Activity
From 06/07/2011 to 07/06/2011
07/06/2011
- 11:45 PM Revision 850f337f (ceph): workunits task: clean up properly if there's an error.
- Previously it would fail out and leave the workunits directory, causing
final cleanup to fail.
Signed-off-by: Greg F... - 11:45 PM Revision 4fcecf01 (ceph): task ceph: set_max_mds so multiple MDS nodes are used
- The current check will be insufficient when we handle standby-replays,
standbys, etc, but it's a lot better than the ... - 10:14 PM rgw Bug #1286: atomic write not working for large files
- Looking at the logs, atomic writes are ok, however, the reads are not atomic.
- 06:16 PM rgw Bug #1286 (Rejected): atomic write not working for large files
- steps to reproduce:
* write a large file (4+ MB) of all A's
* start overwriting the file with all B's
* before ove... - 10:02 PM Revision 03b6c40b (ceph): qa: mds rename: account for not being in the ceph root dir
- We need to know the Ceph absolute path. We can't actually
derive that for sure (if we aren't mounted into the root), ... - 09:20 PM Revision e483bb6b (ceph): qa: mds rename: Rework so it will function in teuthology as a workunit:
- work in current directory, not hardcoded mnt path
use CEPH_TOOL variable rather than hardcoded local executable
pass ... - 09:20 PM Revision b5e4a319 (ceph): move mds rename tests to workunits directory
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:17 PM Revision 9655042d (ceph): Skip s3-tests marked fails_on_rgw, they will fail anyway.
- 09:17 PM Revision 1b2c9641 (ceph): Skip s3-tests marked fails_on_rgw, they will fail anyway.
- 09:15 PM Revision 3665f753 (ceph): The shell exits after the command, hence there is no need for pushd/popd.
- 09:15 PM Revision 06fb9b95 (ceph): The shell exits after the command, hence there is no need for pushd/popd.
- 08:49 PM Revision 088013b8 (ceph): mds: cleanup: use enum for lock states
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:49 PM Revision 528b6151 (ceph): Merge branch 'next'
- 08:34 PM Revision 8f9eaf0d (ceph): rgw: when listing objects, set locator key only when needed
- 07:36 PM rgw Bug #1287 (Resolved): Setting metadata with unreadable characters is not consistent with amazon S3
- If you have a string like '\x04world', Amazon will encode it using MIME encoded-word syntax. Currently, our S3 implem...
- 06:40 PM Revision 7ea8ecdc (ceph): Don't set unix env vars in fastcgi env.
- 05:47 PM Revision 0c6d2e6f (ceph): rgw: async rados delete should be called with the correct object
- 05:19 PM Bug #1285 (Resolved): rgw: copy object sets bad policy
- Fixed by commit:4761317b754c6a903862d1913b53c8ab5e06b8eb.
- 04:52 PM Feature #1219 (Closed): qa: migrate mds rename tests to teuthology
- Success. Had to fix a few bugs in teuthology too.
- 09:28 AM Feature #1219: qa: migrate mds rename tests to teuthology
- I've got this working I think, but it's exposed some bugs in cfuse and the kclient :( so I want to get those patched ...
- 04:19 PM CephFS Bug #1284 (Can't reproduce): client: leaked inode ref with cfuse_workunit_snaps.yaml
- Seems to be pretty rare. ...
- 04:13 PM Bug #1283 (Resolved): rados df produces negative numbers and other incorrect bucket sizes
- rados df produces negative numbers and other incorrect bucket sizes! this means we have no way of easily knowing the...
- 03:59 PM Revision 1d7fbed6 (ceph): rados: rename load-gen options
- No abbreviations, update usage().
Signed-off-by: Sage Weil <sage@newdream.net> - 02:30 PM Bug #1282 (Resolved): rados CLI segfault
- Horrible subject -- sorry, couldn't think of a better summary off hand. :(...
- 01:39 PM Bug #1279 (Resolved): osd: [WRN] bad locator
- Ok, the issue happened actually when listing objects inside a bucket. From what I understand this warning in this sit...
- 12:11 PM Bug #1279: osd: [WRN] bad locator
- ...
- 11:09 AM Bug #1279: osd: [WRN] bad locator
- I can't reproduce (this specific problem), what's your rgw commit version?
- 05:56 AM Bug #1279 (Resolved): osd: [WRN] bad locator
- Like I mentioned yesterday, I'm seeing bad locator messages on my OSD's after uploading a file through the RADOS gate...
- 11:28 AM Feature #1281 (Resolved): s3-test should test removal of owner's bucket grants behavior
- When owner doesn't have any grant to a bucket, it shouldn't be able to read or write to that bucket, however, it shou...
- 11:25 AM rgw Bug #1241 (Rejected): rgw denies owner access to bucket when acl is empty
- That's the expected behavior.
Will open a different issue for adding a test in s3-tests to verify that we do that. - 10:30 AM Bug #1280 (Closed): gceph: missing dependencies
07/05/2011
- 11:48 PM Revision f80a2f69 (ceph): Remove quotes from apache conf.
- 11:47 PM Revision f164dd79 (ceph): nuke: sudo for the final rm -rf
- 11:47 PM Revision 7e603ceb (ceph): adjust ceph.conf from yaml input
- 11:38 PM Revision a3b42e65 (ceph): Let apache kill rgw.
- It seems to be getting killed correctly now.
- 11:32 PM Revision ba6545f8 (ceph): Set LD_LIBRARY_PATH for rgw.
- 10:14 PM Revision d04256aa (ceph): set flavor to 'local' if pushing from local src dir
- always record in summary
- 09:57 PM Revision 5b2de2b9 (ceph): mkcephfs: Only create OSD journal dir if we have a journal.
- Thanks to huang jun <hjwsm1989@gmail.com> for finding the bug.
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamho... - 09:56 PM Revision 1da8f817 (ceph): honor CINIT_FLAG_NO_DEFAULT_CONFIG_FILE
- Don't use CEPH_CONF_FILE_DEFAULT when CINIT_FLAG_NO_DEFAULT_CONFIG_FILE
is set.
Signed-off-by: Colin McCabe <colin.m... - 09:22 PM Revision 2aa146a7 (ceph): mds: always clear_flushed() after finish_flush()
- The scatter_writebehind_finish() is always followed up by an eval_gather(),
which does the clear_flushed(). For ever... - 08:58 PM Revision fb7696f3 (ceph): client: fix num_flushing_caps accounting
- This only affects debug output, fwiw.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:48 PM Revision 003492a2 (ceph): Retry connecting based on more error codes.
- 08:43 PM Revision e9e3883d (ceph): client: don't call flush_snaps when nothing to flush
- Otherwise we fail an assert.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:43 PM Revision 933e3495 (ceph): mds: kill stray break
- This broke with the gatherbuilder addition.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:33 PM Revision 7e1f09ff (ceph): context: implement complete()
- finish() requires the caller to delete. complete() does that for you by
calling finish() and then doing delete this.... - 05:07 PM Revision b11b5826 (ceph): Merge branch 'stable'
- 05:00 PM Revision 531f46c3 (ceph): logrotate.conf: Mark stat/*.log as "missingok"; it's not always there.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 05:00 PM Revision 8459f80b (ceph): rgw: remove temp objects asynchronously
- 04:59 PM rgw Bug #1249 (Can't reproduce): radosgw does not exit when apache is killed
- 04:49 PM Feature #1254 (Resolved): teuthology: ceph.conf additions in ceph stanza of config.yaml
- 04:29 PM Bug #1278 (Resolved): teuthology: make nuke nuke kernel mounts
- usually the daemons are already down and the umount hangs. instead, umount -f, or just reboot (and then wait for the...
- 04:18 PM Revision 529df5db (ceph): Merge branch 'stable'
- 04:18 PM Revision b670f31d (ceph): Move stat/*.log to the end of logrotate.conf
- Logrotate ignores entries after a rule that doesn't match any files.
Signed-off-by: Alexandre Oliva <oliva@lsd.ic.un... - 04:01 PM Revision 2b168b03 (ceph): nuke: do not escape fusermount .../mnt.*
- 03:58 PM Revision 6feab3cb (ceph): mds: fix file_excl assert
- If we are in XSYN state and want to move to anything else, we must go via
EXCL, but we may not be loner anymore. Wea... - 03:30 PM Feature #1277 (Resolved): teuthology: manage known_hosts
- (02:21:43 PM) tv: host locking db gets an extra column, sshpubkey
(02:22:12 PM) tv: reinstalling script does ssh-key... - 02:40 PM CephFS Feature #1276 (Resolved): client: expose mds partition via virtual xattrs
- our options are basically and ioctl or virtual xattrs. virtual xattrs are easier to use and portable across fuse and...
- 02:36 PM CephFS Bug #1252 (Resolved): regression: cfuse fails snaptest-2 with 2-MDS system
- commit:e9e3883d0db4cc03183c0432773825f84170c42b
- 01:26 PM Feature #1271: proflogger: query interface
- - rip out current logging
- keep two copies of the counter vectors
- query via unix domain socket
- on each query,... - 12:40 PM Feature #1271 (Resolved): proflogger: query interface
- query via unix domain socket (instead of logging to a file and scraping that)
- 12:42 PM Bug #1275 (Resolved): fix testrados
- Fix testrados to always succeed, never fail or segfault. I probably will split the multi-threaded stuff into a separa...
- 12:41 PM Feature #1274 (Resolved): qa: test watch/notify races
- What happens if we put a watch on an object, and then another user destroys the pool it's in? Do we handle this case ...
- 12:41 PM Feature #1273 (Resolved): test pool creation and pool destruction races
- What happens if one guy destroys a pool while the other guy is using it? We should give an error code back to the guy...
- 12:40 PM Feature #1272 (Resolved): test rados list-objects in parallel with object modifications
- We need to test modifying a pool while its objects are being listed. Do we handle all the cases correctly? (Objects g...
- 11:57 AM Feature #1270 (Closed): qa: tests for splitting pgs vs read/write workload
- 11:33 AM Feature #1269 (Resolved): collectd: monitor cluster stats
- 10:55 AM rgw Feature #1268 (Rejected): rgw: optimistically fetch acl and first chunk
- 10:50 AM Feature #1267 (Rejected): osd: rgw class to do acl check
- 10:38 AM Documentation #1266 (Resolved): qa: document ceph-qa-suite
- 10:36 AM Feature #1265 (Resolved): osd: test osd recovery clone handling
- 10:33 AM Feature #1264 (Resolved): teuthology: allocate+lock ad/or verify nodes locked
- 10:32 AM Feature #1263 (Resolved): teuthology: scan cluster log for badness
- error out on any ERR (and possibly any WARN)? This will catch the errors the system notices but handles/ignores.
- 10:32 AM Feature #1262 (Closed): teuthology: monitor health during run
- 09:45 AM Revision d4fdaba9 (ceph): a few basic kclient workunits, reorg
- 09:45 AM Revision 3dd95f6c (ceph): .gitignore: ignore emacs backups
- 09:35 AM Feature #801 (Resolved): librados: allow access to multiple clusters
- 09:34 AM Bug #1185: rados: export caught in loop on 'buck' bucket (1.5M objects)
- See #1258
- 09:30 AM Subtask #815 (Resolved): Remove globals & partition g_conf
- This bug was really all about refactoring stuff to be thread-safe. Now, that task has been completed.
It's time to... - 09:28 AM Subtask #1227 (Rejected): write tests of libceph, librgw, librados library thread-safety
- this isn't really a subtask of #815. Will create more detailed bugs for all these.
- 09:26 AM Subtask #1231 (Rejected): NUM_THREADS=3 testrados segfaults
- Moving this into issue #1261
- 09:23 AM Linux kernel client Bug #1253 (Resolved): null point deref in __dentry_open
- 09:13 AM Bug #1257 (Resolved): test/fix concurrency issues with RADOS
- In general, we need to make sure that concurrent operation works properly with RADOS / librados / DHO.
We have never...
07/04/2011
- 10:18 PM Bug #1256 (Won't Fix): mds dies on ESHUTDOWN under too-high mon load (time-outs?)
- It's relatively common for the active mds to die while I run backups from/to the filesystems that hold the mon data i...
07/03/2011
- 10:29 PM Bug #1255 (Duplicate): configure does not detect missing g++
- When g++ is missing, it fails on some other check (boost). Should verify that g++ exists first.
- 03:57 AM Revision e1db8e9e (ceph): new workunits
07/01/2011
- 11:18 PM Revision 03a08eb3 (ceph): Add rgw task with hardcoded apache config.
- 11:18 PM Revision ae87abf3 (ceph): Add s3tests task.
- 11:18 PM Revision cd524a69 (ceph): Add s3tests task.
- 10:44 PM Revision 924a3225 (ceph): obsync: improve formatting a little bit
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:28 PM Revision da917ade (ceph): obsync: add man page, documentation line
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:51 PM Bug #1185: rados: export caught in loop on 'buck' bucket (1.5M objects)
- The original process is still running (but suspended). Unfortunately the binary is an old build so there are no debu...
- 04:05 PM Bug #1185: rados: export caught in loop on 'buck' bucket (1.5M objects)
- This is something where a core file or a backtrace would be really, really helpful. I reviewed the code in librados::...
- 08:47 PM Feature #1215 (Resolved): qa: ability to run radosgw
- 08:47 PM Feature #1216 (Resolved): qa: run s3-tests in suite
- 05:56 PM Revision f5cca2e8 (ceph): buffer: remove do_cow, clone_in_place
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:34 PM Revision 6e736071 (ceph): More idiomatic Python.
- 04:33 PM Revision ce041a52 (ceph): summary is used outside the try: except:, move it outside it.
- 04:32 PM Revision c1cd141b (ceph): Only ignore ENOENT errors in teuthology-ls.
- 04:29 PM Revision fdbf591c (ceph): Whitespace cleanup.
- 04:28 PM Revision e697e4b8 (ceph): pyflakes cleanup.
- 04:22 PM Revision 2137da2c (ceph): Add final slash to path to enforce it being a dir.
- 04:09 PM Revision 53fc6920 (ceph): use local source dir for teuthology run
- - ceph:
path: /home/sage/ceph
...
This will 'make install' to a temp dir, tar it up, and ship that to the
target... - 03:52 PM Feature #1172 (Resolved): obsync: man page
- implemented by commit:da917ade4ab1f9e5dc7674a439bb212d71d616f2 and commit:924a3225acf7ab5542058f8b427385b7ec5c16b8
- 03:52 PM Revision bd79ae82 (ceph): Merge remote branch 'origin/wip-client'
- 02:45 PM Feature #1254 (Resolved): teuthology: ceph.conf additions in ceph stanza of config.yaml
- 02:42 PM Feature #1204 (Resolved): qa: add existing workunits to suite
- 10:18 AM Linux kernel client Bug #1253 (Resolved): null point deref in __dentry_open
- commit:4f62174bf07425d9d815652e078bed73397a5f00
workunits/misc/trivial_sync.sh on sepia
doesn't trigger under uml... - 10:15 AM CephFS Bug #1252 (Resolved): regression: cfuse fails snaptest-2 with 2-MDS system
- On head of master, it asserts: ...
