spurious executable permissions on source code files
Reviewed-by: Sage Weil <sage@inktank.com> - 06:26 PM Revision efb4cf64 (ceph): doc: Removed extraneous comment from snapshot text.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:15 PM Revision 70cf8d6c (ceph): Merge branch 'master' of https://github.com/ceph/ceph
- 05:31 PM Revision 3a4111a1 (ceph): Merge pull request #1063 from ceph/install-erasure-headers
- osd: add missing header to install list
Reviewed-by: Loic Dachary <loic@dachary.org> - 05:15 PM Revision 50705974 (ceph): osd: add missing header to install list
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 05:00 PM Revision a5f8cc7e (ceph): rgw: convert bucket info if needed
- Fixes: #7110
In dumpling, the bucket info was separated into bucket entry point and
bucket instance objects. When set... - 04:05 PM Bug #7122: push 0/hit_set_... v 0'0 failed because local copy is 818'7131
- 12:47 AM Bug #7122: push 0/hit_set_... v 0'0 failed because local copy is 818'7131
- ubuntu@teuthology:/var/lib/teuthworker/archive/sage-2014-01-08_22:39:23-rados:thrash-wip-cache-snap-testing-basic-pla...
- 12:42 AM Bug #7122 (Resolved): push 0/hit_set_... v 0'0 failed because local copy is 818'7131
- ...
- 03:57 PM Revision 029b9ef5 (ceph): remove spurious executable permissions on files
- Fedora's rpmlint complains that some of the source code files in the
tree happen to be executable. Remove the execute... - 03:39 PM Bug #7126 (Pending Backport): mon: returns wrong version from mon_commands
- 02:30 PM Bug #7126 (Resolved): mon: returns wrong version from mon_commands
- manifested as a broken LibRadosTier.Promote test, but is uspect this has been brekaing things for ages.
ubuntu@teu... - 02:03 PM rbd Bug #6548 (Resolved): tgt: Kernel panic putting zpool on iSCSI LUN using bs_rbd
- 12:36 PM rbd Bug #6548: tgt: Kernel panic putting zpool on iSCSI LUN using bs_rbd
- I have not seen a repeat of this issues with the new code base. I would recommend that we close this ticket.
- 01:43 PM Revision 4c049b37 (ceph): Merge pull request #173 from ceph/wip-ceph-deploy
- fix ceph-dpeloy tests to use 2x replication
- 01:14 PM rbd Bug #7125 (Resolved): Assertion failure in rbd_img_obj_callback()
- My system hung while stress testing an RBD backed XFS file system. After power cycling the system the error message ...
- 12:54 PM Revision 995a3cc5 (ceph): mds/MDSMap: include max_file_size in plaintext print output
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:49 PM Revision a3eb935f (ceph): mailmap: add Yehuda Sadeh fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 6d0ba4d9 (ceph): mailmap: add Greg Farnum fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 7aac3504 (ceph): mailmap: add Samuel Just fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 96d9b993 (ceph): mailmap: add Dan Mick fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 43fa0555 (ceph): mailmap: add João Eduardo Luís fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 17e0a5eb (ceph): mailmap: add Noah Watkins fixes for Reviewed-by:
- Also acknowledge noah.watkins@inktank.com as the primary email.
Signed-off-by: Loic Dachary <loic@dachary.org> - 12:49 PM Revision 3fe5f117 (ceph): mailmap: add Gary Lowell fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 62fbcacd (ceph): mailmap: add Loic Dachary fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision 2de21486 (ceph): mailmap: add Yan Zheng fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:49 PM Revision f6b52fd4 (ceph): mailmap: add Sage Weil fixes for Reviewed-by:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:31 PM Revision ca9acb96 (ceph): mon: implement 'mds set max_mds|max_file_size'
- First, add the ability to modify max_file_size. While we are at it, move
to a more sensible interface for adjusting ... - 10:59 AM Bug #7121: nightlies: failed to become clean before timeout expired due to degraded pg
- crush map rule
# rules
rule replicated_ruleset {
ruleset 0
type replicated
min_size 1
... - 10:45 AM Documentation #6234 (Resolved): all our new-user paths need to document name-resolution restrictions
- http://ceph.com/docs/master/start/quick-start-preflight/
- 09:08 AM Revision 45418e97 (ceph): Merge pull request #1056 from apeters1971/wip-jerasure-vectorop
- EC-JERASURE: rewrite region-xor function using vector operations
Reviewed-by: Loic Dachary <loic@dachary.org> - 09:06 AM Revision 5893469b (ceph): Merge pull request #1057 from dachary/wip-buffer-aligned
- erasure-code: ensure that coding chunks are page aligned
Reviewed-by: Andreas Peters <andreas.joachim.peters@cern.ch> - 01:35 AM Revision 1b4368b3 (ceph): schedule_suite: 2x replication for ceph-deploy
- This avoids having to update all of the tests.
Signed-off-by: Sage Weil <sage@inktank.com> - 01:09 AM Revision 69a99e6e (ceph): osd: ignore OSDMap messages while we are initializing
- The mon may occasionally send OSDMap messages to random OSDs, but is not
very descriminating in that we may not have ... - 01:09 AM Revision 1545ddbd (ceph): mon: only send messages to current OSDs
- When choosing a random OSD to send a message to, verify not only that
the OSD id is up but that the session is for th... - 01:08 AM Revision 741c64ec (ceph): osd: ignore OSDMap messages while we are initializing
- The mon may occasionally send OSDMap messages to random OSDs, but is not
very descriminating in that we may not have ... - 01:07 AM Revision 4fae5f84 (ceph): mon: only send messages to current OSDs
- When choosing a random OSD to send a message to, verify not only that
the OSD id is up but that the session is for th... - 01:05 AM Revision 7a89c678 (ceph): Merge pull request #1045 from ceph/wip-7093
- fix for #7093
Reviewed-by: Samuel Just <sam.just@inktank.com> - 12:44 AM rgw Bug #6462: Cannot update object HTTP headers using swift API
- I noticed that Content-Encoding has the same problem.
- 12:37 AM rgw Bug #6462: Cannot update object HTTP headers using swift API
- Using Emperor, I'm still having a hard time getting POST (as well as copy via PUT) to update or remove Content-Dispos...
- 12:33 AM Revision 695255ec (ceph): osd: Interim working version with backfill reserve state changes
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:33 AM Revision 0940d8fa (ceph): osd: Change waiting_on_backfill to a set<int>
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:33 AM Revision f9a11ef2 (ceph): osd: Determine if compatibility is required due to back level OSDs
- fixes: #7039
Signed-off-by: David Zafman <david.zafman@inktank.com> - 12:33 AM Revision 7837490a (ceph): osd: Fix problems in ReplicatedPG::do_op() logic
- Fix assert(is_degraded_object(soid)) in ReplicatedPG::wait_for_degraded_object()
Use last_backfill_started as the b... - 12:33 AM Revision f89d4105 (ceph): osd: Return false for func that returns bool
- Signed-off-by: David Zafman <david.zafman@inktank.com>
- 12:33 AM Revision 25a798ed (ceph): messages, os, osd: Clean-up g/hobject_t MAX handling
- Add check of is_max() before updating an unset pool
Use efficient is_max() instead of compare with hobject::get_max()... - 12:33 AM Revision a07d6820 (ceph): osd: Recovery reservations need to include backfill peers
- Caused by 86e4fd4f7a9848511e1bf9ae433fcd3a35a0818f
Signed-off-by: David Zafman <david.zafman@inktank.com>
01/08/2014
- 10:53 PM Revision ba19006d (ceph): osd: Config OSD when CEPH_FEATURE_OSD_ERASURE_CODES feature first seen
- Set the OSD superblock CEPH_OSD_FEATURE_INCOMPAT_SHARDS and
notify the filestore through set_allow_sharded_objects() ... - 10:38 PM Revision 09d021b4 (ceph): doc: Updated for hostname -s and resolving to non-loopback IP address.
- fixes: #6234
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 10:37 PM Revision ed605f0d (ceph): doc: Fixed hyperlink.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:26 PM Revision b780f4bf (ceph): osd: Remove redundant incompat feature
- We can remove this CompatSet bit without worry because the only
way it could have been set is if an erasure coded poo... - 09:47 PM Bug #7121 (Resolved): nightlies: failed to become clean before timeout expired due to degraded pg
This issue has been seen before on master branch as well as testing of wip-5858-rebase testing.
teuthology:/a/te...- 09:45 PM Bug #7120 (Duplicate): osd: EEXIST on mkcoll on dumpling
- ...
- 08:11 PM Revision 18b158e3 (ceph): add another host so that replicas meet the minimum required
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
- 07:41 PM Bug #7093 (Resolved): osd: peering can send messages prior to auth
- 08:02 AM Bug #7093: osd: peering can send messages prior to auth
- Patches look good to me.
- 07:14 PM Revision 294cdfb9 (ceph): erasure-code: relax zero copy requirements in tests
- The input buffer is copied when padding is required. Remove the tests
that ensured the input buffer was not copied en... - 07:13 PM Revision 656de1b6 (ceph): erasure-code: ensure that coding chunks are page aligned
- When coding chunks are allocated for jerasure, their address must be
aligned to page boundaries. The requirement is a... - 05:57 PM CephFS Bug #6914 (Resolved): Accessing specific file on CephFS causes client hang
- fixed by f292992b82d042a9f738e4698802e3b205b82c83
- 02:54 PM rgw Bug #6830: S3 CompleteMultipartUploadResult has empty ETag element
- Still happens on RHEL 6.5 with ceph-0.72.2-0.el6.x86_64.
- 02:15 PM Revision 9ba65996 (ceph): EC-JERASURE: rewrite region-xor function using vector operations to get...
- 11:13 AM Revision b6adf597 (ceph): Merge pull request #1055 from kri5/master
- mailmap: Adds Christophe Courtaut
git shortlog -nes
shows only one Christophe Courtaut proof that the .mailmap ... - 10:46 AM Revision 31f3745a (ceph): mailmap: Adds Christophe Courtaut
- Signed-off-by: Christophe Courtaut <christophe.courtaut@gmail.com>
- 10:26 AM Revision 65a15039 (ceph): Merge pull request #1052 from dachary/wip-large-output-daemon-config
- common: fix large output in unittest_daemon_config
Reviewed-By: Christophe Courtaut <christophe.courtaut@gmail.com> - 09:40 AM Bug #7117: centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_common.sh'
- I replicated this problem locally using the most current `master` branch. The issue is not really `gatherkeys` but th...