- 09:23 AM rgw Feature #1251: rgw: consistent caching of bucket objects (acls)
- This is going to be implemented using watch/notify. Each node keeps its own cache but notifies other nodes about chan...
- 09:07 AM rgw Feature #1251 (Resolved): rgw: consistent caching of bucket objects (acls)
- use notify with payload on an object to sent out cache invalidation messages to all rgw instances
- 09:00 AM CephFS Bug #1239 (Resolved): client: hang on umount after snaps workunit
- 07:23 AM Revision a6ffcc8d (ceph): librados: close very small race condition
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:17 AM Revision 0e6de715 (ceph): mon: add 'osd create [id]' command
- If the id is specified, mark a non-existant osd rank as existant. The id
must fall within the current [0,max) range.... - 05:24 AM Revision 1af8998c (ceph): client: clean up cap flush methods
- We grew several copies of this code, and it turns out none of them were correct.
- assign flush tid in send_cap() he... - 05:15 AM Revision 984e5a0a (ceph): Makefile: libmds.a, not libmds.la
- We never link this into a .so, so avoid building it again with -fPIC.
Signed-off-by: Sage Weil <sage.weil@dreamhost.... - 05:04 AM Revision 038a754f (ceph): mds: fix off-by-one in cow_inode vs snap flushes
- We need to wait for the client to flush snapped caps if the client has
not already flushed for the given snap. If th... - 12:13 AM Revision e6eef5e9 (ceph): rgw: sync cache between rgw instances
06/30/2011
- 11:19 PM Feature #1244 (Resolved): mon: allocate osd ids
- 11:05 PM Revision effee7ff (ceph): Make kernel a separate entity outside of tasks.
- It is run before anything other than checking for conflicts.
This way it can't step on the connections used by other ... - 10:51 PM Revision 4b1d536a (ceph): Don't recreate ctx.cluster when reconnecting after installing kernels.
- 10:48 PM Revision 6f86de44 (ceph): Save kernel version in summary instead of a separate file.
- 10:47 PM Revision 7b1b3322 (ceph): Kernel task does not need to be a context manager.
- 09:45 PM Revision 1206625b (ceph): rgw: fix of users are created suspended
- 09:43 PM Revision ca6d2390 (ceph): Fix handling of CEPH_CONF
- Formerly, CEPH_CONF was not respected by libraries. But now it is.
It overrides the default when reading the config f... - 09:00 PM Revision 31d49735 (ceph): rados: fix warning
- 08:44 PM Revision 6e49415c (ceph): client: only send one flushsnap once per mds session
- This mirrors a kclient change a while back (e835124).
We only want to send one flushsnap cap message per MDS session... - 08:33 PM Revision 133904d7 (ceph): Merge branch 'rados-load-gen'
- 08:32 PM Revision 860c6657 (ceph): rados tool: load generator
- 06:26 PM Revision e16556e3 (ceph): Archive dir removal has to be unconditional.
- Even when ctx.archive is False, ceph logging
need the destination directory exist, so
/tmp/cephtest/archive has to be... - 06:25 PM Revision cb4ffc31 (ceph): Use idiomatic python.
- 05:24 PM Revision 9771a8e2 (ceph): C_GatherBuilder: more uses, add set_finisher
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:24 PM Revision 728c132a (ceph): C_Gather: hide constructor, convert uses
- Note: this fixes a small memory leak in MDCache::open_snap_parents.
Signed-off-by: Colin McCabe <colin.mccabe@dreamh... - 05:24 PM Revision 16b65678 (ceph): C_Gather: remove unused "any" option
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:24 PM Revision f69fcc70 (ceph): C_GatherBuilder: add C_GatherBuilder::activate()
- Add an activate() function that must be called before we call the
onfinish callback. This is especially important in ... - 05:24 PM Revision 2f5925ea (ceph): Add "How to use C_GatherBuilder" comment
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:24 PM Revision a157bbb8 (ceph): Add compiler_extensions.h for warn_unused_result
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:24 PM Revision 5f53131f (ceph): mds/journal.cc: remove deadcode
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:24 PM Revision 4772bb69 (ceph): Add C_GatherBuilder
- C_Gather objects are deleted by the last sub-context to execute.
If you create a C_Gather object manually, you must w... - 05:24 PM Revision 562a04df (ceph): Filer.h: use C_GatherBuilder
- Filer.h now uses C_GatherBuilder to avoid memory leaks.
Also, C_GatherBuilder's constructor now takes a Context.
Si... - 04:58 PM Revision 648e50e6 (ceph): obsync: Depend on python-lxml on Debian derived platforms
- Signed-off-by: Wido den Hollander <wido@widodh.nl>
Signed-off-by: Sage Weil <sage@newdream.net> - 04:21 PM Revision 2fbba81f (ceph): osd: don't spew spurious scrub unreserve messages
- The past primary was sending out scrub unreserve messages to all the
non-primary OSDs in the acting set on a PG state... - 02:52 PM Bug #1250 (Resolved): rgw users are created suspended
- This was already fixed, but found its way into 0.30, so commit:1206625b57a35da9b6fa67e1ce6a41c771ef06f3 fixes it, pus...
- 02:33 PM rgw Bug #1249 (Can't reproduce): radosgw does not exit when apache is killed
- When apache receives SIGKILL, its fastcgi helpers should die as well. The configuration used is the one setup by the ...
- 02:20 PM Feature #1219 (In Progress): qa: migrate mds rename tests to teuthology
- I'm trying to better acquaint myself with teuthology and this seems pretty rote.
- 11:26 AM Bug #1185: rados: export caught in loop on 'buck' bucket (1.5M objects)
- trying to reproduce this (with logs) and having a hard time. :/
cd /mnt/backup/dhobjects
rados -n client.dhoba... - 11:23 AM Bug #1242 (Resolved): OSD cleans up scrubs wrong
- There was less stuff going on in the on_change area than I thought. Pushed this in commit:2fbba81f6459cdfbc447eab94a9...
- 10:27 AM Bug #1248 (Resolved): teuthology: setup fails horribly if you don't have a mon on each node
- ...
- 09:52 AM Linux kernel client Bug #1247 (New): kicking capsnaps can send dups
- we put the capsnaps on the list, but resend (all) by inode.
probably just pass which capsnap to reflush instead o... - 09:17 AM Feature #1246 (Resolved): teuthology: bundle/install ceph from a local source tree
- (instead of gitbuilder) for more rapid development/debugging!
should be pretty straightforward.. just need to gene... - 06:48 AM Bug #1245 (Resolved): obsync: Issues with python-xattr (AttributeError: 'module' object has no at...
- While trying obsync under Ubuntu 10.04 I got:...
- 03:32 AM Revision 7779ca15 (ceph): client: more inode ref counting debugging
- blech
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 03:17 AM Revision 490f7e95 (ceph): client: pin dentries referenced by MetaRequest
- Pin dentries referenced by MetaRequest.
Signed-off-by: Sage Weil <sage@newdream.net> - 03:17 AM Revision 553c8a9f (ceph): client: do not assume MetaRequest's dentries are linked
- The dentries we reference may have been unlinked prior to us sending this
request. That's fine as long as we don't d... - 03:17 AM Revision 9da44e67 (ceph): client: do not leak MetaRequests on get_or_create() failure
- Avoid leaking in the error paths.
Signed-off-by: Sage Weil <sage@newdream.net> - 03:17 AM Revision 3ef86711 (ceph): client: fix dentry ref counting
- Before nobody ever pinned dentries except Dirs. The only valid ref
counts were 0 and 1, and unlink, rename, etc woul...
06/29/2011
- 10:52 PM Revision bec489d4 (ceph): rados: some more load-gen work
- 10:23 PM Revision 310032ee (ceph): mds: fix mds scatter_writebehind starvation
- scatter_writebehind is called by eval_gather on dirty locks, and
eval_gather is called by wrlock_finish on unstable l... - 10:23 PM Revision d1c5fb1d (ceph): messenger: rename is_blank_addr to is_blank_ip.
- This is a better description of the function.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 10:23 PM Revision d918e675 (ceph): msgr: simply set_ip by using is_blank_ip
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:03 PM Revision 0b252e99 (ceph): client: make tdout() work in header
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:44 PM Revision aad529e5 (ceph): rgw: merge chunks for clone range, truncate to 0 before
- The truncation to 0 has two purposes: so that if dest was bigger than
source, and we also hit a btrfs bug which we ca... - 07:58 PM Revision 902f699c (ceph): qa: add clone_range tool
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:55 PM Revision 7d57b355 (ceph): teuthology-ls: summarize results from an archive-dir
- ugly but basically works
- 07:55 PM Revision 43763169 (ceph): workunits: set LD_LIBRARY_PATH
- Needed by rbd workunit to find librbd.so
- 07:23 PM Revision b95e61ae (ceph): teuthology-nuke
- Take in a full config (or just targets: portion) and do a destructive
cleanup.
Still need to clean up kernel mounts. - 07:09 PM Revision 052f43c9 (ceph): pass owner, optional description through to summary.yaml
- Owner can be overridden explicitly, otherwise it's the running unix user.
The description is optional and passed str... - 07:09 PM Revision 770edc3a (ceph): suite: set each run description to combination of facet names
- 07:09 PM Revision 2125e8dc (ceph): include @hostname in owner
- 05:57 PM Revision bc91e1b5 (ceph): workunits: set CEPH_ARGS so 'ceph' etc can find config
- 05:49 PM Revision 7d754419 (ceph): qa: file_layout.sh is kclient only
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:47 PM Revision 8bcc639a (ceph): messenger: add a set_ip function to initialize the IP
- Previously we only filled in IPs if they were set by the config file
(at startup) or after we connected to the monito... - 05:41 PM Revision a9a27336 (ceph): workunit: include ceph bin dir in path
- Some workunits need these binaries (e.g. kclient/file_layout.sh needs
cephfs).
Signed-off-by: Sage Weil <sage@newdre... - 04:51 PM Revision b5510a66 (ceph): qa: move file_layout into workunits/misc
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:48 PM Revision 1247a224 (ceph): qa: fix file_layout.sh mode
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 04:46 PM Revision cde5dc22 (ceph): client: improve debug output a bit
- ...to help track down inode ref count leaks.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 04:46 PM Revision 92e0fac0 (ceph): mds: fix snaprealm split for dir inode
- The snaprealm root directory inode belongs to the snaprealm, at least
currently. This make split_at() consistent wit... - 04:37 PM Revision c292cce3 (ceph): radosgw_admin: fix cli test
- 04:34 PM Feature #1244 (Resolved): mon: allocate osd ids
- for chef
- 03:32 PM CephFS Bug #1243 (Resolved): inest lock blocks dir create for a long time
- commit:310032ee8128f6417ac302c0f5ecd27c691cbcc7
I haven't been able to figure out why this doesn't impact cfuse, t... - 01:43 PM CephFS Bug #1243 (Resolved): inest lock blocks dir create for a long time
- From the mailing list:
steps are:
we mount ceph on /mnt/test/
then create dir "/mnt/test/a/b/"
1) in dir "b" , us... - 01:30 PM Bug #1242: OSD cleans up scrubs wrong
- Sam and I looked at this and I have a patch to test, but I want to audit the role change code a bit to look for simil...
- 10:55 AM Bug #1242 (Resolved): OSD cleans up scrubs wrong
- The logs and core are in vit:/home/joshd/startup_crash/ - the relevant OSD is osd.1 on sepia82.
Specifically, it cal... - 10:55 AM Messengers Bug #1220 (Resolved): failed assert: peer_addr > msgr->ms_addr
- I fixed the self-connection problems in commit:8bcc639ab2171827286dafb42ef4635477dee8f1.
Then I created bug #1242 ... - 10:12 AM Messengers Bug #1220: failed assert: peer_addr > msgr->ms_addr
- But not just an OSD bug -- the messenger shouldn't form connections to itself. There is code to prevent this and asse...
- 09:57 AM Messengers Bug #1220 (In Progress): failed assert: peer_addr > msgr->ms_addr
- Actually, this is an OSD bug:...
- 10:52 AM rgw Bug #1241 (Rejected): rgw denies owner access to bucket when acl is empty
- Currently the @test_bucket_acl_revoke_all()@ test creates a bucket that you can't delete on the rgw (until you change...
- 10:02 AM Feature #1240 (Resolved): teuthology: fail/clean up when daemons crash
- currently if cmds crashes you see the error dump but teuthology doesn't shut everything down and clean up.
- 09:35 AM CephFS Bug #1232 (Closed): client: inode ref leak?
- 09:34 AM CephFS Bug #1239 (Resolved): client: hang on umount after snaps workunit
- two inodes still have pins. looks like a ref count leak somewhere....
- 09:27 AM Bug #1221 (Resolved): journal dump and reset don't work with cephx on
- This doesn't occur in the current stable, although I can still reproduce it on the named commit. Don't know what the ...
- 09:11 AM CephFS Bug #1238 (Resolved): mds: ~elist crash on untar_snap_rm
- 09:09 AM Bug #1032: osd: Marked down and become zombies after killing
- If you can reproduce this, the next step is to figure out what the kernel threads are doing. Do several rounds of
... - 04:05 AM Bug #1032: osd: Marked down and become zombies after killing
- I'm seeing it again. Trying to upgrade from v0.29.1 to v0.30.
My cluster was bouncing all over again, "wrongly mar... - 09:07 AM Tasks #1165 (Closed): snap collection clone removal
- 09:07 AM Tasks #1157 (Closed): create teuthology tests for updated snap_trimmer
06/28/2011
- 11:33 PM Revision a97f5eeb (ceph): Merge branch 'master' of ssh://ceph.newdream.net/git/ceph
- 11:32 PM Revision 69952c18 (ceph): Merge branch 'master' into rgw
- 11:29 PM Revision 4b16d6f1 (ceph): radostool: load gen, wip
- 11:07 PM Revision f7fb5475 (ceph): qa: make kernel_untar_build.sh output more helpful
- - rm -rv
- list remaining files on failure
to help debug #1230
Signed-off-by: Sage Weil <sage@newdream.net> - 10:04 PM Revision 4815f512 (ceph): Add rgw testing with s3tests.
- 09:51 PM Revision 0904ce55 (ceph): Move get_clients method into misc to avoid duplication.
- 09:51 PM Revision 79956538 (ceph): Add kernel task for controlling kernel version for different roles.
- 09:19 PM Revision 016cf673 (ceph): testceph: test file opens with extra slashes
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:18 PM Revision f5cbe50e (ceph): testceph: expand test cases
- Signed-off-by: Richard Sharpe <realrichardsharpe@gmail.com>
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 09:18 PM Revision 256e6a2b (ceph): tasks/clock: tolerate ntp daemon
- 09:13 PM Revision a851a155 (ceph): filepath: don't parse multiple slashes as multiple dname bits.
- This causes all kinds of trouble if it occurs because most of the
code isn't prepared for it. So prevent that from ha... - 07:37 PM Revision 2c70bb8f (ceph): Journaler: pay attention to return codes from read head.
- Previously we ignored them, except for printing them out. This could
lead to bad things like creating new journals fo... - 07:37 PM Revision d1fcffad (ceph): Resetter: Check return values from Journaler.
- And clean up the shutdown a bit.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 06:23 PM Revision b418896d (ceph): PG: fix add_next_event and merge add_event
- Previously, we would assume that we had an object at the prior_version
in the log event if we encounter it but don't ... - 06:23 PM Revision 33aa5786 (ceph): PG: simplify merge_log, remove old objects
- Previously, some old objects could escape removal during a log merge.
This patch also simplifies merge_log somewhat.
... - 06:23 PM Revision a440fd09 (ceph): ReplicatedPG: Fix some obc leaks
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 05:54 PM Revision fab24c80 (ceph): filestore: allow FIEMAP to be disabled via conf
- The ext4 fiemap ioctl is buggy in some older kernels. Allow the admin to
disable it.
Fixes: #1222
Signed-off-by: Sa... - 05:36 PM Revision 0568d5cb (ceph): mds: make cast explicit to suppress compiler warning.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 05:36 PM Revision 033f2d63 (ceph): qa: add file_layout.sh workunit
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 04:41 PM Revision 7612c762 (ceph): Merge branch 'next'
- 04:18 PM Revision fb81442c (ceph): testradospp: fix compilation
- 03:29 PM Revision 1446d6de (ceph): mon: Guard injectargs parsing to prevent crashes on bad input
- While we're at it, reply to injectargs commands instead of
doing it silently.
Signed-off-by: Greg Farnum <gregory.fa... - 03:19 PM Revision 6a7e635a (ceph): mds: Explicitly initialize layout fields, and to the correct values.
- We were previously encoding an fl_pg_preferred of 0, which did
horrible things to the kernel client since 0 is a vali... - 03:09 PM CephFS Bug #1193 (Resolved): cmds crashes with stacktrace when using --reset-journal and an invalid rank
- Should be fixed by a pair of commits ending in commit:d1fcffad6c038c6712f1d83735e629017829a2fb
- 03:07 PM Bug #1235 (Resolved): mds crashes if the client sends an empty component in _lookup
- Should be resolved by commit:a851a1556f00820ff17ebae03cc6928f9df9657c.
- 11:13 AM Bug #1235: mds crashes if the client sends an empty component in _lookup
- Apparently I need to be careful about this because the client uses empty path names to tell the MDS to go into the sn...
- 09:17 AM Bug #1235 (Resolved): mds crashes if the client sends an empty component in _lookup
- Samba submits names like .//some-file.txt.
Libceph in path_walk fails to notice the empty component and calls _loo... - 02:11 PM CephFS Bug #1238 (Resolved): mds: ~elist crash on untar_snap_rm
- INFO:teuthology.task.workunit.client.0.out:removed `k/coreutils-8.5/lib/stripslash.c'
INFO:teuthology.task.workunit.... - 01:43 PM Feature #1190: codingstyle document
- Looks basically fine to me (tho i didn't read the whole thing). It leaves a lot of stuff unspecified, though. Notab...
- 12:56 PM rgw Feature #1224 (Resolved): s3-tests should test 100-continue
- Done, commit:abcb67dec0a4f656016efb8b54ea595ac15c9c07.
- 12:28 PM CephFS Feature #1237 (Resolved): mds caps limit mount to some subdir
- About a year ago the ceph.conf supported something like:...
- 12:23 PM Bug #325 (Can't reproduce): writes starve reads
- 11:18 AM Bug #1191: FAILED assert(!missing.is_missing(soid))
- This is probably caused by a bug in missing set construction during log merging. It may be fixed in 33aa578656f64606...
- 11:08 AM Feature #1222 (Resolved): configurable way to disable fiemap use in the osd
- 10:58 AM Bug #1187: OSD: OSDMap::decode
- Oh yes, I had to rebuild since I screwed up my single monitor...
- 10:43 AM Bug #1187 (Can't reproduce): OSD: OSDMap::decode
- It looks like the cluster has been rebuilt since then? Epoch 25743 (that it couldn't get) is >> the current 912. Or...
- 10:51 AM CephFS Feature #1236 (Resolved): libceph: set layout via virtual xattrs (libceph/cfuse)
- Apparently there actually is support in cfuse for ioctls, although it's hairy. Given that, we should support them in ...
- 10:45 AM rgw Feature #1198 (Resolved): rgw user suspension
- 10:42 AM Feature #1202 (Resolved): qa: workunits to set file layout and verify it works.
- Okay, set up a workunit that's reasonably comprehensive. Being a workunit it doesn't cover cross-client checking; we'...
- 09:18 AM Bug #1234 (Resolved): mon: Monitor::handle_command
- commit: 1446d6de01e5c8b7d0ffd3ed83cd1a602b5768b1
- 08:27 AM Bug #1234 (In Progress): mon: Monitor::handle_command
- Yeah, the monitor just doesn't look at what it's doing before it does it with injectargs. Fixing.
- 07:13 AM Bug #1234 (Resolved): mon: Monitor::handle_command
- I was playing with "injectargs" a bit today and was able to crash my monitor a few times by just doing:...
- 05:23 AM Revision 842ec226 (ceph): debian: drop python-support version req
- 05:23 AM Revision 64b1b2c7 (ceph): v0.30
- 12:21 AM Revision e27a8939 (ceph): rados: encode bufferlist in watch-notify
- 12:21 AM Revision 7bbdcdba (ceph): rgw: radosgw_admin user rm --purge-data
- 12:21 AM Revision dbee6992 (ceph): rgw: atomic put
- 12:21 AM Revision d0340426 (ceph): rgw: implement atomic multipart upload, atomic copy
- 12:21 AM Revision e5801728 (ceph): rgw: radosgw_admin --help typo
- 12:21 AM Revision a0dcf682 (ceph): rgw: fix clone_range
- 12:21 AM Revision 6d6b05d8 (ceph): rgw: fix some issues with call to clone_range
- 12:21 AM Revision 01118352 (ceph): rados: encode bufferlist in watch-notify
06/27/2011
- 10:48 PM Bug #995 (Can't reproduce): unrecoverable 0.*p* PGs
- 10:48 PM Messengers Bug #882 (Closed): misc leaks in librados
- 10:48 PM Bug #990 (Closed): osd: PG::replay_queued_ops
- 10:48 PM Bug #991 (Closed): osd: OSD::_process_pg_info
- 10:48 PM Bug #992 (Closed): osd: OSD::_lookup_lock_pg
- 09:57 PM Revision ef25dbbd (ceph): config: remove unused iat options
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 07:07 PM Revision 92b0e8e8 (ceph): qa: initial commit of file_layout.sh
- This tests the file layout functionality. Right now you need to
check the comparisons yourself. Expect that to change... - 07:07 PM Revision 19614bb8 (ceph): testceph: actually call closedir on opened directory.
- Otherwise testceph hangs because it's still got things open!
Fixes #1228
Signed-off-by: Greg Farnum <gregory.farnum@... - 06:40 PM Revision 58e4dc19 (ceph): filer: no need for tid_t here
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:10 PM Linux kernel client Bug #1226 (Resolved): kclient can't read newly-created file with a custom layout
- Fixed by commit:9735451e4b1ec197545de5daf222d518cbf7369c
This did turn out to be an MDS/cephfs bug, actually -- it... - 02:24 PM Linux kernel client Bug #1226: kclient can't read newly-created file with a custom layout
- ...and in fact it can't write it either, for the same reason. It's just that writes are buffered so you don't see the...
- 01:05 PM Linux kernel client Bug #1226 (In Progress): kclient can't read newly-created file with a custom layout
- Aha, or perhaps not the caps......
- 05:10 PM Revision c6f2c32a (ceph): client: print cache size during trim for debug
- Signed-off-by: Sage Weil <sage@newdream.net>
- 01:49 PM Bug #1233 (Resolved): ceph -v reports incorrect version
- Not a bug... there's something hosed in the way I'm storing my git repo such that git describe is reporting an old ve...
- 01:06 PM Bug #1233: ceph -v reports incorrect version
- I am indeed betting that .git_version is a non-git file... though I think the name is pretty unfortunate... might be ...
- 01:04 PM Bug #1233 (Resolved): ceph -v reports incorrect version
- I'm not sure what the .git_version file is supposed to be, as I see no reference in git docs to such a file.
From ne... - 12:14 PM Bug #1228 (Resolved): mds bug: testceph hangs forever in Client::unmount
- Well, this was a testceph problem. Fixed by commit:19614bb868124fcb51308f62f7c0cfa13a31038c so it actually closes the...
- 09:23 AM CephFS Bug #1225 (Resolved): client: recallstate removing caps when it shouldn't
06/26/2011
- 10:33 PM CephFS Bug #1232 (Closed): client: inode ref leak?
- see ginormous cfuse log on vapre
06/25/2011
- 08:22 AM Revision ea73a5cd (ceph): FileStore: fix error in ALIGN and ALIGN_UP
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 04:04 AM Revision 034aa086 (ceph): radsogw_admin: fix clitest
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 03:53 AM Revision f7b2555e (ceph): testrados_delete_pool_while_open: fix printf args
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 03:39 AM Revision bd7e3e01 (ceph): radosgw_admin: --help/-h
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:30 AM Revision 6a3626d3 (ceph): create testrados_delete_pool_while_open
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
06/24/2011
- 11:51 PM Revision 8ba05c44 (ceph): PG: assemble_backlog - fix comparison error, tail is non-inclusive
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 11:49 PM Revision 00fa99b9 (ceph): client: touch dentries during path traversal
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:49 PM Revision 4e2a8fab (ceph): client: touch cap on lookup even if we use the dentry lease
- Touch the dir cap for the lease's mds even if we use the lease to traverse.
This makes the trim_caps() behave better ... - 11:49 PM Revision 4c47df35 (ceph): client: prefer auth cap in caps_issued_mask()
- If we have an auth cap, prefer to touch that one. This helps consolidate
caps on a single mds and allows mds replica... - 11:49 PM Revision 15dd4f72 (ceph): client: fix mds touch
- 11:49 PM Revision 24a26c66 (ceph): client: fix trim_caps
- We can't blindly remove caps from inodes because we need at least one cap
for any inode in our cache. Try to trim no... - 11:41 PM Revision 008e02f5 (ceph): .gitignore: update for test_libceph_build
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 11:38 PM Revision 283da885 (ceph): testrados: add NUM_THREADS
- We can now test librados access with multiple simultaneous threads from
a single process.
Signed-off-by: Colin McCab... - 09:30 PM Revision 13909c24 (ceph): global_init: make global_init_set_globals static
- Now that the libraries have been de-globalized, this funtion should only
be used by global_init itself.
Signed-off-b... - 09:08 PM Revision 7dd919c2 (ceph): CephFSInterface: de-globalize
- CephFSInterface: de-globalize
liceph: add ceph_get_mount_context
Signed-off-by: Colin McCabe <colin.mccabe@dreamhos... - 08:14 PM rgw Feature #1223 (Resolved): user removal should be (optionally) removing all user buckets
- Done, commit:ba436ed6d0b89510d2f150d0da208fad737d7915.
- 09:37 AM rgw Feature #1223 (Resolved): user removal should be (optionally) removing all user buckets
- 08:09 PM rgw Bug #1208 (Resolved): rgw: make PUT atomic
- Done, as of commit:4298b59ac5686b1739f87b88407b205dfdecee8e. Also implemented atomic copy and atomic multipart complete.
- 06:55 PM Revision 1ddc5521 (ceph): Fix C_NotifyComplete::notify
- C_NotifyComplete::notify was broken in an earlier change where I removed
a derr() without removing the if statement b... - 06:15 PM Revision 178d63e4 (ceph): FileStore: ensure that CLONE_RANGE ranges are aligned
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 06:15 PM Revision f6320fee (ceph): ReplicatedPG,OSD: clone srces need to be non-degraded
- do_op: check that clone_range src sobjects are present on
replicas as well. Also, clean up src_obc map if we encount... - 06:15 PM Revision 5175851a (ceph): ReplicatedPG: make sure that rollback target is not degraded
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 06:15 PM Revision 2fbf215b (ceph): PG: assemble_backlog - fix comparison error, tail is non-inclusive
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 06:05 PM Revision 0115855a (ceph): debian: obsync depends on python-xattr
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:00 PM Revision a1c9bb35 (ceph): cleanup
- 06:00 PM Revision b00eb9f9 (ceph): osd: clean up pool/pg_stat_t structs slightly
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 06:00 PM Revision 5c65661b (ceph): mon: order degraded before unfound
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:52 PM Revision 5af5b9f0 (ceph): libceph: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:52 PM Revision 6696b0ff (ceph): libceph: build without globals
- libceph, libosdc, libclient: depend on libcommon rather than libglobal.
No more globals in libceph.
Signed-off-by: C... - 05:44 PM Revision bfef6c86 (ceph): filestore: remove warning
- 05:37 PM Subtask #1231: NUM_THREADS=3 testrados segfaults
- Wrote a small test, commit:6a3626d373f42cb1750edbdecd050a3cf0606dd7, that also seems to be exhibiting odd behavior. I...
- 05:13 PM Subtask #1231 (Rejected): NUM_THREADS=3 testrados segfaults
- *NUM_THREADS=3 ./testrados* segfaults most of the time.
It seems to segfault more when the threads are actually in... - 05:35 PM Revision 00c5539d (ceph): Filer, Journaler: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:56 PM Bug #1230 (Can't reproduce): cfuse: rm -r missed a directory
- kernel_untar_build
logs on fatty - 04:52 PM Linux kernel client Bug #1226: kclient can't read newly-created file with a custom layout
- From a brief look I think there's a problem with the way it's dropping and refreshing caps to effect the layout change.
- 01:05 PM Linux kernel client Bug #1226 (Resolved): kclient can't read newly-created file with a custom layout
- After using cephfs to set a custom layout on a file, the kernel client can't read that file until the file has been w...
- 04:43 PM Bug #1228: mds bug: testceph hangs forever in Client::unmount
- *steal*
- 03:48 PM Bug #1228 (Resolved): mds bug: testceph hangs forever in Client::unmount
- Sometime between 0.29 and 0.28.1, testceph broke.
I tried to bisect it, but there were other problems in testceph ... - 04:24 PM Feature #1229 (Resolved): rados load generator
- - read/write mix
- data set size
- object size distribution
- workload skew (hot vs cold objects)
- "thread" coun... - 02:39 PM Subtask #846 (Resolved): De-globalize SimpleMessenger, etc.
- I just did a short review and didn't find any more globals in the libraries, except trivial things like an incrementi...
- 02:33 PM Subtask #1227 (Rejected): write tests of libceph, librgw, librados library thread-safety
- write tests of libceph, librgw, librados library thread-safety
- 02:21 PM Subtask #845 (Resolved): g_conf should not be defined in library code.
- resolved
- 12:48 PM Bug #1192 (Resolved): multi-client stall under heavy write load
- 12:45 PM Bug #1213 (Can't reproduce): test case demonstrating read vs write throughput
- Well I ran a couple tests locally and on a 3-node cluster and didn't actually see any evidence of this. Combined with...