- 05:54 AM Bug #7117 (Resolved): centOS: init script silently dies due to missing '/usr/lib64/ceph/ceph_comm...
One of the failing tests: http://qa-proxy.ceph.com/teuthology/teuthology-2014-01-05_01:10:06-ceph-deploy-master-t...- 09:36 AM rgw Bug #7110: rgw: cannot set acls on pre-dumpling buckets
- 09:26 AM Feature #7114: Hinted recovery
- Instead of "a defined bucket in the CRUSH hierarchy", it probably makes more sense to say uses the nearest common anc...
- 07:57 AM Revision 6eea3133 (ceph): Merge pull request #1044 from ceph/port/onexit
- create on_exit replacement
Reviewed-by: Sage Weil <sage@inktank.com> - 01:11 AM Revision 7d5674cf (ceph): doc: Added librados introduction doc. Still wip.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 01:11 AM Revision 33e78a4a (ceph): doc: Updated terminology. Added librados intro to index.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
01/07/2014
- 11:09 PM Revision c4b9dae4 (ceph): Merge pull request #172 from ceph/wip-teuthvmcon-wusui
- Fix a bug where ctx.config['targets'] was looped through again
- 10:22 PM Revision f32adf05 (ceph): Fix a bug where ctx.config['targets'] was looped through again
- in connect(). The bug caused vm behavior to happen for a
target if any of the machines in the cluster was a vm. The... - 07:23 PM Revision 6fac2acc (ceph): Merge pull request #1050 from ceph/wip-7083
- radosgw-admin: fix object policy read op
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 07:14 PM Revision 7acb0a17 (ceph): on_exit: remove side effects from asserts
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 05:28 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- Try the for-stable-3.10.24 branch of https://github.com/ceph/ceph-client.git
- 05:11 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- Where can I find the code to build against kernel-devel-3.10.0-54.0.1.el7 ?
- 05:10 PM Bug #7116 (Resolved): mon: pg_temp left behind if pool deleted while pg_temp exist
I bumped the pg_num/pgp_num for pool 3 and then deleted pool 3 quickly afterwards:
./ceph osd dump
....
pg_tem...- 03:51 PM Revision e3d0b0ae (ceph): common: fix large output in unittest_daemon_config
- All tests in daemon_config use the global g_ceph_context
object. Introducing an expansion loop in it will impact all ... - 03:44 PM Subtask #6895 (Fix Under Review): omap needs to be disablable on a per-pool basis, this needs to ...
- "work in progress":https://github.com/ceph/ceph/pull/1053
- 10:18 AM Subtask #6895: omap needs to be disablable on a per-pool basis, this needs to be required to cach...
- * add a can_omap() predicate in "pg_pool_t":https://github.com/ceph/ceph/blob/5300ed942809c963ccbe4fc1f90e7767f7b74b5...
- 07:07 AM Subtask #6895 (In Progress): omap needs to be disablable on a per-pool basis, this needs to be re...
- 03:04 PM Feature #7114 (New): Hinted recovery
- In the case where you are constructing multi-site RADOS object stores, where inter-site connectivity is a premium, it...
- 01:29 PM rgw Bug #7110: rgw: cannot set acls on pre-dumpling buckets
- The fix in wip-7110-dumpling looks good to me.
- 10:26 AM Subtask #6888: EC/Tiering: Disallow omap writes on pools which are backed by EC pools
- It should be able to; this will be stuff in the caching and tiering code to look at the properties on the backing poo...
- 06:54 AM Subtask #6888 (Need More Info): EC/Tiering: Disallow omap writes on pools which are backed by EC ...
- Is it possible to implement this before implementing ErasureCodePG ?
- 07:04 AM Subtask #7048 (Resolved): mon: erasure crush rule vs pool process
- 07:01 AM Subtask #6027 (Resolved): ensure that erasure coded pools don't work until the osds can handle it
- 07:00 AM Revision 5300ed94 (ceph): Merge pull request #1043 from ceph/port/misc
- misc portability patches
Reviewed-by: Sage Weil <sage@inktank.com> - 06:59 AM Subtask #6119 (Won't Fix): replace PG::object_contexts with SharedPtrRegistry
- This is obsoleted by the work done on PGBackend.
- 06:54 AM Documentation #6465 (In Progress): admin/build-doc should have some kind of build check for broke...
- I pushed some changes to a wip-6465 branch but it is failing with what looks like unrelated problems.
This is a li... - 06:52 AM Subtask #6434 (Need More Info): review and test PGBackend
- 06:29 AM Revision 4eeee0e6 (ceph): Merge pull request #1051 from kdreyer-inktank/automake-1.12
- configure: support Automake 1.12
- 05:30 AM Revision ffc9d72d (ceph): Merge pull request #1046 from dachary/wip-7103
- common: recursive implementation of config::expand_meta …
Reviewed-by: Sage Weil <sage@inktank.com> - 05:26 AM Revision 822ad588 (ceph): configure: support Automake 1.12
- Signed-off-by: Jonathan Dieter <jdieter@lesbg.com>
Signed-off-by: Ken Dreyer <ken.dreyer@inktank.com> - 04:02 AM Revision 12a0f513 (ceph): Merge pull request #1047 from dachary/wip-duplicated-headers
- autoconf: trim duplicated headers
01/06/2014
- 10:30 PM rgw Bug #7110: rgw: cannot set acls on pre-dumpling buckets
- Buckets created on versions prior to Dumpling cannot have ACLs set, even after upgrading to Dumpling.
- 08:28 PM rgw Bug #7110 (Won't Fix): rgw: cannot set acls on pre-dumpling buckets
- 06:28 PM rgw Bug #7110 (Resolved): rgw: cannot set acls on pre-dumpling buckets
- 09:10 PM Revision 0d5763eb (ceph): add missing/required OSDs to the rest of ceph-deploy tasks
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit 2e721f0d7c5f344168d70959bb1478a0f33... - 09:05 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- RPM kmods are a special kind of pain. Red Hat has a kABI, and in theory if you rely on that, you can re-use your kern...
- 08:07 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- Ken Dreyer wrote:
> Do we have some form of kernel module packages available already? Or will these be developed fro... - 08:00 PM devops Feature #6986: Build packages containing RBD and CephFS kernel modules for installation on RHEL7-...
- Do we have some form of kernel module packages available already? Or will these be developed from scratch?
- 08:53 PM Revision b1976dd0 (ceph): radosgw-admin: fix object policy read op
- Fixes: #7083
This was broken when we fixed #6940. We use the same function to both
read the bucket policy and the obj... - 08:25 PM Revision 0c9129bf (ceph): error after 15 minutes of waiting for gatherkeys
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit ad18717505c407f2c8bb995af8fd516b619... - 08:18 PM devops Bug #6746: ceph-release rpm not playing well with yum-plugin-priorities
- Hi John, What priority value would work for you?
- 07:35 PM Feature #7108: build: error out on inappropriate use of dout()
- alternatively, do not define dout or g_ceph_context at all for library code.
- 10:38 AM Feature #7108 (New): build: error out on inappropriate use of dout()
- Right now, if you use dout() instead of ldout() in a library context, the compilation phase goes fine and then you ge...
- 06:31 PM Feature #7039 (Fix Under Review): Check other OSD CEPH_FEATURE_OSD_ERASURE_CODES to enable acting...
- Part of review of wip-5858-rebase
- 06:28 PM Bug #7014 (Can't reproduce): rados: stuck degraded, possibly related to acting_backfill changes
- This might have been fixed by fix for #6905 which is to increase the timeout in suites/rados/thrash/thrashers/mapgap....
- 06:18 PM rgw Bug #7073: "rgw gc max objs" should have a prime number as default value
- Hi Yehuda,
Thanks for the update.
Should I provide a pull request for this or you will fix it at your side?
Th... - 01:13 PM rgw Bug #7073: "rgw gc max objs" should have a prime number as default value
- Right. We can just do a mod of some other larger prime, and then mod 32 (or whatever is configured) so that we don't ...
- 05:08 PM Revision 5c049663 (ceph): Enable reporting of entire runs as dead
- 03:05 PM Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap
- There was previously no "w-only" permission supported, indeed. I'm not sure it makes sense even for allow new snaps....
- 10:00 AM Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap
- Ah, I did the allow_new_snaps command. I'm not very familiar with the REST api, but I think it needs to be able to ha...
- 02:18 PM Subtask #6888 (In Progress): EC/Tiering: Disallow omap writes on pools which are backed by EC pools
- 12:59 PM rgw Bug #7083 (Fix Under Review): rgw: public-read acl test failing on master
- 12:47 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- First tested the kernel rebuild (-j 8 for the make command) twice, this went fine. (Except for performance: 1 kernel ...
- 11:24 AM CephFS Fix #7109 (Resolved): disallow removing the first data pool
- Our lookup-by-ino code requires that we be able to look for all existing inodes in a well-known location — the defaul...
- 10:35 AM Documentation #6234: all our new-user paths need to document name-resolution restrictions
- This is also missing from the new pre-flight checklist: http://ceph.com/docs/master/start/quick-start-preflight/
T... - 09:36 AM Bug #6824: Removal of an OSD that is not down should set non-successful status code
- I missed this when it was going by, but it looks like that patch makes things like "osd down" and "osd out" no longer...
- 09:26 AM CephFS Feature #3866: mds: Add lazily-updated backtraces to hard links
- If we create a hardline that references the "real" location of the file (as a path), and then the "real" location mov...
- 06:43 AM devops Bug #7107 (Resolved): bootstrap script should not install requirements
- Opened pull request: https://github.com/ceph/ceph-deploy/pull/148
Merged to master with hash: 07f903c - 06:40 AM devops Bug #7107 (Resolved): bootstrap script should not install requirements
- There is no need to install from requirements file anymore. All libraries defined there are only for
development onl...