- 11:16 AM CephFS Bug #1225 (Resolved): client: recallstate removing caps when it shouldn't
- - can't have an inode with no caps
- cap lru needs to align with dentry lru to get reasonable results - 11:15 AM Feature #1217 (Resolved): identify key performance/health metrics for osd
- the per-cluster stats are all included in pg dump: per-pg missing, degraded, unfound. sampling that and tracking per...
- 10:57 AM rgw Feature #1224 (Resolved): s3-tests should test 100-continue
- should try to PUT with 'expect: 100-continue' and bad credentials.
- 03:35 AM Bug #325: writes starve reads
- This problem seems to be fixed with the latest kclient and server. Cannot reproduce it now.
- 12:25 AM Revision 86b41ff9 (ceph): ObjectCacher.cc: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:06 AM Revision 1fc60a2b (ceph): filestore: don't bother doing fiemap on small ranges
- 12:06 AM Revision 757926a1 (ceph): rgw: restructure permissions verification
- apparently 100-continue was broken
06/23/2011
- 11:27 PM Revision bd5b62aa (ceph): Merge remote branch 'origin/osd_metrics'
- 11:26 PM Revision c260b254 (ceph): update .gitignore
- 11:17 PM Revision f347ea14 (ceph): client/Client.cc: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:57 PM Revision 81900994 (ceph): qa: wget -q for kenrel_untar_build.sh
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:46 PM Revision d1a75559 (ceph): librgw: eliminate globals
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:46 PM Revision 88e9004f (ceph): update .gitignore
- 10:28 PM Bug #1076 (Resolved): avoid sparse read for small reads
- Effectively done in commit:1fc60a2baa6c4b5039ffe61e6443ccce53c0cc9c. The filestore will not do a fiemap on small (con...
- 10:04 PM Revision 76dae02d (ceph): librados, librbd, librgw: remove dep on libglobal
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:03 PM Revision 089255eb (ceph): osd: reorder, cleanup prologger metrics
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:57 PM Revision 74aaa022 (ceph): osd: add sub_op instrumentation
- pull subop
push subop
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:57 PM Revision 6e78e245 (ceph): osd: instrument pushes, pull requests
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:57 PM Revision 13b02890 (ceph): osd: fix loadavg check
- Use the syscall, stupid.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:57 PM Revision a9b107ff (ceph): osd: remove lame rlsum/num
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:57 PM Revision 6a3741a6 (ceph): osd: instrument readable latency too
- Time before a write is readable (not necessarily on disk). Note that if we
get the commit first (e.g. writeahead jou... - 09:57 PM Revision 560c0826 (ceph): osd: clean up numpg*, hb*
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:57 PM Revision 2506500f (ceph): osd: instrument osdmap updates
- Messages, epochs, dup epochs.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:57 PM Revision dfa6c8d2 (ceph): osd: rename peer_stat_lock -> stat_lock
- It protects osd_stat too.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:57 PM Revision d3794840 (ceph): osd: rip out old peer stat sharing
- This never really worked, and has bitrotted, and the stats it's calculating
are mostly redundant anyway.
Signed-off-... - 09:57 PM Revision c72cfe05 (ceph): osd: rename push applied, committed completions
- Be consistent.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:57 PM Revision 74f4b4fe (ceph): osd: fix up log_op_stats to actually work
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:57 PM Revision 6c5a7d5f (ceph): osd: clean up client op stats
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:25 PM Revision 09f6d9f1 (ceph): osd: fix compilation
- 09:25 PM Revision d593b394 (ceph): Revert "osd: make it compile on older boost libraries"
- This reverts commit 116241ef106e6a88a84482f7f0f21219a6d29e33.
- 08:45 PM Revision cbc06f0c (ceph): test_ceph_argparse: fix build
- When this was merged, libcommon.a was not changed to libcommon.la
Signed-off-by: Colin McCabe <colin.mccabe@dreamhos... - 08:39 PM Revision 80df03d8 (ceph): push global_init_set_globals into preinit callers
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:32 PM Revision 116241ef (ceph): osd: make it compile on older boost libraries
- 08:24 PM Revision c01a734d (ceph): Merge branch 'next'
- Conflicts:
src/librbd.cc - 08:16 PM Revision 7cef3923 (ceph): Merge remote branch 'origin/stable' into next
- Conflicts:
src/librbd.cc - 08:13 PM Revision 7e33d122 (ceph): create test_librados_build
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:33 PM Revision ede3a0a3 (ceph): librados: fix end_block calculation for aio_read
- Conflicts:
src/librbd.cc - 07:32 PM Revision 39da1076 (ceph): librados: fix end_block calculation for aio_read
- Conflicts:
src/librbd.cc - 07:31 PM Revision 44759452 (ceph): librbd: track block_ofs instead of ofs in aio_sparse_read callback
- 06:29 PM Revision e477e65b (ceph): Add ceph_argparse unit test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:45 PM Revision f1f75dff (ceph): ceph_argparse_witharg: fix dashes in args
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:40 PM Revision 4549501c (ceph): common/Throttle: Remove unused return type on Throttle::get()
- Signed-off-by: Jim Schutt <jaschut@sandia.gov>
Signed-off-by: Sage Weil <sage@newdream.net> - 05:11 PM Revision c23d4c20 (ceph): libceph: delogbalize, again
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:54 PM Revision 92a3a472 (ceph): Makefile: add missing common/dout.h header to dist tarball
- Signed-off-by: Sage Weil <sage@newdream.net>
- 03:17 PM Feature #1217: identify key performance/health metrics for osd
- the current set of osd metrics:...
- 01:19 PM Feature #1222 (Resolved): configurable way to disable fiemap use in the osd
- Just hit an ext4 fiemap bug, we should be able to disable it through ceph.conf.
- 12:51 PM Bug #1186 (Closed): Cluster won't recover, OSD's go up and down again (and stay down)
- I'm going to close this one as well. I formatted my cluster today and started with a fresh v0.29.1 cluster, had to du...
- 11:35 AM Bug #1221 (Resolved): journal dump and reset don't work with cephx on
- Got a user report and tested this myself with current stable (commit:b34e195a46e8fc6eba0099b517685a205ce86061).
- 12:40 AM Revision e2e96b31 (ceph): Enable core dumps when running the rbd cli tool.
- 12:34 AM Revision 9eea6b93 (ceph): Handle rbd config with no properties.
- 12:32 AM Revision 122ed286 (ceph): Wait for rbd devices to be removed before removing the module.
- 12:13 AM Revision 5da06628 (ceph): librados: fix end_block calculation for aio_read
06/22/2011
- 11:41 PM Revision a8eac4fa (ceph): uclient: change snaprealm refcounting to erase at 0 refs.
- This is more idiomatic and less confusing than having 0 refs be
still alive, and -1 be the point of destruction.
Sig... - 11:41 PM Revision e8b45732 (ceph): AnchorServer: fix return values in add().
- I accidentally didn't set the return value in my new branch. (Not
that they're read anywhere.) While I'm at it, docum... - 11:40 PM Revision 823a05c9 (ceph): proflogger: only register one event for all loggers
- If we already did logger_start(), don't schedule a new event.
Signed-off-by: Sage Weil <sage@newdream.net> - 11:36 PM Revision efeac9f3 (ceph): Add status logging into task rbd.
- 11:36 PM Revision fb016548 (ceph): Wait for block device to show up in task rbd.dev_create.
- 11:04 PM Revision 9a3a685b (ceph): OSDMap: kill some deadcode, deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:04 PM Revision 629ac0da (ceph): librados: deglobalize, again
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:03 PM Revision dfc3a851 (ceph): Add rbd modprobe method.
- 11:03 PM Revision 2d644619 (ceph): Add rbd -> /dev mapping method.
- 11:03 PM Revision deb2fe5c (ceph): Add rbd.mkfs method.
- 11:03 PM Revision 789de09f (ceph): Add rbd.mount method.
- 11:03 PM Revision c0482793 (ceph): Add combined rbd task that runs all the subparts.
- 11:03 PM Revision d9e343ea (ceph): Remove rbd from TODO file.
- 11:03 PM Revision bd46d753 (ceph): Add rbd task with create_image function.
- 11:03 PM Revision b0c8b1d7 (ceph): Ensure suite components are run in alphabetical order.
- 11:03 PM Revision 1a68b6fb (ceph): Remove unused import.
- 10:53 PM Revision 0559f84b (ceph): librados: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:48 PM Revision 8fd24723 (ceph): mount.ceph, rbd: Output errors to stderr, not stdout.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 10:48 PM Revision 21df29d0 (ceph): mount.ceph, rbd: Avoid clobbering errno with printf and strerror calls.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 10:48 PM Revision 151beff4 (ceph): mount.ceph, errno: Avoid printing binary, clarify message.
- Payload is a binary key, printing it to stdout is not helpful.
Key name is often "client", make the message easier t... - 10:23 PM Revision 02a2518e (ceph): run-cli-tests: Pass through CCACHE_DIR and such env vars.
- Commit 7cd50f29d5cbf8deb64d00318b39c281119c0e03 makes the binaries
use libtool's "executable wrappers", which will tr... - 10:00 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- 0.3 seconds for kernel mount... 0.01s for cfuse.. I'm going to work on tracking this down..
- 09:05 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- ok, but are you saying that a native mount is also .5s for touch? that's no good either!
- 04:59 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- Ahh I don't seem to have access to close it...
- 04:58 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- Ahh okay... file create isn't really any slower than on native ceph kernel mount (rather than re-export)... so this i...
- 03:42 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- That's faster... 0.5s for a touch create.
As a reference local filesystem and another cfs exported over nfs runs sub... - 03:14 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- It looks like after the mknod nfsd is calling write_inode via commit_metadata() in fs/nfsd/vfs.c. This triggers cap ...
- 09:49 PM Revision 2305a7a4 (ceph): librbd: Remove derr uses
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:49 PM Revision c9cf19c5 (ceph): Remove "#include debug.h" from config.h
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:31 PM Bug #1184 (Resolved): osd: snapset, pg info stats needs to be projected like obs
- 09:22 PM Revision c9da3076 (ceph): uclient: Update statfs to match the kernel client and its block sizing.
- Make it better match the kernel client, and its scheme to use a large
block size so we don't overflow 32-bit systems.... - 09:22 PM Revision bc273957 (ceph): AnchorServer: overwrite old Anchor backpointers when proper.
- Sometimes when we do an AnchorTable update, it's because the inode
in question got moved. However, if the inode had m... - 09:22 PM Revision fd9bf2d9 (ceph): mds: Include the pinner in nested auth pin debugging output.
- This will make identifying stuck pins a lot easier.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 09:22 PM Revision 787f03cb (ceph): uclient: only change the auth_cap if the mseq is newer
- Previously we just trusted the MDS' reported auth status, which meant
that even if the MDS was always right it could ... - 09:22 PM Revision 7fcf29d6 (ceph): uclient: fix flush_caps(Inode*,int)
- This function was just broken before. You need to be setting
flushing_cap_tids for the caps you're actually flushing,... - 08:58 PM Revision fc25647e (ceph): Merge branch 'next'
- Conflicts:
src/librbd.cc - 08:57 PM Revision 6a1572c7 (ceph): client: always use get_snap_context() accessor
- There were a few places where we were using the member directly and not the
accessor, which rebuilds the internal val... - 08:40 PM Revision 0bf1dc33 (ceph): librbd.cc: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:37 PM Revision 30c47566 (ceph): librbd: track block_ofs instead of ofs in aio_sparse_read callback
- 06:42 PM Revision cabec5cb (ceph): Merge branch 'wip_snap_trimmer'
- 06:41 PM Revision 05c3edfe (ceph): ReplicatedPG: update pull_info and data_subset in sub_op_push
- data_subset and data_needed in pull_info need to be updated with the
results of running calc_clone_subsets in sub_op_... - 06:41 PM Revision 79f76dca (ceph): ReplicatedPG,PG: update snap_collections on replica
- Previously, snap_collections did not get updated on the replica. As a
result, snap collections would not necessarily... - 06:41 PM Revision 33395c8b (ceph): ReplicatedPG: split get_obs_to_trim from snap_trimmer
- Factor code for getting next collection/objects to trim into
get_obs_to_trim.
Signed-off-by: Samuel Just <samuel.jus... - 06:41 PM Revision e53c60db (ceph): ReplicatedPG: factor object trimming out of snap_trimmer
- object trimming has been moved to trim_object which is called from
snap_trimmer.
Signed-off-by: Samuel Just <samuel.... - 06:41 PM Revision 34cb737f (ceph): ReplicatePG,PG: SnapTrimmer state machine skeleton
- Currently, snap_trimmer does not wait for replicas to apply the object
removal repops before updating the info with t... - 06:41 PM Revision 923617dc (ceph): ReplicatedPG, PG: fill out state machine
- Move snap_trimmer functionality into snap_trimmer_machine.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com> - 06:41 PM Revision 3f4e11e1 (ceph): ReplicatedPG: Replica collection removal
- Previously, snap collections and their contents on the replica would be
removed by the primary via a shipped transact... - 06:41 PM Revision 965aadc4 (ceph): ReplicatedPG: clean up OpContext use in do_op
- do_op, make_writeable, and friends now update new_obs, new_snapset, and
new_stats rather than updating those structur... - 06:40 PM Revision 7f298660 (ceph): FileStore: die if we get an ENOENT on a clone
- During recovery, ignoring this would allow a bug in the osd to
erroneously mark the pg clean with incorrectly recover... - 06:07 PM Revision 52e03b0a (ceph): librbd: fix read_iterate for the case where block ofs is not 0
- Conflicts:
src/librbd.cc - 06:05 PM Revision 9edb566c (ceph): librbd: fix read_iterate for the case where block ofs is not 0
- Conflicts:
src/librbd.cc - 06:04 PM Revision 59ea9f8b (ceph): Revert "debian: ceph-client-tools depends on librados2, librbd"
- This reverts commit 96ef8a67bcf4a4a43f0a5c38224314abdd88a12c.
- 06:01 PM Revision b34e195a (ceph): librbd: fix read_iterate for the case where block ofs is not 0
- 05:47 PM Revision 7cd50f29 (ceph): Makefile.am: use libtool convenience libraries
- Use libtool convenience libraries rather than explicitly forcing .a
files (static code archives) to be generated or i... - 05:10 PM Revision 10c08e1c (ceph): Merge branch 'next'
- 05:09 PM Revision 0fe6a15d (ceph): Merge remote branch 'origin/stable' into next
- Conflicts:
src/librbd.cc - 04:39 PM CephFS Bug #1199 (Resolved): 4-MDS fsstress: remote ino lookup assert
- Pushed to master in commit:bc273957af2224d028626a961fdb1bf3bc153f40 and commit:e8b45732a1e77bf4811335ae7b8c97c72cd342ce.
- 04:37 PM CephFS Bug #1178 (Resolved): cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- Pushed to master in commit:787f03cbd7489cb31f5feeec08f84d0b68aa5c0a.