01/05/2014
- 03:25 PM Bug #7072: logrotate not sending HUP signal
- Looking at the logrotate source code it would seem that logrotate depends on whatever is defined in postrotate to per...
- 01:06 PM Bug #7072: logrotate not sending HUP signal
- I can confirm this on Ubuntu 13.04.
If you check with lsof, you can see that the radosgw + ceph-osd files are move... - 12:52 PM Bug #7072: logrotate not sending HUP signal
- Distributor ID: Ubuntu
Description: Ubuntu 12.04 LTS
Release: 12.04
Codename: precise
logrotate version: 3.7.8
... - 03:11 PM Revision 5bb1545e (ceph): common: multiple variable occurence expansion test
- http://tracker.ceph.com/issues/7103 refs #7103
Signed-off-by: Loic Dachary <loic@dachary.org> - 03:11 PM Revision 87db534c (ceph): common: cosmetic inversion of tests arguments
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:11 PM Revision 94854090 (ceph): common: recursive implementation of config::expand_meta
- Using a recursive implementation of variable expansions make it easier
to protect against loops and provide human rea... - 03:11 PM Revision 3f34dc7d (ceph): common: unit tests for config::expand_meta
- Part of the config.cc tests are in test/confutils.cc but they do not
cover meta variable expansion. Create unittest_c... - 09:56 AM Bug #7103 (Fix Under Review): config: variable substitution happen only once
- "proposed fix":https://github.com/ceph/ceph/pull/1046
- 06:44 AM Revision 98ed9ac5 (ceph): mon: only send messages to current OSDs
- When choosing a random OSD to send a message to, verify not only that
the OSD id is up but that the session is for th... - 06:44 AM Revision 35da8f9d (ceph): osd: do not send peering messages during init
- Do not send any peering messages while we are still working our way
through init().
Fixes: #7093
Signed-off-by: Sage... - 06:44 AM Revision f68de9f3 (ceph): osd: ignore OSDMap messages while we are initializing
- The mon may occasionally send OSDMap messages to random OSDs, but is not
very descriminating in that we may not have ...
01/04/2014
- 10:46 PM Bug #7093 (Fix Under Review): osd: peering can send messages prior to auth
- 09:03 AM Bug #7093: osd: peering can send messages prior to auth
- several instances in teuthology-2014-01-02_19:40:02-upgrade:parallel-master-testing-basic-plana
- 08:54 PM Revision 2181d25f (ceph): onexit: add an on exit callback utility
- Adds a class that executes registered callbacks in its destructor. Since
static duration objects have their destructo... - 08:54 PM Revision 7aa98052 (ceph): log: use on exit manager to flush logs on exit
- on_exit is not portable, and atexit doesn't allow parameters to be
passed to callbacks.
Signed-off-by: Noah Watkins ... - 05:18 PM Revision ef6a56aa (ceph): keyutils: handle non-linux platform
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 05:18 PM Revision 26c07d97 (ceph): ipaddr: use IN6_ARE_ADDR_EQUAL for comparison
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 05:18 PM Revision 6342d051 (ceph): pipe: handle missing MSG_MORE and MSG_NOSIGNAL
- On OSX (and currently any platform missing the MSG_MORE
macro) the MSG_MORE optimization is disabled. The MSG_NOSIGNA... - 05:18 PM Revision 3d19f7c8 (ceph): test: disable cross process sem tests on non-Linux
- How to make this portable:
- MAP_ANONYMOUS -> MAP_ANON (OSX)
- sem_init (anonymous semaphore) needs to be replaced ... - 07:18 AM Cleanup #7105: There are three different ways to retrieve an authentication key
- Ahh.. forgot the ...
- 07:16 AM Cleanup #7105 (Closed): There are three different ways to retrieve an authentication key
- Is it necessary to have three different commands that do the exact same thing ?
Reference:
https://github.com/cep... - 06:50 AM Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap
- Okay, I'm starting to understand how it works..
This is how the mapping is done between the API and the commands a... - 06:32 AM Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap
- I left out a line at the end of the error, the complete error is:...
- 06:13 AM Feature #7104 (New): rest-api: support commands requiring 'w' cap without 'rw' cap
- Right now the commands must be one of "r", "rw", "rwx". We should support each of those as capabilities, not as named...
- 02:05 AM Bug #7103 (In Progress): config: variable substitution happen only once
- 02:04 AM Bug #7103: config: variable substitution happen only once
- The "test demonstrating the problem":https://github.com/dachary/ceph/commit/84374f3d60b3be37f74bbf65b4c5568f301e2832 ...
- 02:00 AM Bug #7103 (Resolved): config: variable substitution happen only once
- In ceph.conf if...
- 12:58 AM Revision 6663a5ee (ceph): Merge pull request #1038 from ceph/wip-objectcacher-backoff
- osdc/ObjectCacher: back off less during flush
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 12:30 AM Revision ab2b2bb0 (ceph): doc: Fixed typo.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:05 AM Revision a231f220 (ceph): Merge pull request #1041 from dachary/wip-mon-test
- mon: get rid of --keyring /dev/null hack
- 12:05 AM Revision 81d71f9d (ceph): Merge pull request #1042 from kdreyer-inktank/gitignore-autotools-test-...
- add autotools-generated files to .gitignore
01/03/2014
- 11:47 PM Revision bb8b7503 (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... - 11:34 PM Revision f2e33e81 (ceph): mon: get rid of --keyring /dev/null hack
- No longer necessary when --auth-supported=none
Signed-off-by: Loic Dachary <loic@dachary.org> - 11:00 PM Revision 3a42d411 (ceph): Merge pull request #1026 from dmsimard/ceph_rest_api_discovery
- Be more explicit how to discover available API calls
Reviewed-by: Dan Mick <dan.mick@inktank.com> - 10:37 PM Revision d110c913 (ceph): doc: Removed dash from --dmcrypt option.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:00 PM Revision 5de7e562 (ceph): Merge pull request #1040 from dachary/wip-ceph-disk-activate-data-dir
- ceph-disk: fix false positive for gitbuilder
- 09:52 PM Revision c772b6d1 (ceph): ceph-disk: fix false positive for gitbuilder
- The output of test/ceph-disk.sh is very verbose which is good for
debugging errors. However it sometime contains stri... - 09:45 PM Revision 3cffea49 (ceph): Re-raise exceptions caught in the watchdog
- 09:08 PM Revision 027929e0 (ceph): Use response.text if response.json is None
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 09:01 PM Revision f92174ff (ceph): Strip stdout lines
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:56 PM Revision 68b259fd (ceph): Catch and log unhandled exceptions in the watchdog
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 08:51 PM Revision e2ee5287 (ceph): osdc/ObjectCacher: back off less during flush
- In cce990efc8f2a58c8d0fa11c234ddf2242b1b856 we added a limit to avoid
holding the lock for too long. However, if we ... - 08:45 PM Revision c6a9de04 (ceph): Add 'emperor' to list of branches with reporting
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 07:41 PM Revision 31210787 (ceph): Merge pull request #1032 from dachary/wip-ceph-disk-activate-data-dir
- ceph-disk : fixes and improvements for the --data-dir code path and associated tests
Reviewed-by: Sage Weil <sage@in... - 06:41 PM Revision eaa3e124 (ceph): Work around a change in pip 1.5 regarding wheels
- The error message was "pip's wheel support requires setuptools >= 0.8
for dist-info support."
Signed-off-by: Zack Ce... - 06:23 PM Revision daefe815 (ceph): Be more explicit how to discover available API calls
- Signed-off-by: David Moreau Simard <dmsimard@iweb.com>
- 05:55 PM Revision d3afebe1 (ceph): Be safer when calling ./bootstrap
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 04:13 PM rgw Bug #7099: Strange Comportments with media files
- MP3 Test : http://p1.13h.com/swift/v1/test/touslesmemes.mp3
- 03:54 PM rgw Bug #7099: Strange Comportments with media files
- I'm using Ubuntu 12.04 LTS
- 03:07 PM rgw Bug #7099 (Resolved): Strange Comportments with media files
- Hello
I'm using ceph 0.72.2 and I have a problem to get media file (mp3 ou mp4 video) with RadosGw and swift emu... - 04:06 PM Bug #7052 (Resolved): saucy make check doesn't clean up
- 03:53 PM Bug #7052 (Fix Under Review): saucy make check doesn't clean up
- Looks like a Automake 1.13 feature with parallel tests. As discussed in the Google+ hangout today, I've added these t...
- 03:33 PM Revision de8522fa (ceph): ceph-disk: tests for the --data-dir code path
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision 306b099a (ceph): ceph-disk: create the data directory if it does not exist
- Instead of failing if the OSD data directory does not exist, create
it. Only do so if the data directory is not enfor... - 03:30 PM Revision 51ee3c04 (ceph): ceph-disk: add copyright notice
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision ca713f48 (ceph): ceph-disk: implement --statedir as /var/lib/ceph
- Replace hardcoded /var/lib/ceph with the STATEDIR global variable and
implement the --statedir option to override the... - 03:30 PM Revision a71025d3 (ceph): ceph-disk: implement --sysconfdir as /etc/ceph
- Replace hardcoded /etc/ceph with the SYSCONFDIR global variable and
implement the --sysconfdir option to override the... - 03:30 PM Revision b82ccfbf (ceph): ceph-disk: fix Error() messages formatting
- Mainly using % instead of ,
Signed-off-by: Loic Dachary <loic@dachary.org> - 03:30 PM Revision de005059 (ceph): ceph-disk: remove noop try:
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision 324804a8 (ceph): ceph-disk: fix activate() indent
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision 55ca7bb2 (ceph): ceph-disk: do not hide main_activate() exceptions
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision 908348b8 (ceph): ceph-disk: make exception handling terse in main_activate_journal
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 03:30 PM Revision ad6b4b4b (ceph): ceph-disk: add --prepend-to-path to control execution
- /usr/bin is hardcoded in front of some ceph programs which makes it
impossible to control where they are located via ... - 03:30 PM Revision 2b935bbf (ceph): ceph-disk: which() uses PATH first
- Instead of relying on a hardcoded set of if paths. Although this has the
potential of changing the location of the bi... - 03:30 PM Revision d09af0fa (ceph): ceph-disk: display the command output on OSD creation failure
- The string form of a CalledProcessError instance does not include the
output datamember. Add it to the Error exceptio... - 03:30 PM Revision 97f516a1 (ceph): ceph-disk: use CalledProcessError.returncode
- CalledProcessError has no errno data member
Signed-off-by: Loic Dachary <loic@dachary.org> - 03:30 PM Revision b65eb377 (ceph): ceph-disk: fsid is a known configuration option
- Use get_conf_with_default instead of get_conf because fsid is a known
ceph configuration option. It allows overriding... - 03:30 PM Revision e773b68f (ceph): ceph-disk: implement --mark-init=none
- It is meant to be used when preparing and activating a directory that is
not to be used with init. No file is created... - 03:30 PM Revision 0fcc0818 (ceph): ceph-disk: run ceph-osd when --mark-init=none
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 01:06 PM rgw Feature #6513 (Fix Under Review): rgw: dr: Service scripts for meta/data sync agents
- 12:53 PM rgw Feature #6513: rgw: dr: Service scripts for meta/data sync agents
- Work in progress is wip-6513 branch in the radosgw-agent tree.
- 01:03 PM devops Fix #7095 (Resolved): fix f19 gitbuilder
- 01:03 PM devops Feature #7094 (Resolved): f20 gitbuilder
- 01:00 PM devops Feature #6020: radosgw-apache opinionated package
- Build job for the package is radosgw-apache on jenkins.
- 01:00 PM devops Feature #6020: radosgw-apache opinionated package
- Work in progress is in the master branch in the radosgw-apache tree.
- 11:45 AM Bug #7093 (Resolved): osd: peering can send messages prior to auth
- we are still authenticating:...
- 10:56 AM rgw Bug #7084: segv in RGWDataChangesLog::RGWDataChangesLog
- recent log: ubuntu@teuthology:/a/sage-2014-01-02_12:58:22-upgrade:parallel-next-testing-basic-plana/22332
- 07:38 AM Bug #7085 (Resolved): ceph-disk cli test fails if /var/lib/ceph is missing
- The corresponding test file been moved to "ceph-disk.sh":https://github.com/dachary/ceph/commit/de8522fa6c82768da1d61...
- 02:30 AM Revision 131b2184 (ceph): Use CentOS Gitbuilder sha1 instead of Fedora for non-ubuntu.
- Signed-off-by: Sandon Van Ness <sandon@inktank.com>
- 01:56 AM Revision 830583f8 (ceph): osd: Correction to #6690 change
- Reorder sequence so min_size check is right
Remove redundant resize of backfill
Correction to 19cff890eb6083eefdb7b7... - 12:52 AM Revision 6b8d4188 (ceph): init-ceph: pass config file path when adjust crush position
- Fixes: #6591
Reported-by: Sean Noonan <sean.noonan@twosigma.com>
Signed-off-by: Sage Weil <sage@inktank.com>
01/02/2014
- 10:09 PM Revision 806c0cb3 (ceph): Merge pull request #1012 from dachary/wip-mon-mkfs
- unit tests for ceph-mon --mkfs
- 10:07 PM Revision 53a4a88f (ceph): Merge pull request #1029 from ceph/wip-6914
- Wip 6914
Reviewed-by: Sage Weil <sage@inktank.com> - 09:52 PM Revision 1f47a5b1 (ceph): Merge pull request #1035 from ceph/port/bitsandints
- endianess and integer types
Reviewed-by: Sage Weil <sage@inktank.com> - 09:49 PM Revision 4843f89e (ceph): Merge pull request #1036 from dachary/wip-7085
- ceph-disk: cannot run unit tests
- 09:45 PM Revision 2ba6930d (ceph): ceph-disk: cannot run unit tests
- Because ceph-disk relies on hardcoded paths. The corresponding test will
be added back when ceph-disk can run from so... - 09:35 PM Revision 24417f99 (ceph): test/cli-integration/rbd: silence stderr
- Signed-off-by: Sage Weil <sage@inktank.com>
- 08:48 PM Revision d88ec89a (ceph): break out of the while loop after 15 minutes
- Signed-off-by: Alfredo Deza <alfredo.deza@inktank.com>
(cherry picked from commit bef6eb74dcaa37b70b1eab4d28bfa10abb0... - 07:09 PM Revision 4c0086ed (ceph): upgrade/parallel/stress-split: api tests needs to run on dumpling host
- Even tho we run the dumpilng branch test.sh, the installed
ceph_test_rados_api_* are still emperor. Add another role... - 07:03 PM Revision 82205493 (ceph): inttypes: detect and define missing integer types
- Working around missing integer types is pretty easy. For example, the
__u32 family are Linux-specific types, and usin... - 07:03 PM Revision 8f91cace (ceph): endian: check byte order on OSX
- Defines CEPH_[BIG|LITTLE]_ENDIAN to use, as BYTE_ORDER etc macros may be
different names on different platforms.
Sig... - 06:06 PM Revision 4d88dd10 (ceph): osd: discriminate based on connection messenger, not peer type
- Replace ->get_source().is_osd() checks and instead see if it is the
cluster_messenger so that we do not confuse ourse... - 05:27 PM Revision 39d6f281 (ceph): upgrade/parallel/stress-split: use dumpling tests against mixed d+e
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:52 PM devops Bug #6591 (Resolved): init script does not pass alternate configuration file to osd crush create-...
- 04:36 PM Revision 46a5674b (ceph): doc/release-notes: fix bobtail version
- The tunables were mostly supported by later argonaut point releases,
but not officially. Let's stick with bobtail th... - 02:56 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- The RPM's built fine now. I upgraded all nodes in the ceph cluster.
- 07:29 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- error: Installed (but unpackaged) file(s) found:
/usr/lib64/erasure-code/libec_example.so
/usr/lib64/erasure-... - 02:10 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- I'm currently building the RPM's using the mentioned ceph.spec which includes the patch.
- 01:22 PM Bug #7085 (Resolved): ceph-disk cli test fails if /var/lib/ceph is missing
- if the ceph package isn't installed locally, the ceph-disk cli test fails.
probably need to pass ceph-disk an al... - 01:09 PM rgw Bug #7084 (Can't reproduce): segv in RGWDataChangesLog::RGWDataChangesLog
- ...
- 12:40 PM rgw Bug #7083 (Resolved): rgw: public-read acl test failing on master
- This test fails:...
- 10:28 AM Bug #6810 (Can't reproduce): very high monitor memory usage after upgrade dumpling -> emperor
- 10:24 AM Subtask #5858 (Fix Under Review): Backfill should be able to handle multiple backfill peers
- 09:35 AM Bug #7061 (Resolved): valgrind: mon and osd leaks in libnss on master
- 09:33 AM Bug #7072 (Need More Info): logrotate not sending HUP signal
- what distribution is this?
- 09:21 AM Revision f6bbcf43 (ceph): mon: tests for ceph-mon --mkfs
- * auth none must not require a keyring
* --key can be used as an alternative to --keyring
* --mkfs is idempotent
* th... - 09:17 AM Bug #7060 (In Progress): mon: osdmap scrub crc mismatch
- 02:48 AM Revision 40018fcc (ceph): Merge pull request #1027 from jdurgin/wip-unprotect-error
- librbd: better error when unprotect fails on unprotected snap
Reviewed-by: Sage Weil <sage@inktank.com> - 02:47 AM Revision 4fe0b1bb (ceph): Merge pull request #1010 from dachary/wip-mon-keyring
- implement --key as an alternative to --keyring when ceph-mon --mkfs
Reviewed-by: Sage Weil <sage@inktank.com> - 02:04 AM Revision e74ab284 (ceph): Merge pull request #1031 from dachary/wip-parse-args
- CEPH_ARGS should trim whitespaces
Reviewed-by: Sage Weil <sage@inktank.com> - 02:03 AM Revision af27f070 (ceph): Merge pull request #1030 from dachary/wip-crush-location
- osdc: = is not a delimiter in --crush-location
Reviewed-by: Sage Weil <sage@inktank.com> - 02:00 AM Revision 906f973e (ceph): Merge pull request #1020 from dachary/wip-ceph-disk-override
- ceph-disk: prepare --data-dir must not override files
Reviewed-by: Sage Weil <sage@inktank.com> - 01:58 AM Revision 41987db8 (ceph): doc/release-notes: v0.74
- Signed-off-by: Sage Weil <sage@inktank.com>
- 01:03 AM Revision 43465d47 (ceph): Merge pull request #950 from ceph/wip-pg-stat
- update pg stats when they are queried; share peer_info on query too
Reviewed-by: Loic Dachary <loic@dachary.org>
01/01/2014
- 08:23 PM Revision 229740f4 (ceph): ceph-conf: display arguments when an option cannot be parsed
- unable to parse option: ''
Is difficult to figure out. It is much better if shown in the context in
which it was fou... - 08:23 PM Revision 07f53994 (ceph): common: CEPH_ARGS should trim whitespaces
- CEPH_ARGS when parsed by env_to_vec did not trim trailing and leading
whitespaces: they would unexpectedly be parsed ... - 07:41 PM Revision b729ed64 (ceph): Merge pull request #1028 from ceph/wip-empty-rbd-ls
- rbd: return 0 and an empty list when pool is entirely empty
Reviewed-by: Sage Weil <sage@inktank.com> - 05:05 PM Bug #6988 (Resolved): pg query doesn't show current 'problem' stats
- 04:45 PM Bug #6988: pg query doesn't show current 'problem' stats
- "work in progress":https://github.com/ceph/ceph/pull/950
- 03:18 PM Revision 40059e12 (ceph): Merge pull request #1025 from pyKun/fix_typos
- Fix typos in erasure code documents
Reviewed-by: Loic Dachary <loic@dachary.org> - 02:32 PM Revision 2b848a91 (ceph): osdc: = is not a delimiter in --crush-location
- When parsing the crush location
--crush-location 'root=default host=hostA'
it must create the vector
[ '... - 10:48 AM Revision 7dfe550c (ceph): ceph-disk: prepare --data-dir must not override files
- ceph-disk does nothing when given a device that is already prepared. If
given a directory that already contains a suc... - 10:07 AM Revision e946df13 (ceph): mon: do not use the keyring if auth = none
- The Monitor::is_keyring_required() predicate is defined to be used in
the mkfs code path and not require the keyring ... - 10:07 AM Revision ab6203f3 (ceph): mon: fix indentation
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 10:07 AM Revision 4c8a3134 (ceph): mon: implement --key for --mkfs
- Allow --key to be used as an alternative to --keyring when ceph-mon
--mkfs runs. The key is wrapped inline into
... - 07:53 AM Revision fe3fd5fb (ceph): Merge branch 'next'
- 03:19 AM Revision f12abbf6 (ceph): mds: don't allow changing layout for non-regular file
- layout is meaningless for non-regular file. using non-default layout
may make backtrace less efficiency.