- 04:34 PM Revision a422d1a5 (ceph): rbd: --help/-h
- 03:56 PM Revision 1635bff4 (ceph): osd: fix merge
- Signed-off-by: Sage Weil <sage@newdream.net>
- 02:52 PM Messengers Bug #1220 (Resolved): failed assert: peer_addr > msgr->ms_addr
- I got this crash soon after OSD startup during a teuthology run:...
- 11:20 AM Revision d42da230 (ceph): signals unit test: fix
- Ensure that the signal waits for us to call sigsuspend, rather than
getting delievered to another thread. We do this ... - 11:16 AM Bug #1201 (Resolved): rbd export doesn't match import
- Fixed with commit:b34e195a46e8fc6eba0099b517685a205ce86061.
- 10:56 AM Feature #1219 (Closed): qa: migrate mds rename tests to teuthology
- qa/mds/rename
- 10:36 AM Feature #1215: qa: ability to run radosgw
- Here's how I think this should work:
- assume apache2 and libapache2-mod-fcgid (or whatever) debs are always insta... - 10:13 AM Feature #1218: perfcounter collectd plugin
- Background: http://collectd.org/
plugins are queried for information regularly (the term used is plugins "dispatch... - 08:46 AM Bug #1152 (Closed): Mon getting killed by OOM killer
- I'm going to have close this one for now. By accident I screwed up the data directory of my only monitor, which is no...
- 06:09 AM Revision e2144205 (ceph): librbd: sparse read cleanup, fixes
- Conflicts:
src/librbd.cc - 05:45 AM Revision 6ddee81e (ceph): librbd: fix and cleanup a bit read_iterate
- 05:43 AM Revision e4f7d8e1 (ceph): librbd: fix and cleanup a bit read_iterate
- 05:43 AM Revision bfe0d55c (ceph): librbd: sparse read cleanup, fixes
- 05:43 AM Revision 52e6e852 (ceph): rgw: some cleanups
- 03:56 AM Revision e1f8125d (ceph): add a bunch of workunits
- 03:31 AM Revision 6725e748 (ceph): qa: simple rbd import/export test
- Fixes: #1211
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 12:05 AM Revision 9761810c (ceph): Objecter.cc: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
06/21/2011
- 11:23 PM Revision df2e3bcb (ceph): initialize g_ceph_context in common_preinit
- Initialize g_ceph_context in common_preinit rather than in a global
constructor.
Add comments to all of the major in... - 11:16 PM Revision a66d0264 (ceph): CephContext: not copyable
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:58 PM Revision 74e5c390 (ceph): make misc items as not copyable
- These structures are all not copyable. Make sure we get a link error if
we try.
Signed-off-by: Sage Weil <sage.weil... - 10:57 PM Revision 4e244f2a (ceph): [dex]list: not copyable
- The list items are embedded in other structures; copying the list will lead
to duplicated references and other badnes... - 10:57 PM Revision d3d11596 (ceph): msgr: avoid copying Pipe* xlist
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:54 PM Revision bdb7991c (ceph): Merge branch 'stable' into next
- Conflicts:
src/osd/OSD.cc - 10:54 PM Revision ce6f5788 (ceph): mds: make avoid copying sessionmap by_state xlist
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:54 PM Revision f94e2ab6 (ceph): osd: do misdirected request checks before pg active test
- We can process+discard a misdirected or invalid request without the pg
being active. Reorder a few other checks too.... - 10:53 PM Revision c17d9c0b (ceph): mds: avoid copying snap client_caps xlist
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:53 PM Revision cdb500e5 (ceph): mon: do not copy subscription xlist
- xlist is not copyable.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 10:52 PM Revision abafef3b (ceph): qa: clean up after snaptest-multiple-capsnaps.sh
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:46 PM Revision 0763221b (ceph): Dispatcher: prevent copying
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:09 PM Feature #1211 (Resolved): qa: rbd test for sparse image import/export
- 11:00 AM Feature #1211 (Resolved): qa: rbd test for sparse image import/export
- import and re-export and verify correct result
maybe also verify sparseness appears correct (by mapping extents vi... - 07:54 PM Revision 5df7b23d (ceph): add kclient_ prefix to trivial_sync, dbench
- 07:47 PM Revision 2e7d06c6 (ceph): msgr: only SO_REUSEADDR when specific port is specified
- In general SO_REUSEADDR is slightly dangerous, but avoids waiting for the
timeout when restarting servers. This is i... - 07:29 PM Revision 888e880c (ceph): mon: add 'log ....' command
- Just sticks something in the central log.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:29 PM Revision 4c1cb28c (ceph): mon: fix log state trimming
- We need to stash last_consumed in order for trimming to work.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:28 PM Revision 282575a2 (ceph): Merge branch 'wip-globalfoo'
- 07:28 PM Revision 112e5692 (ceph): Convert g_ceph_context to a pointer.
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:07 PM Revision 89dca916 (ceph): gitignore: add test_libcommon_build
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:40 PM Revision fbf3f4ac (ceph): common_init: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:40 PM Revision 209d1407 (ceph): Thread.cc: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:40 PM Revision b64a48b7 (ceph): DecayCounter: take a utime_t in ::decode
- We need to initialize the last_decay time of the DecayCounter when
decoding it. This is not found in the encoded info... - 06:40 PM Revision b02c8d65 (ceph): Create test_libcommon_build
- Create test_libcommon_build to test for references to g_ceph_context and
other libglobal symbols from within libcommo... - 06:40 PM Revision 1ef90997 (ceph): inode_load_vec_t: de-globalize
- Don't call ceph_clock_now from the constructor of inode_load_vec_t.
Update the classes that contain inode_load_vec_t ... - 06:31 PM Revision c48540ae (ceph): filestore: fix fiemap
- 06:31 PM Revision 39e2c6e8 (ceph): rbd: fix read_iterate for sparse read
- 06:22 PM Revision 5ad52aff (ceph): rgw: use adapting window for put obj
- 06:16 PM Revision 9e9cec69 (ceph): AuthNone: encode entity name in authorizer
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:02 PM Revision 05e2ba9b (ceph): Save the flavor of a run in the summary file.
- 05:00 PM Revision 330ec41f (ceph): Add teuthology-suite, to run multiple tests in a batch.
- 04:59 PM Revision fb1a8757 (ceph): Initial import.
- 04:51 PM Revision 7f92d370 (ceph): osd: warn instead of error if cluster addr specified but not public addr
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:48 PM Subtask #1164 (Resolved): initialize g_ceph_context in common_preinit
- implemented in df2e3bcb2ac790e179e97f6b8017b6fa6a8087bf
- 03:53 PM CephFS Bug #1195: ceph mds crash on version upgrade
- pushed fix commit:cc644b842261dbeefde804ed999061b8733a9190 to stable branch
- 02:59 PM Feature #1218 (Resolved): perfcounter collectd plugin
- 02:58 PM Feature #1217 (Resolved): identify key performance/health metrics for osd
- per node-
ops/sec
bw/sec
read/write latency
per cluster-
missing/lost/degraded objects
pg status over time
o... - 02:53 PM Feature #1216 (Resolved): qa: run s3-tests in suite
- 02:53 PM Feature #1215 (Resolved): qa: ability to run radosgw
- set up apache, radosgw
- 02:34 PM Feature #1214 (Resolved): qa: flock/fnctl lock tests
- 02:33 PM Bug #1141 (Closed): osd: misc snap bugs
- 02:33 PM Tasks #1155 (Resolved): Replica does not remove snap collections
- 02:26 PM Bug #1144 (Resolved): snap_trimmer
- 02:24 PM Bug #1213 (Can't reproduce): test case demonstrating read vs write throughput
- 01:03 PM Feature #1212 (Resolved): teuthology: ability to restart daemons while other tasks are running
- e.g., ability to restart osd, mds while a workload is running
- 01:01 PM Bug #1018 (Can't reproduce): error on building ceph on red hat 5.5
- 01:01 PM Bug #1032 (Closed): osd: Marked down and become zombies after killing
- i'm going to close this one. if you see a zombie with a btrfs error in dmesg, it's a btrfs problem. if you don't, l...
- 12:33 PM Bug #250: mon: delete old states to avoid filling disk
- log trimming fixed by commit:4c1cb28cb6cbb3d23a1736d02f6af524481c886a
osdmap, mdsmap, monmap still need trimming, ... - 11:10 AM Bug #250: mon: delete old states to avoid filling disk
- We should probably just do this in the monitor. The logm is never read so it's safe to delete whenever, unlike the ot...
- 10:59 AM Bug #250: mon: delete old states to avoid filling disk
- I keep running into this, my monitors keep running out of inodes which causes them to stall.
Right now I have cron... - 12:29 PM Subtask #1160 (Resolved): introduce CephContext to some structures in common/
- 10:58 AM Feature #1210 (Resolved): qa suite: schedule nodes and run suite in parallel
- extend or wrap teuthology-suite to schedule the right number of nodes for each test and run them in parallel. probab...
- 10:56 AM Feature #1209 (Resolved): qa: node locking/reservation
- a simple client/server protocol (REST?)
lock/unlock N nodes (of type T?)
record owner and lock time
anything else? - 10:55 AM Tasks #1175 (Resolved): qa: multi-test runner
- 10:55 AM Feature #1123 (Resolved): qa: small but completely functional suite
- 10:53 AM rgw Bug #1208 (Resolved): rgw: make PUT atomic
- using clonerange?
- 09:58 AM Feature #1207 (Resolved): qa: run daemon(s) through valgrind
- Run one or more components through valgrind. Maybe in teh ceph: section something like...
- 09:21 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
> - when the client sends caps back to the MDS, some of them are dirty. The MDS, as a rule, will always make sure...- 05:40 AM Revision 37a97622 (ceph): include/atomic.h:fix build for non-atomic-ops case
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:09 AM Revision 8f1beb1b (ceph): rgw: put data using a window
06/20/2011
- 11:41 PM Revision d5ec8b77 (ceph): MonClient: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 4615913c (ceph): hex.cc: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision f6c7343f (ceph): auth: more deglobalization
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 2b22c32b (ceph): OSDMap: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 40532049 (ceph): DoutStreambuf: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision fae8dde3 (ceph): Messenger: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 83020ad0 (ceph): assert: de-globalize
- Implement a simple registration system so that assert output can still
show up non-interleaved in daemons and utiliti... - 11:41 PM Revision 8147350c (ceph): MDSMap: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 6ff09ef5 (ceph): Move signal, pidfile to global
- Move the global parts of signal and pidfile handling into global/
Signed-off-by: Colin McCabe <colin.mccabe@dreamhos... - 11:41 PM Revision fc434386 (ceph): Create global_context.cc to house g_ceph_context
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 91065f48 (ceph): auth: AuthSupported: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision b0cb4808 (ceph): KeyRing: derr -> lderr
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision ba611910 (ceph): msg: remove globals
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision 0fc6ef2e (ceph): LogClient: remove accidental g_conf references
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision ba5b7331 (ceph): tcp.cc: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:41 PM Revision dcc64482 (ceph): ProfLogger: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:37 PM Revision 63bbaddd (ceph): move common_init to global_init
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision b44ead76 (ceph): mon/MonMap: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 02c2ee16 (ceph): mon/MonClient: use ldout instead of dout
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision c95062c7 (ceph): libclient: remove client/SyntheticClient.cc
- client/SyntheticClient.cc is only of interest to csyn. The other users
of libclient don't need it.
Signed-off-by: Co... - 11:35 PM Revision c46814db (ceph): Makefile: put LIBGLOBAL LDFLAGS into LIBGLOBAL_LDA
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision da2d4297 (ceph): Create ProfLoggerCollection to hold ProfLoggers
- Rather than having global variables, create ProfLoggerCollection to
store the current set of ProfLoggers. This collec... - 11:35 PM Revision 17d22a96 (ceph): auth: CryptoKey, CryptoAES: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision ab75f2ab (ceph): auth: KeyRing, RotatingKeyRing: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 76c5f248 (ceph): auth: AuthAuthorizeHandler: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 51ffa7b1 (ceph): auth: CephxKeyServer: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 01038faf (ceph): auth: remove remaining globals
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 5d5a7116 (ceph): common_init: misc globals cleanup
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision f82611d7 (ceph): common: de-globalize LogClient, ProfLogger, Timer
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision d224876e (ceph): ThreadPool: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 8069e83c (ceph): common/DecayCounter: deglobalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:35 PM Revision 8a0d4864 (ceph): SimpleMessenger: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:34 PM Revision 7735b35f (ceph): g_clock.now -> ceph_clock_now
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 9085597b (ceph): run_cmd: ret empty string on success;err otherwise
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 1dbb0ac8 (ceph): auth: de-globalize TicketManager, TicketHandler
- de-globalize CephXTicketManager and CephXTicketHandler. Also
de-globalize two static functions.
Signed-off-by: Colin... - 11:22 PM Revision be8fe54a (ceph): auth: de-globalize AuthClientHandler and friends
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision d03c96e5 (ceph): auth: de-globalize AuthServiceHandler and pals
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision d6adc4e7 (ceph): auth: de-globlize CephXAuthorizer
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 2452ef44 (ceph): common/Clock: add deglobalized clock stuff
- Add clock functions which don't use globals. Remove Clock::recent_now.
Signed-off-by: Colin McCabe <colin.mccabe@dre... - 11:22 PM Revision c0e804db (ceph): ObjectStore: clean up includes a little bit
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 07ff69a7 (ceph): include/Completion: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision ff626f0e (ceph): bufferlist: get rid of derr usage
- bufferlist doesn't need to use dout/derr. Only one function used it, and
that was only to return an error string.
Si... - 11:22 PM Revision bf2c226a (ceph): include/atomic.h: whitespace cleanup
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 03c83799 (ceph): include/atomic cleanup
- * Don't allow copying of class atomic_t.
* Remove common/Spinlock.h because it's unecessary
* SimpleMessenger: use at... - 11:22 PM Revision 0fa77948 (ceph): common/Finisher: use ldout
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision f2bd39fa (ceph): common/MemoryModel: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision 6c58da04 (ceph): include/Context.h: de-globalize
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:22 PM Revision d80b331f (ceph): lockdep: code cleanup and de-globalization
- common/Mutex.cc: this is a duplicate of the code in common/lockdep.cc.
Delete the duplicate code.
common/lockdep.cc:... - 11:17 PM Revision 823eb267 (ceph): Thread.cc: clean up
- Don't implement everything in the header file. It slows down compilation
and usually runtime.