Signed-off-... - 03:08 AM Revision f292992b (ceph): mds: always store backtrace xattr in the default pool
- when creating file in non-default pool, we need to store a backtrace
in the default pool in addition to the specified... - 01:21 AM Revision cae663af (ceph): osd/ReplicatedPG: improve debug output from check_local
- This will help with #6685
Signed-off-by: Sage Weil <sage@inktank.com> - 01:12 AM Revision ac15f1ad (ceph): Merge pull request #1011 from dachary/wip-mon-daemonize
- do not daemonize ceph-mon when not necessary
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 01:02 AM Revision ac547a5b (ceph): rbd: return 0 and an empty list when pool is entirely empty
- rbd_list will return -ENOENT when no rbd_directory object
exists. Handle this in the cli tool and interpret it as suc... - 12:26 AM Revision e91fb910 (ceph): librbd: better error when unprotect fails on unprotected snap
- This will show up on the command line and logs, making it more
clear than EINVAL.
Fixes #6851 and #4045
Signed-off-b...
12/31/2013
- 09:48 PM Revision 42e98aca (ceph): Be more explicit how to discover available API calls
- I believe it cannot hurt to be slightly more explicit about how
to retrieve API calls and provide examples.
Signed-o... - 08:25 PM Revision b4f524eb (ceph): Sleep once outside of the watchdog loop
- Hopefully this will prevent the double-posting of jobs.
Signed-off-by: Zack Cerza <zack.cerza@inktank.com> - 08:21 PM Revision 7d27793f (ceph): Merge pull request #1022 from ceph/wip-listomapvals
- fix rados listomapvals for >512 keys
Reviewed-by: Sage Weil <sage@inktank.com> - 07:44 PM Revision 04fe7271 (ceph): rados: add rados tool test
- For now this just tests listomapvals, but it may be expanded in the
future.
Signed-off-by: Josh Durgin <josh.durgin@... - 07:32 PM Revision e1d20d14 (ceph): Merge pull request #1009 from dachary/wip-mon-idempotent
- make ceph-mon --mkfs idempotent
Reviewed-by: Sage Weil <sage@inktank.com> - 07:06 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- previous patch doesn't work. please try the new one, thank you.
- 05:58 PM Revision f0ae4abd (ceph): mon: create mon-data directory on --mkfs
- If the mon-data directory does not exist when ceph-mon --mkfs runs, it
is created.
Signed-off-by: Loic Dachary <loic... - 05:58 PM Revision 1eafe8dc (ceph): mon: make ceph-mon --mkfs idempotent
- A mon is considered to exist if the mon-data directory exists and is not
empty. If ceph-mon --mkfs is run twice, it w... - 05:58 PM Revision eeba294f (ceph): mon: remove fixture directory between runs
- Instead of just removing the store.db directory, remove everything so
that --mkfs can re-create it instead of inferin... - 05:52 PM rbd Bug #6693 (Fix Under Review): "rbd ls" returns error if the pool empty
- https://github.com/ceph/ceph/pull/1028
- 05:42 PM rbd Bug #4045 (Fix Under Review): snap unprotect on a snapshot that is already unprotected throws ina...
- https://github.com/ceph/ceph/pull/1027
- 05:42 PM rbd Bug #6851 (Fix Under Review): rbd snap unprotect test failing
- https://github.com/ceph/ceph/pull/1027
- 05:40 PM rbd Bug #2689 (Resolved): qemu iozone test hangs
- Retested with qemu 1.5 from the havana cloud archive for ubuntu, and ceph 0.67.4. It worked fine. I'm not sure exactl...
- 04:18 PM Revision d77101cc (ceph): Merge pull request #1016 from bydsky/bugfix
- Fix Issue #6992: stop the accepter and mark all pipes down before rebind
Backport: emperor, dumpling
Reviewed-by: Sa... - 04:17 PM Revision 49a96d13 (ceph): Merge pull request #1024 from jdurgin/wip-7067
- osd_types: add missing osd op flags
Reviewed-by: Sage Weil <sage@inktank.com> - 01:53 PM Revision fc815f75 (ceph): Merge pull request #168 from ktdreyer/readme-formatting
- format bullets in README
- 12:47 PM Revision d4438410 (ceph): mon: add flags argument to common_init_finish
- It is the same flag that is given to common_preinit. The service thread
is not initialized if CINIT_FLAG_NO_DAEMON_AC... - 12:46 PM Revision 10aa220a (ceph): mon: set CINIT_FLAG_NO_DAEMON_ACTIONS when appropriate
- --mkfs, --inject_monmap and --extract-monmap are no daemon actions.
Signed-off-by: Loic Dachary <loic@dachary.org> - 12:46 PM Revision a1945135 (ceph): mon: do not daemonize if CINIT_FLAG_NO_DAEMON_ACTIONS
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:46 PM Revision 403f777b (ceph): mon: fix indentation
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:34 PM rbd Bug #7067 (Resolved): rbd read-flags.sh test is failing
- commit:2b0a435920daab8de27402e00b58c6b00ccb5663
- 11:17 AM Revision e0bae952 (ceph): Fix typos in erasure code documents
- Signed-off-by: Kun Huang <academicgareth@gmail.com>
- 11:10 AM rbd Bug #7076 (Resolved): rbd: check for watchers before removing an image
- ~/ceph/src$ ./rbd showmapped
id pool image snap device
0 rbd img2 - /dev/rbd0
~/ceph/src$ ./rbd rm img... - 08:18 AM Bug #6992 (Pending Backport): OSD assert fails after it found it was marked as down by monitor du...
- 07:30 AM devops Bug #6503 (Rejected): ceph-deploy: nosetests fail
- Actually, I thought there was some mechanism in ceph-deploy that was using nose, but there isn't. Closing this as we ...
- 06:17 AM devops Bug #6503: ceph-deploy: nosetests fail
- We no longer use nose to test anything in ceph-deploy. The correct way is to use tox:...
- 05:04 AM Revision 34e50364 (ceph): Merge pull request #1023 from ceph/port/misc
- misc portability patches
Reviewed-by: Sage Weil <sage@inktank.com> - 04:34 AM Revision 2b0a4359 (ceph): osd_types: add missing osd op flags
- These were accidentally removed in:
85282319ee3e0d535d1ffc0a6ae8f763a41628b7
Fixes: #7067
Signed-off-by: Josh Durgin... - 03:37 AM Revision 3d895a0f (ceph): format bullets in README
- 02:57 AM Revision f8e413f9 (ceph): msgr: fix rebind() race
- stop the accepter and mark all pipes down before rebind to avoid race
Fixes: #6992
Signed-off-by: Xihui He xihuihe@...
12/30/2013
- 11:30 PM Revision e985ca2a (ceph): Port from httplib2 to requests module
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 11:30 PM Revision ba8f9999 (ceph): Split out ResultsSerializer.job_info()
- Signed-off-by: Zack Cerza <zack.cerza@inktank.com>
- 11:30 PM Revision 56c70e11 (ceph): Set the content-type in report_job()
- 11:28 PM rgw Bug #7073 (Resolved): "rgw gc max objs" should have a prime number as default value
- Recently when we trouble shoot latency increasing on our ceph cluster, we observed a couple of gc objects were hotspo...
- 10:52 PM Revision ece8b770 (ceph): Merge pull request #1006 from yuyuyu101/wip-7066
- Fix rbd bench-write improper behavior
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 09:36 PM Revision 8fcfc911 (ceph): qa: test rados listomapvals with >512 keys
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 09:36 PM Revision be5afa2e (ceph): rados: allow listomapvals to list all k/v pairs
- Current code only lists first 512 k/v pairs.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 09:27 PM rbd Bug #7067 (Fix Under Review): rbd read-flags.sh test is failing
- Good thing the test caught this before we were very confused by logs in the future missing some flags. Bisected and e...
- 09:20 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- I think I know what was going on. please try the attached patch, it should solve your issue. Thank you for your detai...
- 09:19 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- For building kernel, try increasing number of jobs that run simultaneously.
- 02:40 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- We plan to do a clean reinstall of the cluster next week (january 6). Of course we're more than happy to provide any ...
- 01:50 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- Zheng Yan wrote:
> are cpu usages of ceph-mds and ceph-osd high?
Hardly any CPU usage; Idle is > 95% on all cluster... - 01:47 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- Zheng Yan wrote:
> Rolf Fokkens wrote:
> > After the mds is in state "replay" for a while an impressive list of "fa... - 09:07 PM Revision db82bfec (ceph): mon: OSDMonitor: fix some annoying whitespace
- Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 42c4137cbfacad5654f02c6608cc0e81b... - 09:07 PM Revision 10ad2e01 (ceph): mon: set ceph osd (down|out|in|rm) error code on failure
- Instead of always returning true, the error code is set if at least one
operation fails.
EINVAL if the OSD id is inv... - 09:07 PM Revision 2354b957 (ceph): qa: test for error when ceph osd rm is EBUSY
- http://tracker.ceph.com/issues/6824 fixes #6824
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from c... - 09:07 PM Revision f862c530 (ceph): mon: OSDMonitor: fix some annoying whitespace
- Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
(cherry picked from commit 42c4137cbfacad5654f02c6608cc0e81b... - 09:07 PM Revision ca4540aa (ceph): mon: set ceph osd (down|out|in|rm) error code on failure
- Instead of always returning true, the error code is set if at least one
operation fails.
EINVAL if the OSD id is inv... - 09:07 PM Revision f75c9734 (ceph): qa: test for error when ceph osd rm is EBUSY
- http://tracker.ceph.com/issues/6824 fixes #6824
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from c... - 09:03 PM Revision c165483b (ceph): v0.74
- 08:58 PM Revision f7a66d6b (ceph): make: restrict use of --as-needed to Linux
- This should really be checked in configure.ac in the future.
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 08:58 PM Revision da5a082e (ceph): make: add top-level libcommon dependency
- On OSX there is consistently a problem with resolving pipe_cloexec and other
symbols through indirect libtool depende... - 08:58 PM Revision 8c8be16d (ceph): json_spirit: don't build json_spirit_value.cpp
- This file just includes the header that is full of templates and
typedefs, so the resulting object file has no symbol... - 08:58 PM Revision 5b775334 (ceph): make: avoid symbol exporting for C++ libs on non-Linux
- This removes export-symbol-regex for installed libraries with C++
interfaces on non-Linux where the hidden symbols ar... - 08:58 PM Revision ce8b26e8 (ceph): erasure: add dummy symbol to avoid warnings
- Empty object files generate warning.
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 08:58 PM Revision fdd85625 (ceph): test_cls_rbd: avoid shift overflow
- warning: signed shift result (0x500000000) requires 36 bits to represent
Signed-off-by: Noah Watkins <noahwatkins@gm... - 08:58 PM Revision 9d41fd2c (ceph): test_cls_rbd: avoid warning -Wno-unnamed-template-args
- According to this https://code.google.com/p/googletest/source/detail?r=446
the use of unnamed types (in this case the... - 08:58 PM Revision b4fc16c7 (ceph): make: conditionally build filestore backends
- Each of btrfs and zfs backends are wrapped in if __linux__ and if
WITH_ZFS, respectively, resulting in empty object f... - 07:10 PM Revision 87db89ea (ceph): librados: read into user's bufferlist for aio_read
- * The 'buf' argument to read() used to be passed into
AioCompletionImpl, and the results would be copied back after... - 05:22 PM Revision 2f25bfea (ceph): librbd: call user completion after incrementing perfcounters
- The perfcounters (and the ictx) are only valid while the image is
still open. If the librbd user gets the callback f... - 05:16 PM Revision 1abb1698 (ceph): librbd: call user completion after incrementing perfcounters
- The perfcounters (and the ictx) are only valid while the image is
still open. If the librbd user gets the callback f... - 05:09 PM rbd Bug #5426: librbd: mutex assert in perfcounters::tinc in librbd::AioCompletion::complete()
- Fix confirmed with the packages from http://gitbuilder.ceph.com/ceph-deb-precise-x86_64-basic/ref/dumpling-5426/ on M...
- 10:13 AM rbd Bug #5426 (Resolved): librbd: mutex assert in perfcounters::tinc in librbd::AioCompletion::comple...
- 04:49 PM Revision 4944618f (ceph): rbd: check write return code during bench-write
- This is allows rbd-bench to detect http://tracker.ceph.com/issues/6938
when combined with rapidly changing the mon os... - 04:49 PM Revision 4dae50d8 (ceph): objecter: don't take extra throttle budget for resent ops
- These ops have already taken their budget in the original op_submit().
It will be returned via put_op_budget() when t... - 04:49 PM Revision a3fdfd77 (ceph): objecter: don't resend paused ops
- Paused ops are meant to block on the client side until a new map that
unpauses them is recieved. If we send paused wr... - 04:49 PM Revision 7b229fca (ceph): objecter: set op->paused in recalc_op_target(), resend in not paused
- When going through scan_requests() in handle_osd_map() we need to make
sure that if an op should not be paused anymor... - 04:49 PM Revision 2b109803 (ceph): objecter: clean pause / unpause logic
- op->paused holds now whether operation should be paused or not, and it's
being updated when scanning requests. No nee... - 04:49 PM Revision f4d0fc8a (ceph): osd: drop writes when full instead of returning an error
- There's a race between the client and osd with a newly marked full
osdmap. If the client gets the new map first, it ... - 04:49 PM Revision 3a7a0353 (ceph): objecter: resend all writes after osdmap loses the full flag
- Now that the osd does not respond if it gets a map with the full flag
set first, clients need to resend all writes.
... - 04:38 PM Revision d97f4ed3 (ceph): objecter: don't take extra throttle budget for resent ops
- These ops have already taken their budget in the original op_submit().
It will be returned via put_op_budget() when t... - 04:38 PM Revision 67595286 (ceph): rbd: check write return code during bench-write
- This is allows rbd-bench to detect http://tracker.ceph.com/issues/6938
when combined with rapidly changing the mon os... - 04:38 PM Revision f5b801eb (ceph): objecter: resend all writes after osdmap loses the full flag
- Now that the osd does not respond if it gets a map with the full flag
set first, clients need to resend all writes.
... - 04:38 PM Revision ae81b156 (ceph): osd: drop writes when full instead of returning an error
- There's a race between the client and osd with a newly marked full
osdmap. If the client gets the new map first, it ... - 04:38 PM Revision 03628e36 (ceph): objecter: clean pause / unpause logic
- op->paused holds now whether operation should be paused or not, and it's
being updated when scanning requests. No nee... - 04:38 PM Revision b9cf0135 (ceph): objecter: set op->paused in recalc_op_target(), resend in not paused
- When going through scan_requests() in handle_osd_map() we need to make
sure that if an op should not be paused anymor... - 04:38 PM Revision 766fba83 (ceph): objecter: don't resend paused ops
- Paused ops are meant to block on the client side until a new map that
unpauses them is recieved. If we send paused wr... - 04:13 PM Revision 15cdcba1 (ceph): Merge pull request #1018 from dachary/wip-vstart-fsid
- vstart: set fsid in [global]
Reviewed-by: Sage Weil <sage@inktank.com> - 04:13 PM Bug #7072: logrotate not sending HUP signal
- It happened on another monitor and this time I was able to capture the output of...
- 04:05 PM Bug #7072: logrotate not sending HUP signal
- Can you identify which files are not being closed by looking at the open files in /proc/`pidof ceph-mon`/fd and seein...
- 03:52 PM Bug #7072 (Resolved): logrotate not sending HUP signal
- I have a cluster where the Ceph Monitors are actually Virtual Machines.
They have relatively small 20GB root drives.... - 04:12 PM Revision 6b871475 (ceph): valgrind.supp: ignore libnss3 leaks
- These just started popping up when I updated the notcmalloc gitbuilder, probably
because of an updated libnss version... - 04:12 PM Revision 85edc9f6 (ceph): Merge pull request #1019 from dachary/wip-show-config-value
- common: evaluate --show-config* after CEPH_ARGS
Reviewed-by: Sage Weil <sage@inktank.com> - 02:55 PM rbd Bug #7066 (Resolved): Improper bench-write implementation in do_bench_write@rbd.cc:924
- commit:fd57d99b6bbd2ec7d5e4c1f4f7cfcf4ba351050b
- 01:07 PM Bug #6824 (Resolved): Removal of an OSD that is not down should set non-successful status code
- 12:43 PM Bug #6824: Removal of an OSD that is not down should set non-successful status code
- Request to backport fix to Dumpling.
- 12:42 PM Bug #6824 (Pending Backport): Removal of an OSD that is not down should set non-successful status...
- 10:40 AM Bug #6725 (Resolved): objecter: kick_requests() resends ops that should be paused
- commit:98ab7d64a191371fa39d840c5f8e91cbaaa1d7b7 and commit:5fe3dc647bf936df8e1eb2892b53f44f68f19821 are related, alon...
- 10:32 AM Feature #6414 (New): filestore: make read() use a bufferlist method
- 10:32 AM Messengers Feature #6411 (New): msgr: move iov + sendmsg into bufferlist::sendmsg method
- 10:14 AM Linux kernel client Bug #6984 (Resolved): RBD volume not mountable after creating 8 or more snapshots
- This has been fixed in 3.12-rc1, commit
03507db631c94a48e316c7f638ffb2991544d617. - 10:12 AM Revision 0b40bbd4 (ceph): common: evaluate --show-config* after CEPH_ARGS
- The content of CEPH_ARGS is appended to the list of arguments. When
--show-config or --show-config-value is also set,... - 09:09 AM Revision b5c17f6f (ceph): vstart: set fsid in [global]
- If not set, commands that rely on --show-config-value fsid or something
equivalent will fail. ceph-disk does, for ins... - 08:26 AM Bug #7071 (Resolved): osd/ReplicatedPG.cc: 5953: FAILED assert(attrs->count("_"))
- ...
- 05:10 AM Revision 78037886 (ceph): Merge pull request #808 from ceph/wip-6152-dumpling
- rgw: Fix S3 auth when using response-* query string params
- 05:09 AM Revision 38d9613b (ceph): Merge pull request #1008 from ceph/port/spinlock
- spinlock: add generic spinlock implementation
Reviewed-by: Sage Weil <sage@inktank.com> - 04:58 AM Revision f9b50cee (ceph): Merge pull request #1014 from ceph/port/misc
- misc portability patches
Reviewed-by: Sage Weil <sage@inktank.com> - 04:44 AM Revision 9a7e85db (ceph): Merge pull request #1015 from yuyuyu101/wip-fix-tests
- Fix some test codes problems
Reviewed-by: Sage Weil <sage@inktank.com> - 04:41 AM Revision fb941d82 (ceph): Merge pull request #1013 from ceph/port/public-hdrs
- use portable types in public headers
Reviewed-by: Sage Weil <sage@inktank.com> - 04:36 AM Revision ee3d2277 (ceph): upgrade/parallel/stress-split: fix test
- Upgrade the second host before restarting the final monitor.
Run the dumpling api tests, since we still have half th... - 04:07 AM Revision d7d7ca84 (ceph): Fix qa/workunits/rados/test_cache_pool.sh typos
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 04:06 AM Revision cc67b7b2 (ceph): Fix test/filestore/store_test.cc error
- Commit
https://github.com/ceph/ceph/commit/d2b661d0efe79f5ec96e64647350cee0b80dc887
change the test result, just chan...