Signed-off-by: Colin M... - 11:06 PM Revision 9678ad3f (ceph): lockdep: add lockdep_register_ceph_context,cleanup
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:06 PM Revision ee5502d2 (ceph): Remove cdout
- cdout just duplicates the functionality of dout.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 11:06 PM Revision 5a612f95 (ceph): Create de-globalized versions of the dout macro
- The de-globalized versions are available in dout.h
Versions relying on globals continue to be available in debug.h
S... - 11:05 PM Revision 620d0784 (ceph): signal: const cleanup, don't use derr
- In signal.cc: const cleanup, don't use derr, add a comment.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 11:01 PM Revision 128bfcfd (ceph): rgw: put_obj_data doesn't need mtime
- 11:01 PM Revision 1e5cb068 (ceph): rgw: put_obj_data uses aio
- This is just a trivial fix, making the etag calculation done while the io
is pending. Still some room for improvement. - 10:58 PM Revision 10171cad (ceph): filestore: fix fiemap
- 10:58 PM Revision ca2f4e28 (ceph): rbd: fix read_iterate for sparse read
- 09:31 PM Revision bc8cc868 (ceph): Fix bug that thought all >1 node clusters always had core dumps.
- Accidentally shared the stdout between all the runs.
- 09:31 PM Revision e481db13 (ceph): Archive syslog messages while the test was in progress.
- 09:06 PM Revision f78de01a (ceph): encoding: add list<shared_ptr<T>> macros
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:06 PM Revision cc644b84 (ceph): mds: avoid EMetaBlob::fullbit copying with shared_ptr
- The fullbit subclass has a pointer, making it unsafe to copy around. Fix
by adding a undefined copy constructor and ... - 09:02 PM Revision 96ef8a67 (ceph): debian: ceph-client-tools depends on librados2, librbd
- It contains rados and rbd command line tools.
Signed-off-by: Sage Weil <sage@newdream.net> - 06:47 PM Revision 04d10d10 (ceph): qa: tiobench workunit: remove timeout
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 06:28 PM Revision 29e64df9 (ceph): Merge remote branch 'origin/stable'
- 05:09 PM CephFS Bug #1206: NFS reexport file creation lags 1-3 seconds
- If I reexport the ceph filesystem via fuse/nfs, there is *no* lag in creating files.
- 05:02 PM CephFS Bug #1206 (Closed): NFS reexport file creation lags 1-3 seconds
- I'm attaching the kernel logs and mds logs for the creation of a file called scale-product/testfoo3 via a touch runni...
- 04:38 PM Feature #183: qa: xfstests workunit
- I looked at this briefly and it appears surprisingly annoying to do since xfstests normally wants to do the mounting ...
- 04:37 PM Feature #1205 (Resolved): qa: run teuthology clients in parallel
- 04:37 PM Feature #1204 (Resolved): qa: add existing workunits to suite
- 04:29 PM CephFS Bug #1199: 4-MDS fsstress: remote ino lookup assert
- The problem is that if the backtrace for a given inode changed, the AnchorServer didn't overwrite the old Anchor that...
- 03:31 PM CephFS Bug #1199: 4-MDS fsstress: remote ino lookup assert
- Okay, this seems to be an anchortable bug. It's getting the last trace that got committed for the inode in question, ...
- 04:28 PM CephFS Bug #1200 (Duplicate): 4-MDS fsstress remote ino lookup cycle
- Yep, this is the same as #1199.
- 04:09 PM Feature #1196 (Resolved): qa: install kernels via teuthology
- 03:15 PM CephFS Bug #1110 (Resolved): mds: ls -l hangs on concurrent writer
- Created a separate task for the request prioritiziation on the osd, #1203
- 09:22 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Okay, so I found one other issue while testing this patch. So there are two reasons this is slow:
- when we revo... - 08:18 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- > Indeed: It was at epoch 5 which has a "modified" timestamp of 2011-06-15 22:37:48. This matches exactly the time wh...
- 03:14 PM RADOS Feature #1203 (New): osd: priority or fairness osd operations
- The idea is be to prioritize MDS messages in the OSD op queue to minimize latency for metadata writes. See #1110.
- 03:08 PM Bug #1194: kclient: NFS reexport does not survive ceph fs remount
- I was able to reproduce this.
ESTALE shows up in the messages log.
I don't see much in the mds log.
At this poin... - 11:26 AM Linux kernel client Bug #1140: balance_dirty_pages makes Fw cap revocation slow
- commit:afc2c255f157c1ff6ecff5d9ad7979a226a19e1c is a patch that avoids the balance_dirty_pages issue. That will solv...
06/18/2011
- 03:17 PM Feature #1202 (Resolved): qa: workunits to set file layout and verify it works.
- This'll only work on the kclient, of course; no ioctl for cfuse.
- 02:18 PM Bug #1201: rbd export doesn't match import
- P.S. v0.29.1-4-g93623fb
- 02:17 PM Bug #1201 (Resolved): rbd export doesn't match import
- rbd import foo bar output:
rbd import file_pos=0 extent_len=12288
reading 12288 bytes at offset 0
rbd import file_...
06/17/2011
- 11:20 PM Revision 7335fbbf (ceph): librados: return value on non void function
- 11:18 PM Revision 830f48d2 (ceph): update clitests with new usage
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:17 PM Revision 93623fba (ceph): common: fix descriptinos for -i, -n too
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 11:01 PM Revision 1b9575ea (ceph): common: fix generic usage for -D vs -d
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 11:01 PM Revision 6a32f723 (ceph): Autotests run no ceph software, so don't try to do code coverage there.
- 11:00 PM Revision b9c36729 (ceph): rgw: suspend/enable buckets through pool async api
- 11:00 PM Revision e2150a08 (ceph): librados: add async pool op operations
- 11:00 PM Revision 57c542b9 (ceph): Archive cores dumped during test, record test as failed if any seen.
- 10:01 PM Revision 6fb971ff (ceph): qa: organize workunits
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:24 PM Revision b2c803db (ceph): secret.c: use safe_read when appropriate
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:59 PM Revision 7b8bf550 (ceph): mds: fix use-after-free on ESession replay
- Reported by: Sam Lang <samlang@gmail.com>
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 04:26 PM Revision 4efdc52a (ceph): rgw: remove stray semicolon
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:26 PM Revision 4ca8054c (ceph): rgw: cast RGW_SUSPENDED_USER_AUID
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:22 PM Revision 9e4adf0d (ceph): qa: make snap-rm-diff workunit clean up
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 04:22 PM Revision e9f308df (ceph): qa: wget -q
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 04:21 PM Revision 20e3ac2c (ceph): mds: CDir::log_mark_dirty needs to flush before waiting
- We need to flush before waiting or else our completion (and mark_diryt())
will be out of order with respect to other ... - 03:30 PM CephFS Bug #1200: 4-MDS fsstress remote ino lookup cycle
- Looks like this is the same root cause as #1199.
- 03:19 PM CephFS Bug #1200 (Duplicate): 4-MDS fsstress remote ino lookup cycle
- Notice how it's missing an expected ino in a dir that's marked complete, and then it just tries to do the lookup agan...
- 03:12 PM CephFS Bug #1199 (Resolved): 4-MDS fsstress: remote ino lookup assert
- ...
- 02:36 PM Subtask #1177 (Resolved): qa: post-process coverage data
- 02:34 PM Feature #1119 (Resolved): qa: gcov/lcov html output
- 01:45 PM Bug #1152: Mon getting killed by OOM killer
- Humm, these heap dumps are never larger than 0.2MB!
Looking at the total memory allocation over the lifetime of th... - 11:48 AM Bug #1152: Mon getting killed by OOM killer
- I hit the bug again it seems. The OOM killer came around and killed my monitor again. I didn't touch the machine yet,...
- 11:36 AM rgw Feature #1198 (Resolved): rgw user suspension
- This should include:
- the ability to suspend/reenable users
- the ability to disable access to a suspended user'... - 11:33 AM rgw Feature #767 (Resolved): rgw: incremental/large file uploads
- Done, as of commit:8b4b83842cdf2e707efd6c3aded9abbf1ab352d7
- 11:31 AM Bug #1197 (Resolved): osd: make inconsistent state durable
- I'm pretty sure that if we ever get an inconsistent PG, that knowledge needs to be in permanent storage so that we do...
- 11:30 AM Feature #900 (Resolved): qa: thrash in/out
- 11:29 AM Feature #1196 (Resolved): qa: install kernels via teuthology
- 11:12 AM Bug #1192: multi-client stall under heavy write load
- I've caught another stall using your patch, but it is slightly different.
Maybe I waited longer, but this one even... - 10:54 AM Bug #1192: multi-client stall under heavy write load
- Any news?
- 11:04 AM CephFS Bug #1178: cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- Will be testing locally, push later. We had to make sure to check the mseq when setting a new auth_cap based on the M...
- 09:41 AM CephFS Bug #1195: ceph mds crash on version upgrade
- Oh, you have multiple MDSs.. can you dump the same object for whichever one(s) crashed in
13: (SessionMap::decode... - 09:38 AM CephFS Bug #1195: ceph mds crash on version upgrade
- Attached log from mds crash with suggested debugging enabled.
- 09:30 AM CephFS Bug #1195: ceph mds crash on version upgrade
- Attached result of above command.
- 09:25 AM CephFS Bug #1195: ceph mds crash on version upgrade
- Can you 'rados -p metadata get mds0_sessionmap /tmp/mds0_sessionmap' and attach? I'm curious what is in the object t...
- 09:14 AM CephFS Bug #1195: ceph mds crash on version upgrade
- Hmm, Ceph *should* be upgradable to newer versions. These backtraces don't look familiar though and i don't see anyth...
- 08:03 AM CephFS Bug #1195 (Resolved): ceph mds crash on version upgrade
- This may just not be something that's handled in Ceph yet, but while trying to upgrade from stable v0.28 to v0.29, I ...
- 08:59 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Sage Weil wrote:
> Can you confirm that it hung when the fs was full? Because that's normal (and unrelated to this ... - 03:02 AM Revision ad5ff230 (ceph): mount.ceph: specify full path for /sbin/modprobe
- Fixes: #1188
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 03:01 AM Revision f5f563dd (ceph): radosgw_admin: fix cli test
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 02:44 AM Revision ca4e2877 (ceph): Merge branch 'stable'
- 02:44 AM Revision 3a59579e (ceph): mds: fix transitions away from XSYN state
- If we are in XSYN, we need to transition first to EXCL before moving on.
Simply going to the other unstable state isn... - 02:44 AM Revision 1c5f6550 (ceph): mds: when prodding head inode on snapped behalf, avoid XSYN
- If we try to rdlock a snapped inode in snap->sync state, we need to prod
the head inode into SYNC to get the data flu... - 02:44 AM Revision 447ea1dc (ceph): osd: fix snap_trimmer obc ref
- As of fc4cc39 new_repop claims the obc reference (from the clonerange
branch) but this caller didn't get fixed.
Sign... - 02:44 AM Revision bc1782a1 (ceph): osd: fix find_object_context debug output
- Erroneously said get_object_context, confusing developers everywhere.
Signed-off-by: Sage Weil <sage@newdream.net>
06/16/2011
- 11:53 PM Revision 9974b7e5 (ceph): rgw: user suspension
- 11:51 PM Revision 2738ce01 (ceph): Better examples for kclient task.
- 11:11 PM Revision 0b28b96e (ceph): Remove done and abandoned TODO entries.
- 11:09 PM Revision 4f46e4b7 (ceph): Make cfuse and kclient default to all clients.
- 11:09 PM Revision 5b237fc5 (ceph): Support running multiple autotest tests.
- Docs had this from the start, it was just buggy. Still
doesn't support using the "autotest" task more than once,
no e... - 11:09 PM Revision b49c7390 (ceph): Archive autotest result output.
- 10:19 PM Revision 7fe89b79 (ceph): add clock task to sync clocks
- 10:19 PM Revision 55cfd929 (ceph): fix typo in workunits example
- 10:19 PM Revision cebf88d9 (ceph): show role in log prefix when running workunits
- 10:11 PM Bug #1150: kclient: ERESTARTSYS from flock/fcntl locks
- On 3.0.0-rc3 kernel (master branch) the same trouble.
- 09:36 PM Revision 78a3c234 (ceph): Move non-ceph logic out of the ceph task: base dir, archive transfer.
- 09:36 PM Revision 259eb350 (ceph): Refactor ceph mds daemon handling into a separate subtask.
- 09:36 PM Revision 445186b0 (ceph): Refactor ceph healthy waiting into a separate subtask.
- Note, not a context manager, just a task. There's no cleanup.
- 09:36 PM Revision 9353fcc9 (ceph): Move summary inside context.
- This will hopefully make run_tasks usable by tasks.
The ceph task can't use it yet, because it has its
yield inside t... - 09:36 PM Revision 594a3d3f (ceph): Refactor ceph log handling into a separate subtask.
- 09:36 PM Revision 629ad447 (ceph): Move /tmp/cephtest/data creation to where it belongs, in ceph.cluster.
- 09:36 PM Revision 301ab567 (ceph): Move non-ceph logic out of the ceph task: host in use check.
- To avoid every config always listing basic tasks, we silently
add internal.* tasks in front of the task list. - 09:20 PM Revision c35770c4 (ceph): Merge branch 'rgw-multipart'
- 09:19 PM Revision abfa872f (ceph): Merge remote-tracking branch 'origin/master' into rgw-multipart
- Conflicts:
src/os/FileStore.cc
src/rgw/rgw_admin.cc
src/rgw/rgw_fs.h
src/rgw/rgw_log.cc
src/rgw/rgw_rados.cc
sr... - 09:07 PM Revision 3aa6a4d6 (ceph): qa: pjd must run as root
- 08:26 PM Revision 8b4b8384 (ceph): rgw: get multipart list marker to work
- 08:10 PM Revision 77d38e42 (ceph): v0.29.1
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:34 PM Revision 3fc00e37 (ceph): Refactor ceph mon daemon handling into a separate subtask.
- 06:34 PM Revision 2da2d258 (ceph): Refactor ceph osd daemon handling into a separate subtask.
- 06:30 PM Revision 062ddf25 (ceph): Merge remote branch 'origin/mds_xsyn'
- 06:17 PM Revision a379c67d (ceph): rgw: some multipart cleanups, fixes
- 06:04 PM Revision 574cfe98 (ceph): Remove the "ceph mds set_max_mds" call, apparently it's not needed.
- We can always re-add it if and when it's useful.
- 05:39 PM Revision 9746e774 (ceph): Allow running of any function in a task.
- 05:39 PM Revision 14ddb419 (ceph): Start splitting the ceph task into smaller units.
- This lets them nicely have independent error handling;
an example case where that is desirable is that the
old code f... - 05:39 PM Revision 915a366a (ceph): Refactor ceph cluster setup into a separate subtask.
- 05:39 PM Revision eaba08fb (ceph): Add debug logging to contextutil.nested.
- 04:51 PM Bug #1194: kclient: NFS reexport does not survive ceph fs remount
- I reproduced the problem several times before submitting this bug but can't reproduce it now.
I'm going to leave thi... - 02:23 PM Bug #1194 (Resolved): kclient: NFS reexport does not survive ceph fs remount
- NFS doesn't survive restarts/remounts
Reproduce:
create new ceph fs
export via NFS
(on NFS client) nfs mount
(on... - 03:27 PM CephFS Bug #1178: cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- I'm pretty sure the client is getting the right import/export messages, there's just something going wrong in the twi...
- 02:56 PM CephFS Bug #1178: cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- oh, interesting. in that case it won't get an IMPORT event because it wasn't exported.. and the client doesn't find ...
- 02:48 PM CephFS Bug #1178: cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- Oh, so it's an inode where the auth got changed, and it only has the cap for the new auth but it doesn't think it is ...
- 01:48 PM CephFS Bug #1178 (In Progress): cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- Looking at this now.
- 02:11 PM CephFS Bug #1114: NFS export extreme slowdown
- I think I've figured out the more general symptom.
File creation via NFS-ontop-of-ceph takes 1-3 seconds per file.
... - 01:05 PM Bug #1192: multi-client stall under heavy write load
- I'm compiling it now.
- 12:28 PM Bug #1192: multi-client stall under heavy write load
- Can you try branch osd_pg_active? The request you mention gets kept around on the waiting_for_active list when it sh...
- 08:11 AM Bug #1192: multi-client stall under heavy write load
- I tried to upload osd.32.log.bz2, but got this:
Internal error
An error occurred on the page you were trying to... - 08:06 AM Bug #1192 (Resolved): multi-client stall under heavy write load
- I've been having trouble with Linux client stalls under heavy
write load. I think I've got the signature isolated. ... - 11:29 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Can you confirm that it hung when the fs was full? Because that's normal (and unrelated to this issue); if the FS is...
- 01:42 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- > > Otherwise, there's nothing there. If the situation is reproducible, the best would be an mds log generated with ...
- 10:47 AM Bug #1152: Mon getting killed by OOM killer
- Unfortunately, it looks like the heap dumps we have are from after it grew too large and they don't capture the growt...
- 10:36 AM Bug #1152 (In Progress): Mon getting killed by OOM killer
- 09:39 AM CephFS Bug #1193 (Resolved): cmds crashes with stacktrace when using --reset-journal and an invalid rank
- Here are the two stack traces I noticed that are different....
- 09:16 AM Bug #1188 (Resolved): modprobe not found by mount.ceph
- Looks like this is done!
- 08:36 AM Bug #998 (Resolved): qemu/librbd race conditon
- Fixed by commit:ea3db17697b0d6fadf9bb26af0ed0b61773d530a and commit:515f0884d4f55efc21365ccbd94e1e5c22788051.
- 08:17 AM Bug #1191: FAILED assert(!missing.is_missing(soid))
- ...
- 08:02 AM Bug #1191 (Can't reproduce): FAILED assert(!missing.is_missing(soid))
- In 30 seconds, that error happened on 4 out of 8 OSDs. This is with today's stable branch.
That happend after some... - 07:56 AM CephFS Bug #1170: 2-MDS fsstress: SR fails subtree assert
- I'm seeing this with the 0.29 packages, using Ubuntu Natty and Linux 3.0rc3 built from git. Only some of the crashed ...
- 01:08 AM Revision 1dd25b47 (ceph): Add task for mounting with the kernel client.
- 01:06 AM Revision 9d70d048 (ceph): Add function to output a secret key to a file.
- 12:43 AM Revision 5e1768ee (ceph): Merge remote branch 'origin/stable'
- 12:32 AM Revision 515f0884 (ceph): librbd: fix block_completion race condition
- If block completions finished before all of them were scheduled, the
pending_count could reach 0, and the callback co... - 12:06 AM Revision e9e3feea (ceph): rgw: implement list multiparts
- doesn't work completely, getting there
06/15/2011
- 10:52 PM Revision 9d603447 (ceph): Output a summary file when archiving a run.
- For now, just record whether the test succeeded.
- 10:22 PM Revision 98a80719 (ceph): Remove unused imports and variable.
- 09:43 PM Revision 2d63264c (ceph): librbd: add AioCompletion debugging
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:24 PM Revision 4a7a42b5 (ceph): uclient: path_walk should notice when it gets absolute paths.
- Previously it assumed everything was relative, but we need to handle
absolute paths on occasion, and already have the... - 08:06 PM CephFS Bug #1158: Unfinished freeze hangs fsstress
- Unfortunately, adjust_nested_auth_pins never sees the person who actually grabbed it. The others print out the grabbi...
- 06:51 PM CephFS Bug #1158: Unfinished freeze hangs fsstress
- If you can reproduce, you can enable the auth pin set define in mdstypes.h, which tracks who the pinners are.
//#d... - 06:04 PM CephFS Bug #1158: Unfinished freeze hangs fsstress
- Well, it's a nested auth pin....
- 07:32 PM Revision 1a20ba11 (ceph): Tee teuthology log into the archive directory, if archiving.
- 07:11 PM Revision e90e0411 (ceph): vstart: put .journal outside of osd data dir
- This prevents it from getting cleaned up during mkfs. Then we can make it
a symlink, or at the very least avoid dele... - 07:10 PM Revision 2bae22f0 (ceph): Create the archive dir earlier, move out of ceph task.
- 06:59 PM Revision faa855dd (ceph): Add task for running ceph workunits, QA tests that expect cwd to be cep...
- 06:51 PM Revision e00b7fc6 (ceph): Clean up after an autotest run.
- 06:50 PM Revision cd8f31db (ceph): Write config.yaml directly to file.
- 06:49 PM Revision d157e323 (ceph): Move imports to the top.
- 06:27 PM Revision af9879de (ceph): qa: Make snaptest-git-ceph.sh clean up after itself.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 05:16 PM Revision 1aa29320 (ceph): rgw: fix abort multipart upload
- 04:38 PM Bug #998: qemu/librbd race conditon
- Can't get kvm to crash anymore during my latest tests. Two days before, with the same amount of testing, I would have...
- 02:32 PM Bug #998: qemu/librbd race conditon
- I updated the wip-librbd-race branch again - I managed to confuse myself into thinking the block completion patch was...
- 01:06 PM Bug #998: qemu/librbd race conditon
- also at 2510a0a31e3cc742f7d7564887fd93da5203967f...
- 12:50 PM Bug #998: qemu/librbd race conditon
- a crash I could reproduce with 2510a0a31e3cc742f7d7564887fd93da5203967f...
- 01:38 AM Bug #998: qemu/librbd race conditon
- Thank you for the new patch.
I still managed to crash 3 of my VMs, each with a different stacktrace:... - 04:32 PM CephFS Bug #1114: NFS export extreme slowdown
- Here's the time sequence with each x-axis increment being 20 seconds.
Plotted separately to avoid scaling axes and a... - 04:16 PM CephFS Bug #1114: NFS export extreme slowdown
- Here's a 20-second interval time evolution of /sys/kernel/debug/ceph/*/bdi/stats
I can go ahead and graph these.. - 02:41 PM Feature #1189 (Resolved): qa: independent powerc and console from ceph network
- 12:43 PM Feature #1189 (Resolved): qa: independent powerc and console from ceph network
- 01:09 PM Bug #1186: Cluster won't recover, OSD's go up and down again (and stay down)
- Failed an assert in ReplicaActive receiving a query of type other than Query::Missing. ( assert(query.query.type == ...
- 12:58 PM Feature #1190 (Resolved): codingstyle document
- http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml ?
- 12:31 PM Bug #1152: Mon getting killed by OOM killer
- I installed both ceph and ceph-dbg on the machine, I build the debs with dpkg-buildpackage and install those.
The ... - 09:54 AM Bug #1152: Mon getting killed by OOM killer
- There are no symbols.. can you run a cmon that's build with -g (not from a stripped .deb)?
- 06:54 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Sage Weil wrote:
> I do see this:
> 2011-06-10 11:41:04.325925 7f931b987700 mds0.objecter FULL, paused modify 0x7f... - 06:29 AM Bug #1150: kclient: ERESTARTSYS from flock/fcntl locks
- Ok, I will write later about the test results.
- 12:53 AM Revision c49f9471 (ceph): Archive teuthology config.
06/14/2011
- 11:34 PM Revision 27f79fc8 (ceph): auth: remove CephCryptoManager
- It doesn't do anything.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 10:55 PM Revision ea3db176 (ceph): librbd: fix AioCompletion race condition
- If block completions finished before all of them where scheduled, the
pending_count would reach 0, and the AioComplet... - 09:47 PM Revision eced4d6b (ceph): monmaptool: implement --help
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:43 PM Revision 41d9335d (ceph): test/bufferlist: don't alloc large array on stack
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:23 PM Bug #1188: modprobe not found by mount.ceph
- Does the mount.ceph is the modprobe branch work okay? It just specifies /sbin/modprobe.. which is hopefully always t...
- 04:31 PM Bug #1188 (Resolved): modprobe not found by mount.ceph
- # mount cephsrv:/ /mount/point
sh: modprobe: command not found
mount.ceph: modprobe failed, exit status 127
modp... - 08:58 PM Revision 8c97e6c6 (ceph): .gitignore man/Makefile
- blech
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 07:53 PM Revision 954e0966 (ceph): qa: do not use automake for workunit makefiles
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 07:35 PM Revision 40f5ab96 (ceph): qa/workunits/direct_io: fit .gitignore
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 07:27 PM Revision 51b3651b (ceph): radosgw_admin: fix clitest help
- 03:38 PM Bug #998: qemu/librbd race conditon
- I pushed a simpler (and I think correct) fix to the wip-librbd-race branch. Does this fix the problem for you, ar Fred?
- 10:49 AM Bug #998: qemu/librbd race conditon
- I'm having a lot of similar crashes these days, they usually happen during or soon after the startup of the VM.
Al... - 03:16 PM Bug #1186: Cluster won't recover, OSD's go up and down again (and stay down)
- The monitor debugging also seems to have been a problem. Turning that down and restarting the machines has allowed 3...
- 01:33 PM Bug #1186: Cluster won't recover, OSD's go up and down again (and stay down)
- atom2 with cosd daemons killed:
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
r b... - 10:05 AM Bug #1186: Cluster won't recover, OSD's go up and down again (and stay down)
- A bit more information:
for i in {0..9}; do ssh root@atom$i 'uptime'; done
19:01:32 up 7:42, 0 users, load a... - 05:09 AM Bug #1186 (Closed): Cluster won't recover, OSD's go up and down again (and stay down)
- Ok, the title might be somewhat confusing, but so is the issue :)
I'm still trying to get my 40 OSD cluster back i... - 01:21 PM Bug #1187 (Can't reproduce): OSD: OSDMap::decode
- During #1186 I did notice one OSD crash, the backtrace gave me:...
- 12:25 PM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Oh! Sorry, I didn't scroll down. :)
I do see this:
2011-06-10 11:41:04.325925 7f931b987700 mds0.objecter FULL,... - 10:08 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Sage Weil wrote:
> Oh, that log is from the 8th, and doesn't include the fix I pushed on the 9th fixing that particu... - 09:39 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Oh, that log is from the 8th, and doesn't include the fix I pushed on the 9th fixing that particular bug. Did you at...
- 01:41 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Sage Weil wrote:
> Andre, any chance you can pass along more information about the hang you were seeing? Log, or th... - 05:22 AM Revision d2b7e291 (ceph): crbdnamer: man page
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:19 AM Revision e6451525 (ceph): crbdnamer: +x
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:14 AM Revision cca3da1f (ceph): qa: radosgw_admin: fix clitest
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:11 AM Revision 66229c7a (ceph): qa: direct_io: fix warnings
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 04:56 AM Revision 71543246 (ceph): configure: check for boost statechart header
- Partially fixes: #1124
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 12:01 AM Revision 883991a0 (ceph): added thrashosds
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:00 AM Revision 398a3338 (ceph): adding radosbench.py to tasks
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
06/13/2011
- 11:48 PM Revision e304fd09 (ceph): radosgw_admin: log show can dump info in xml, json
- 11:25 PM Revision 86844671 (ceph): Merge branch 'wip-workunits'
- 11:24 PM Revision b91c2c99 (ceph): librados: use empty(), not size()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:23 PM Revision 8ce98b22 (ceph): qa: test_sync_io: zero buffers prior to read
- Be extra paranoid to ensure we read back the correct data and didn't
just find it in undefined memory.
Signed-off-by... - 11:23 PM Revision efe1aa71 (ceph): radosgw_admin: pool create command
- 11:23 PM Revision 28e1a894 (ceph): radosgw_admin: fix log show (with new poolid param)
- 11:18 PM Revision d89c9a13 (ceph): qa: test_sync_io improvements
- Exit (without removing foo) on error.
Allow read- or write-only operation.
Signed-off-by: Sage Weil <sage@newdream.... - 11:03 PM Bug #1150: kclient: ERESTARTSYS from flock/fcntl locks
- which version of the kernel client are you using?
Can you try the latest for-linus branch of ceph-client.git? T... - 10:59 PM Feature #1174 (Resolved): crbdnamer man page
- 10:52 PM Linux kernel client Bug #1148 (Resolved): o_direct crash in msgr
- 10:37 PM Bug #1124 (Resolved): Depend on new enough Boost
- the version dependency in the .spec file is nontrivial (something about epochs?) and i'm too lazy to figure it out pr...
- 10:30 PM Revision bc8e1209 (ceph): rgw: log pool id, and store/retrieve pool id info
- 10:30 PM Revision 34098a39 (ceph): librados: expose pool id
- Conflicts:
src/librados.cc - 10:21 PM CephFS Bug #1137 (Can't reproduce): MDS Crash
- If this turns up again, let us know! I suspect it may be related to the rename journaling changes; I'll be testing f...
- 10:19 PM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Andre, any chance you can pass along more information about the hang you were seeing? Log, or the specific workload?
- 09:17 PM Revision 1e96f71f (ceph): qa: fix test_shorT_dio_read
- No args, hardcode filename.
Signed-off-by: Sage Weil <sage@newdream.net> - 08:57 PM CephFS Bug #1084: blogbench won't finish: waiting for Fr cap forever
- It is working great. I haven't see this problem after applying the patches.
- 09:40 AM CephFS Bug #1084 (Resolved): blogbench won't finish: waiting for Fr cap forever
- I can't reproduce this on my mds_issue_caps, so I've merged it into master, commit:0f8fbd562cc58351541ac4f60a1dc98bdd...
- 08:36 PM Revision f9e9490d (ceph): PG: clear scrub_received_maps in scrub_clear_state
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 08:31 PM Revision 9d1e8971 (ceph): qa: fix loopall, runallonce scripts
- Run anything executable.
Signed-off-by: Sage Weil <sage@newdream.net> - 08:29 PM Revision d5bba79e (ceph): qa: move compiled workunits into organized directory
- No need for a separate qa/src/ dir; just put them where we want them. We
need to ignore non-executable files when do... - 08:22 PM Revision 953bb605 (ceph): qa: move compiled items into qa/src/
- 08:22 PM Revision f2aef3e1 (ceph): qa: organize workunits a bit
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:22 PM Revision adc98c0e (ceph): qa: expand test_sync_io
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:22 PM Revision 8147076d (ceph): qa: add direct and sync io workunits
- Just run the test programs in ../src
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:22 PM Revision 7087b2da (ceph): qa: include $basedir/src/ (qa/src/) in path
- This lets the workunits find stuff build in qa/src/, in case you are
running using these (old) scripts.