12/29/2013
- 11:18 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- Rolf Fokkens wrote:
> FYI one other thing: the performance is really bad when building a kernel on cephfs. "ceph sta... - 11:09 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- Rolf Fokkens wrote:
> I managed to get me a whole new list of "failed to reconnect caps for missing inodes" by doing... - 09:35 PM Revision 356d71a2 (ceph): inttypes: use portable types in public headers
- loff_t -> int64_t
__u8 -> uint8_t
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 09:35 PM Revision e8e174e7 (ceph): rados: include struct timeval definition
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:32 PM Revision 891801ed (ceph): warning: fix typo and -Wmismatched-tags
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:32 PM Revision 0c09abd9 (ceph): make: use c++ flags for building librados
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:32 PM Revision 424d7c78 (ceph): gtest: disable tr1/tuple
- Not all compilers are supporting tr1/tuple. This forces libgtest to use
an internal implementation of tuple. Alternat... - 09:32 PM Revision a48d0386 (ceph): test: fix VLA of non-POD type
- Variable length array of non-POD type is not supported by clang.
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 09:32 PM Revision 9a7a33ec (ceph): test: only build some tests on linux
- These tests have portability problems that have yet to be solved on
non-Linux systems.
Signed-off-by: Noah Watkins <... - 09:32 PM Revision a625b280 (ceph): test: print warning about missing fs features
- - sync
- posix_fadvise
Signed-off-by: Noah Watkins <noahwatkins@gmail.com> - 09:32 PM Revision 93c125ce (ceph): test: remove platform specific header
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:32 PM Revision 76ad85d8 (ceph): test: include headers for struct statfs
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:31 PM Revision bbcb0223 (ceph): kvstore: only build on linux
- There are several non-standard errno values used. There is still work to
do on addressing errno portability in Ceph, ... - 09:31 PM Revision c7e1c4bf (ceph): c++11: fix std::lock naming conflicts
- Unfortunately, 'using namespace std;' is in pretty widespread use in the Ceph
tree, so we need to rename to avoid the... - 08:26 PM Bug #6992: OSD assert fails after it found it was marked as down by monitor during high system lo...
- https://github.com/ceph/ceph/pull/1016/
From the log the rebind was at 2013-12-10 10:51:51.909718
2013-12-10 10:5... - 08:04 PM Revision f881157e (ceph): Merge pull request #1007 from ceph/wip-misc-fixes
- misc fixes
Reviewed-by: Loic Dachary <loic@dachary.org> - 05:03 PM Bug #7065: omap_clear() also removes all of object's xatts
- I think we should update test/osd/RadosModel (ceph_test_rados) so that it stores separate data in xattrs and omap (ri...
- 12:20 PM Bug #6003: journal Unable to read past sequence 406 ...
- ubuntu@teuthology:/a/teuthology-2013-12-28_23:00:01-rados-next-distro-basic-plana/18074...
- 08:31 AM Revision 16699f40 (ceph): Merge pull request #1004 from ceph/wip-snaps
- make ceph_test_rados read from snaps; resulting bugs found
Reviewed-by: Loic Dachary <loic@dachary.org> - 03:17 AM Linux kernel client Bug #7069 (Resolved): CephFS hang when using fscache - several "blocked for more than 120 seconds...
- Wanting to accellerate CephFS access I added fscache. This resulted in a CephFS hang to the point that processes acce...
12/28/2013
- 10:43 PM Revision 1fec818f (ceph): spinlock: add generic spinlock implementation
- Adds a ceph_spinlock_t implementation that will use pthread_spinlock_t
if available, and otherwise reverts to pthread... - 08:43 PM Revision a448f818 (ceph): rename upgrade/upgrade-* -> upgrade/*
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:36 PM Revision 12f46315 (ceph): qa/workunits/rest/test.py: rbd pool ruleset is now 0
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:34 PM Revision b286e4fe (ceph): ceph_test_rados_api_tier: retry EBUSY race checks
- ...or else these will occasionally fail against a thrashing cluster.
Signed-off-by: Sage Weil <sage@inktank.com> - 06:25 PM Revision b88af07e (ceph): libcephfs: get osd location on -1 should return EINVAL
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:22 PM Revision 250ecf66 (ceph): qa/workunits/mon/crush_ops.sh: fix in-use rule rm test
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:55 PM Revision d4f07cd9 (ceph): crush: fix get_full_location_ordered
- This should return -ENOENT when an id is not present. Broken by
746069ee62c74ecf04ed45988029d5c3382a38d2.
Signed-of... - 10:33 AM Revision fd57d99b (ceph): Fix rbd bench-write improper behavior
- "rbd bench-write" eject all write operations with the same offset at the same
time. It will result in non-objective p... - 10:25 AM Revision 30078e63 (ceph): autoconf: trim duplicated headers
- And sort the header list to help maintain uniqueness.
Signed-off-by: Loic Dachary <loic@dachary.org> - 10:16 AM Bug #7068 (Can't reproduce): os/FileStore.cc: 4035: FAILED assert(omap_attrs.size() == omap_aset....
- ...
- 08:55 AM rbd Bug #7067 (Resolved): rbd read-flags.sh test is failing
- teuthology-2013-12-27_23:00:59-rbd-master-testing-basic-plana has 3 instances
- 01:27 AM rbd Bug #7066 (Resolved): Improper bench-write implementation in do_bench_write@rbd.cc:924
- for (off = 0; off < io_bytes; off += io_size) {
b.wait_for(io_threads - 1);
uint64_t i = 0;
uint64_t... - 12:52 AM Revision 23f715ba (ceph): Merge pull request #1005 from ceph/wip-rgw-leak
- rgw: fix leak of RGWProcess
Reviewed-by: Yehuda Sadeh <yehuda@inktank.com> - 12:36 AM Revision f9f5c371 (ceph): rgw: fix leak of RGWProcess
- Introduced by a3e50b09a1fa22b80dea014d4b7bd96c23904f22.
Signed-off-by: Sage Weil <sage@inktank.com> - 12:32 AM Revision 96fe80db (ceph): osd: preserve user_version in snaps/clones
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:32 AM Revision 80b54876 (ceph): ceph_test_rados: test read from snapshots
- This was disabled back in 2011, c54aa7db3bc6e4c763e3b08d2ae98f89afe5a246.
Whoops!
Signed-off-by: Sage Weil <sage@ink...
12/27/2013
- 10:39 PM Bug #7065: omap_clear() also removes all of object's xatts
- It's amazing that just notice it now.
I think it's a little related to feature(http://tracker.ceph.com/issues/7059... - 10:16 PM Bug #7065 (Resolved): omap_clear() also removes all of object's xatts
- 09:45 PM Revision 2f8b6029 (ceph): osd/OSDMap: observe 'osd crush chooseleaf type' option for initial rules
- This option was dropped by 2a7fcc35b8ceeff1e07da28b10ced4a2a4ed09ec.
Signed-off-by: Sage Weil <sage@inktank.com> - 05:56 PM Revision 9c068939 (ceph): Merge branch 'rbd-map-options'
- Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
- 05:56 PM Revision 9b7364d2 (ceph): rbd: expose options available to rbd map
- Add a -o / --options option, which would allow users to specify
rbd-specific and generic ceph client and osd options ... - 03:59 PM Revision b8f42b6b (ceph): Merge pull request #1001 from dachary/wip-forward-tid
- messages: add tid to string form of MForward
Reviewed-by: Sage Weil <sage@inktank.com> - 01:19 PM Bug #7061: valgrind: mon and osd leaks in libnss on master
- teutholgoy commit:9a17bb5134ae7072089f780cd975761e61c708db
- 06:48 AM Bug #7061: valgrind: mon and osd leaks in libnss on master
- Is there a branch with the patch? Looked on github but nothing obvious caught my eye.
- 12:42 PM rbd Feature #6264 (Resolved): rbd: expose all options available to rbd map
- commit:9b7364d2450c7b51043bd39a0f586cf1c6041f4d
- 12:28 PM Revision 542f8d30 (ceph): Merge pull request #1002 from yuyuyu101/wip-7062
- Lack of "start" member function declare in WBThrottle.h
make check runs ok
Reviewed-by: Loic Dachary <loic@dachary.org> - 10:20 AM Bug #7062 (Resolved): Dead lock happend if daemonize
- 10:11 AM Revision b3bda085 (ceph): Lack of "start" member function declare in WBThrottle.h
- Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
- 06:32 AM rgw Bug #7064 (Resolved): S3 request to non-existent object in non-existent bucket results in 30 seco...
- A request to:
@curl -v http://192.168.2.16:8080/doesnotexist/missingkey@
Results in a 30 second timeout from Apac... - 06:30 AM Revision 4a9c7709 (ceph): messages: add tid to string form of MForward
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 05:33 AM Revision b8fb366e (ceph): Merge pull request #837 from ceph/port/fallocate
- FileJournal: zero-fill in-lieu of posix_fallocate
We may want to change that to a #warning later...
Reviewed-by: Sa... - 05:31 AM Revision 4f078489 (ceph): Merge pull request #908 from ceph/wip-6940
- rgw: fix reading bucket policy in RGWBucket::get_policy()
Reviewed-by: Sage Weil <sage@inktank.com> - 05:29 AM Revision f5b698cb (ceph): Merge pull request #982 from dachary/wip-default-crush-rule
- osd: add default crush rule for erasure pools
Reviewed-by: Sage Weil <sage@inktank.com> - 05:26 AM Revision 39a9c323 (ceph): Merge pull request #974 from dachary/wip-build-depends
- packaging: make check needs argparse and uuidgen
Reviewed-by: Sage Weil <sage@inktank.com> - 05:25 AM Revision a39226ac (ceph): Merge pull request #994 from yuyuyu101/wip-7062
- Fix WBThrottle thread disappear problem
Reviewed-by: Sage Weil <sage@inktank.com> - 05:18 AM Feature #7063: Replace lfn_find call in omap* functions with more lighter implematation
- It seemed that "IndexManager::get_index" protect concurrent write operations on the same object. So if remove it, mul...
- 05:14 AM Revision 67f99f34 (ceph): packaging: make check needs argparse and uuidgen
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 02:53 AM Revision 8f3ad4e3 (ceph): Merge pull request #1000 from ceph/wip-rbd-tinc-5426
- fix #5426 race in librbd
Reviewed-by: Sage Weil <sage@inktank.com> - 01:40 AM Revision 4cea7895 (ceph): librbd: call user completion after incrementing perfcounters
- The perfcounters (and the ictx) are only valid while the image is
still open. If the librbd user gets the callback f...
12/26/2013
- 11:28 PM Revision f8a40014 (ceph): osd: create default ruleset for erasure pools
- The ruleset --osd_pool_default_crush_erasure_ruleset is created to be
suitable for erasure coded pools when OSDMap::b... - 11:13 PM Revision a10fc025 (ceph): osd: build_simple creates a single rule
- The three rules created by build_simple are identical. They are replaced
by a single rule named replicated_rule which... - 11:13 PM Revision 2a7fcc35 (ceph): osd: use CrushWrapper::add_simple_ruleset
- Replace the manually crafted ruleset in OSDMap::build_simple_crush_map*
with calls to add_simple_ruleset. The generat... - 11:13 PM Revision dd81858c (ceph): mon: implement --osd-pool-default-crush-replicated-ruleset
- --osd-pool-default-crush-replicated-ruleset replaces
--osd-pool-default-crush-rule
If --osd-pool-default-crush-rule ... - 11:13 PM Revision 8b2b5a33 (ceph): mon: implement --osd-pool-default-crush-erasure-ruleset
- It must be different from the replicated default.
Signed-off-by: Loic Dachary <loic@dachary.org> - 11:10 PM Revision 15b69593 (ceph): crush: set min_rep and max_rep depending on mode
- Assuming firstn is for replica and indep is for erasure. This is a
strong constraint but it is unlikely to make the r... - 09:34 PM Bug #7061 (Pending Backport): valgrind: mon and osd leaks in libnss on master
- 09:34 PM Bug #7061: valgrind: mon and osd leaks in libnss on master
- 06:35 AM Bug #7061: valgrind: mon and osd leaks in libnss on master
- recent changes touching both the osd and the mon on master as of late, maybe the erasure coding stuff?
- 09:29 PM rgw Bug #6940 (Pending Backport): rgw: reading bucket policy through bucket restful api fails
- 07:38 PM Feature #7063: Replace lfn_find call in omap* functions with more lighter implematation
- Maybe we can maintain a hashmap<object, state>, "state" can be int and 0 mean unknown, 1 mean existing. If unknown, i...
- 06:54 AM Feature #7063 (Closed): Replace lfn_find call in omap* functions with more lighter implematation
- find_index plays a important role in handling transactions with FileStore. Although it's not a heave implementation, ...
- 06:57 PM rbd Bug #5426 (Pending Backport): librbd: mutex assert in perfcounters::tinc in librbd::AioCompletion...
- I cherry-picked the fix to dumpling-5426 and emperor-5426 branches based on dumpling and emperor, respectively. The f...
- 06:40 PM rbd Bug #5426 (Fix Under Review): librbd: mutex assert in perfcounters::tinc in librbd::AioCompletion...
- https://github.com/ceph/ceph/pull/1000
- 06:50 PM Revision da67f7c3 (ceph): crush: add rule_type argument to add_simple_ruleset
- Instead of hardcoded pg_pool_t::TYPE_REPLICATED
Signed-off-by: Loic Dachary <loic@dachary.org> - 05:56 PM Revision 2ae9c1c0 (ceph): partially rename rule to ruleset
- Where code is changed, get the opportunity to rename rule to ruleset to
improve naming consistency.
Signed-off-by: L... - 05:55 PM Revision e244be18 (ceph): Merge branch 'leseb-doc-rbd-havana'
- Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
- 05:55 PM Revision 8b0b32bd (ceph): doc: Add OpenStack Havana documentation
- New features appeared during the Havana cycle.
This patch offers a general update of the doc.
Signed-off-by: Sébasti... - 09:01 AM Revision c6d876aa (ceph): mon: osd-pool-create must not loop forever on kill
- Looping forever on kill does not serve any useful purpose.
Reduce the verbosity of the exit trap to help diagnose err... - 09:01 AM Revision 5cf2cdc0 (ceph): qa: remove osd pool create erasure tests
- Creating an erasure pool will crash the OSD because OSD::_make_pg
asserts if the type is not replicated. The tests re... - 09:01 AM Revision 6e92ed1e (ceph): osd: factorize build_simple and build_simple_from_conf
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 08:58 AM rbd Feature #6264 (Fix Under Review): rbd: expose all options available to rbd map
- wip-rbd-map-options; https://github.com/ceph/ceph/pull/999
- 08:12 AM Revision 20b3da05 (ceph): mon: MDS data and metadata pool numbers are hardcoded
- The MDS assumes pool 0 and 1 are suitable for data and metadata
respectively. Instead of relying on the CEPH_DATA_RUL... - 08:12 AM Revision 272eed35 (ceph): client: SyntheticClient uses the first available pool
- It is unrelated to CEPH_DATA_RULE which is replaced by
SYNCLIENT_FIRST_POOL.
Signed-off-by: Loic Dachary <loic@dacha... - 03:33 AM Revision bf24317b (ceph): Fix WBThrottle thread disappear problem
- New ceph_osd.cc code did ObjectStore init work before global_init_daemonize(),
and WBThrottle thread is created when ... - 12:09 AM Revision 21a64c17 (ceph): Merge pull request #995 from dachary/wip-deprecated
- rados: deprecated attribute has no argument
12/25/2013
- 07:41 PM Revision 87b8e54f (ceph): ceph_argparse: kill _daemon versions of argparse calls
- Commit c76bbc2e6df1, which introduced _daemon versions of some of the
argparse calls, also changed the behaviour of n... - 06:47 PM Feature #7059: filestore: avoid leveldb check for xattr when possible
- I'm not understand the first way clearly. A flag on the filestore when creating? The flag is associated to FileStore?
- 04:59 AM Feature #7059: filestore: avoid leveldb check for xattr when possible
- May I ask about the idea or method?
I also want to reduce redundant calls such as lfn_find() in omap_* functions ... - 12:12 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- FYI one other thing: the performance is really bad when building a kernel on cephfs. "ceph status" shows in general <...
- 12:04 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- I managed to get me a whole new list of "failed to reconnect caps for missing inodes" by doing the following:
- re... - 09:44 AM Revision ea4724d5 (ceph): rados: deprecated attribute has no argument
- The deprecated attribute argument was introduced in gcc 4.5
http://gcc.gnu.org/gcc-4.5/changes.html and centos6 has a... - 09:07 AM Revision ab75df3c (ceph): Merge pull request #988 from ceph/wip-crush-location
- add 'crush location' config option
make check is ok
Reviewed-by: Loic Dachary <loic@dachary.org>
12/24/2013
- 11:30 PM CephFS Feature #3866: mds: Add lazily-updated backtraces to hard links
- what does 'lazily-updated" mean?
- 11:24 PM CephFS Bug #3601 (Resolved): client: With multiple clients, file remove doesn't free up space
- 09:41 PM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- Rolf Fokkens wrote:
> The Ceph cluster was built using Ceph 0.67. On nov 13 it was upgraded to Ceph 0.72
looks li... - 03:14 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- We'll try to access the repaired FS and see if it still hangs.
- 03:13 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- I mounted the broken CephFS and did this (because it's easier than locating the specific inodes):
find -type d -ex... - 02:45 AM CephFS Bug #6914: Accessing specific file on CephFS causes client hang
- The Ceph cluster was built using Ceph 0.67. On nov 13 it was upgraded to Ceph 0.72
- 07:27 PM Bug #7062: Dead lock happend if daemonize
- https://github.com/ceph/ceph/pull/994
- 07:12 PM Bug #7062 (Resolved): Dead lock happend if daemonize
- New ceph_osd.cc code did ObjectStore init work before global_init_daemonize(), and WBThrottle thread is created when ...
- 06:51 PM Revision 7c9638f2 (ceph): Merge pull request #993 from ceph/wip-librados-lock
- Wip librados lock
Reviewed-by: Sage Weil <sage@inktank.com> - 05:00 PM Revision e7bf5b29 (ceph): librados: lockless get_instance_id()
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 05:00 PM Revision 771da13b (ceph): objecter, librados: create Objecter::Op in two phases
- (currently only in some librados operations)
First create the op, only then lock and submit so that we reduce lock
co... - 04:01 PM Revision 5ff30d6c (ceph): crush/CrushWrapper: note about get_immediate_parent()
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:58 PM Revision 0cdbc976 (ceph): librados: mark old get_version() as deprecated
- Use the newly-discovered (for me) deprecated attribute to mark the old
get_version() method and point users toward ge... - 03:58 PM Revision 746069ee (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> - 03:58 PM Revision ac14d4ff (ceph): osdc/Objecter: maintain crush_location multimap
- Observe and parse the 'crush location' config option.
Signed-off-by: Sage Weil <sage@inktank.com> - 03:58 PM Revision 22df7732 (ceph): osdc/Objecter: use crush location and distance for LOCALIZE_READS
- Use the hierarchy in the CRUSH map to determine what the closest
replica is.
Signed-off-by: Sage Weil <sage@inktank.... - 03:58 PM Revision 909f8a42 (ceph): librbd: localize or distribute parent (snap) reads
- The parent is always a snapshot. We may want to treat it differently
than other snaps by virtue of it (likely) being... - 03:58 PM Revision 006449dd (ceph): librados: deprecate aio_operate() read variant that takes snapid
- The argument was ignored.
Signed-off-by: Sage Weil <sage@inktank.com> - 03:57 PM Revision dcc5e355 (ceph): crush/CrushWrapper: add get_common_ancestor_distance()
- Calculate closest common ancestor (type) in the hierarchy.
Signed-off-by: Sage Weil <sage@inktank.com> - 08:22 AM Bug #7061 (Resolved): valgrind: mon and osd leaks in libnss on master
- not on next. unclear what changed...
- 06:08 AM Revision 2247a3df (ceph): upgrade/upgrade-parallel: don't test copy-from on dumpling
- It wasn't there yet
Signed-off-by: Sage Weil <sage@inktank.com> - 01:02 AM Revision a6852afd (ceph): Merge pull request #990 from ceph/wip-fix-mon-fwd
- mon: fix forwarded request features when requests are resent
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
Also available in: Atom