Signed-off-b... - 08:22 PM Revision 8a12a01c (ceph): qa: add henry's direct io tests
- Not sure what the second one is testing, actually, but including it anyway.
Signed-off-by: Sage Weil <sage.weil@drea... - 08:22 PM Revision d851deea (ceph): qa: add short directio read test
- This tests for the bug fixed by linux commit:c3cd628.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:02 PM Revision f3617e51 (ceph): osd: fix head_exists fix
- Needs to default to false, or else deletions don't work.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:05 PM Revision 056a0c5a (ceph): osd: don't update snapset head_exists unless full transaction succeeds
- Make the update at the end of make_writeable, along with the other snapset
updates.
Signed-off-by: Sage Weil <sage@n... - 07:04 PM Revision 11ad7658 (ceph): osd: fix make_writeable
- Look at the ObjectState to determine if the head exists, not the snapset
(which doesn't reflect the prior object stat... - 05:08 PM rgw Feature #1180 (Resolved): radosgw_admin log show dumps info in alternate formats
- Done, commit:e304fd0970492e38fd95ec71d00a221b6ec460c9. Can dump info in xml, json.
- 10:05 AM rgw Feature #1180 (Resolved): radosgw_admin log show dumps info in alternate formats
- In order to be able to add additional information without breaking compatibility.
- 05:07 PM rgw Feature #1179 (Resolved): rgw log should be able to show bucket owner
- That turned to be problematic, as we don't look at the bucket owner on all operations and we don't want to do that as...
- 09:45 AM rgw Feature #1179 (Resolved): rgw log should be able to show bucket owner
- Currently the log shows the object owner on the relevant operations.
- 05:04 PM Revision 35663dee (ceph): monc: protect msg source check with monc_lock
- Push the locking up into ms_dispatch, so that we protect the source check
and don't race with session reopens. Also ... - 04:43 PM Revision 0f8fbd56 (ceph): Merge branch 'mds_issue_caps'
- 04:20 PM Feature #1147 (Resolved): qa: tests for O_DIRECT, sync
- 02:09 PM Bug #1185 (Can't reproduce): rados: export caught in loop on 'buck' bucket (1.5M objects)
- dumped an object list, watched strace, and periodically checked the current file/object name against the list, and it...
- 01:13 PM Bug #1184: osd: snapset, pg info stats needs to be projected like obs
- also ObjectState& obs -> new_obs, so it's clear in the code this is the new, projected change.
each of these conve... - 01:03 PM Bug #1184 (Resolved): osd: snapset, pg info stats needs to be projected like obs
- We need a new_snapset to go along with new_obs in OpContext. And all the transaction code needs to be updated to mat...
- 12:00 PM CephFS Bug #1183 (Resolved): mds: snaptest-snap-rm-cmp.sh failing
- osd regression, fixed by commit:056a0c5a008fef0924b17c7e24c85cf73b3b390e
- 11:03 AM CephFS Bug #1183 (Resolved): mds: snaptest-snap-rm-cmp.sh failing
- this used to work :(
- 10:32 AM Tasks #1182 (Resolved): qa: lock machines
- Teuthology (or a layer above it) should lock machines while tests are executed.
- 10:13 AM Bug #1138 (Resolved): need to package rados.py in the debian .deb
- 10:07 AM CephFS Bug #1181 (Can't reproduce): mds: old_inodes crash
- ...
- 10:04 AM Bug #1151 (Resolved): OSD: CephxClientHandler::handle_response
- Fixed locking in mon_client, commit:35663dee2f7128df642594d0ebdf16640fe1f29a
06/12/2011
- 05:10 PM CephFS Bug #1178 (Resolved): cfuse doesn't shut down -- Fw cap doesn't flush on one inode
- Logs in kai:~gregf/logs/fsstress/hung_client2. This is on my mds_rename branch with 4 MDSes in fsstress. I suspect it...
- 04:26 PM CephFS Bug #1158: Unfinished freeze hangs fsstress
- I managed to reproduce this on my mds_rename branch.
06/11/2011
- 12:37 AM Revision 83a55f4d (ceph): Merge remote branch 'origin/stable'
- Conflicts:
src/dumpjournal.cc - 12:35 AM Revision 78012fde (ceph): Merge remote branch 'origin/deb-pybind'
- 12:10 AM Revision aa5c2244 (ceph): debian: Move python-ceph into section python.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 12:10 AM Revision efb82828 (ceph): debian: Strip and do -dbg magic for librgw too.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 12:09 AM Revision 061460b7 (ceph): debian: Fix copy-pasto in librgw1 control paragraph.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
06/10/2011
- 11:58 PM Revision 6fc8819b (ceph): debian: Depend on python-ceph in obsync.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:58 PM Revision 822e7770 (ceph): debian: Depend on librgw1 in python-ceph.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:43 PM Revision 762c87e0 (ceph): debian: Package librgw.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:15 PM Revision ab9dda87 (ceph): pybind: Open shared libs by their major version.
- The *.so files are only in the -dev packages, and normal
operation should not require those. The major version
number... - 11:15 PM Revision d444d5b8 (ceph): debian: Make git ignore obsync, gceph* package dirs.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:15 PM Revision ddf0263a (ceph): debian: Prevent git ignore from recursing on *.substvars etc.
- Should never have triggered but this is faster simpler
and correcter.
Signed-off-by: Tommi Virtanen <tommi.virtanen@... - 11:15 PM Revision 0a0493de (ceph): debian: Make git ignore all of the *.debhelper files.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:15 PM Revision 5cf62c1c (ceph): debian: Dev packages no longer have soname in their names.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 11:15 PM Revision 0da30c35 (ceph): debian: Properly package the python bindings.
- Build-depend on python-support. Add binary package
python-ceph, making it contain all the ceph python
packages, regar... - 10:02 PM Revision edc6659b (ceph): rgw: list bucket display the correct objects owners
- 09:55 PM Revision 3e284a72 (ceph): boto_tool: add put_bucket_acl
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:42 PM Revision 5cbe10c3 (ceph): added testsnaps
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 08:24 PM Revision 7fa8b973 (ceph): rgw: don't allow handing out object ownership
- 07:26 PM Revision 52e9e5ec (ceph): heap_profiler: if log_dir is empty, don't try and log to root dir!
- If log_dir was empty then the prefix would be of the form /mon.a
That isn't good, so fill it in so it writes to the c... - 06:55 PM Revision 98a3b546 (ceph): librados: get reference to the io context for the pending async ops
- 06:55 PM Revision b2826035 (ceph): librados: remove useless reference holding
- 06:31 PM Revision f41773be (ceph): coverage.sh: use .lcov instead of .info to avoid confusion
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 06:28 PM Revision 08d0d0ae (ceph): Merge remote branch 'origin/wip_coverage'
- Conflicts:
src/mon/Monitor.cc
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com> - 06:24 PM Revision d09ae0b2 (ceph): With coverage, use SIGTERM to make the daemons exit(0).
- 06:17 PM Revision e52bb0cf (ceph): Put all coverage data in the archive dir.
- 06:12 PM Revision 41e19f5e (ceph): Initialize BASE before using it.
- 05:23 PM Revision a6afb051 (ceph): rgw: remove required indexes when modifying user info
- Conflicts:
src/rgw/rgw_admin.cc - 05:21 PM Subtask #1177 (Resolved): qa: post-process coverage data
- Combine from multiple machines per test, and from multiple tests per commit.
- 05:17 PM Tasks #1176 (Closed): qa: html test results
- Generate a page with a table of commit id, tests passed/run, and links to coverage data (aggregate and per-test).
- 05:14 PM Tasks #1175 (Resolved): qa: multi-test runner
- This will be a cron job that picks the latest valid master at e.g. midnight and runs a suite of tests via teuthology.
- 05:13 PM Feature #1174 (Resolved): crbdnamer man page
- this needs to go in the librbd1 package
- 05:12 PM Revision 9a705c74 (ceph): mon: Follow the Message rules when forwarding tells
- My initial commit was naughty, fix that up here by
creating a new message for each mon and just copying
the command.
... - 05:11 PM Feature #1173 (Rejected): gceph: man page
- 05:10 PM Feature #1172 (Resolved): obsync: man page
- 05:07 PM Revision 842f3ac6 (ceph): rgw: remove required indexes when modifying user info
- 04:25 PM Revision c2de9e6a (ceph): mds: clear correct state bit in do_file_recover
- The RECOVERING bit is set, not NEEDSRECOVER. This way we match
_recovered().
Signed-off-by: Sage Weil <sage@newdrea... - 03:07 PM rgw Bug #1171 (Resolved): rgw list bucket displays wrong object owner
- Fixed, commit:6f02b4220e9ef925787772132c01328ccf4d07ae.
- 02:51 PM rgw Bug #1171 (Resolved): rgw list bucket displays wrong object owner
- It just dumps the id of the current user.
- 02:30 PM CephFS Bug #1169: 2-MDS fsstress: Active fails adjust_subtree_auth
- Wait, this wasn't on mds_rename. It was on my wip-sanity branch.
- 02:27 PM CephFS Bug #1169 (Closed): 2-MDS fsstress: Active fails adjust_subtree_auth
- This is on mds_rename branch...
- 02:30 PM CephFS Bug #1170: 2-MDS fsstress: SR fails subtree assert
- Wait, this wasn't on mds_rename. It was on my wip-sanity branch.
- 02:28 PM CephFS Bug #1170 (Closed): 2-MDS fsstress: SR fails subtree assert
- This is on mds_rename branch....
- 01:33 PM rgw Bug #894 (Resolved): openstack pseudo folders
- I'm closing this one. Pseudo folders were implemented, the only thing that was missing is testing it using the older ...
- 01:31 PM rgw Bug #1115 (Resolved): rgw allows users to "give away" s3 objects
- Fixed, commit:859462298170520d53a167c1da214e378ae2e78a.
- 01:31 PM Bug #1152: Mon getting killed by OOM killer
- Attached are the heap dumps from the mon process.
The log is a big, 1.4GB, so I didn't upload it. It can be found ... - 11:42 AM Bug #1152: Mon getting killed by OOM killer
- It'll dump a summary in the clog and try to dump the analysis data into a file named something like osd.1.0001.heap. ...
- 10:22 AM Bug #1152: Mon getting killed by OOM killer
- I've started my mon with memory profiling enabled, but it will take some hours before it starts eating memory.
At ... - 01:11 PM CephFS Bug #1168 (Closed): fsstress 2-mds Standby-replay adjust_subtree asserts
- ...
- 11:19 AM CephFS Bug #1167 (Resolved): Fsstress: bad rstat assert
- Got a bad rstat assert in fsstress. This is with 4 MDSes. Logs in kai:~gregf/logs/fsstress/bad_rstats...
- 11:11 AM CephFS Bug #1166: mds: standby-replay bound asserts
- Logs in kai:~gregf/logs/fsstress/standby_bound_asserts
- 10:55 AM CephFS Bug #1166 (Resolved): mds: standby-replay bound asserts
- Got two of them on a 4-mds run of ffsstress:...
- 11:10 AM CephFS Bug #1114: NFS export extreme slowdown
- I've verified this problem continues with 2.6.39 kernel with today's (5/26/11) build from ceph-client git repo (maste...
- 10:23 AM rgw Bug #1162 (Resolved): rgw_admin user modify --email
- Fixed with commit:3e8da6c71de01330867ad16b8e19b903f94bb906.
- 10:21 AM Bug #1151: OSD: CephxClientHandler::handle_response
- The full OSD log is a bit large to attach, so I've left it on my server at the office.
You can access it by SSH'in... - 10:11 AM Tasks #1165 (Closed): snap collection clone removal
- On recovery, new pg members won't have seen the transactions removing old clones from snap collections. When they pr...
- 09:54 AM CephFS Bug #1110 (In Progress): mds: ls -l hangs on concurrent writer
- Andre Noll wrote:
> Andre Noll wrote:
> > Sage Weil wrote:
> > > I pushed commit:5922de29e7414be3aaa6eb04be8bb0ae5... - 03:50 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Andre Noll wrote:
> Sage Weil wrote:
> > I pushed commit:5922de29e7414be3aaa6eb04be8bb0ae5c943ccf, which should fix... - 09:26 AM Feature #1154 (Resolved): Make tcmalloc profiler commands work with monitor
- Done, and properly tested this time.
- 05:15 AM Revision 35beabf6 (ceph): rgw: remove clutter
- 05:05 AM Revision c863f522 (ceph): rgw: fix user anonymous permissions
- 01:32 AM Revision dadf2a3f (ceph): Thread: remove globals. Thread create must succeed
- Remove the references to global variables from Thread.h. They are really
unecessary. In every case, the printout is f... - 01:04 AM Revision 4e0ddae4 (ceph): LogClient: pass in CephContext explicitly
- Again, the Messenger passed to LogClient may be NULL, so pass in the
CephContext explicitly.
Signed-off-by: Colin Mc... - 12:56 AM Revision 03aef0b6 (ceph): Monitor: pass in CephContext explicitly
- Sometimes we create a Monitor without a Messenger. So we can't pull the
CephContext out of the Messenger, because it ... - 12:43 AM Revision 2b5e0111 (ceph): mds, mon, osd: with coverage enabled, exit(0) when SIGTERM is received
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 12:43 AM Revision befe4226 (ceph): Revert "mon: add all_exit and exit commands"
- This reverts commit 69092735c4b25c88ded58b17b1c198fafcf5931b.
- 12:42 AM Revision cf129ff7 (ceph): Revert "mds: allow mds to 'exit immediately'"
- This reverts commit c01961ff4b712a12e25e437adceae264273d7428.
- 12:42 AM Revision dd9ea9c6 (ceph): Revert "osd: add command to exit cleanly"
- This reverts commit 1cc78b6a864d36cbe1eda45890d7f361adb8201d.
- 12:22 AM Revision 3a582f99 (ceph): mon: add new "tell" infrastructure.
- This lets you issue commands
like "ceph mon tell 0 heap start_profiler". Hurray!
Signed-off-by: Greg Farnum <gregory... - 12:22 AM Revision b6e5c080 (ceph): mds: fix up MDCache::path_is_mine to remove a totally bogus assert.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 12:17 AM Revision 43f37b8d (ceph): Add coverage support.
- 12:17 AM Revision 7c5eed2f (ceph): Check daemon exit codes, even when using gcov.
- 12:17 AM Revision c45488c2 (ceph): Show status message before gcov-style shutdown of daemons.
- 12:17 AM Revision 18b11739 (ceph): Add TODO notes.
- 12:17 AM Revision 2533edde (ceph): Save ceph sha1 in coverage dir.
- This is needed to postprocess the coverage data.
- 12:17 AM Revision ae77af08 (ceph): Simplify ceph sha1 saving. No need to round-trip, store N copies.
- 12:17 AM Revision 798f3cc6 (ceph): Can now run specific ceph version based on sha1.
- 12:16 AM Revision f162f07e (ceph): Support ceph tarball "favors", that is different gitbuilders.
- Use that for picking the gcov gitbuilder, when told to record code
coverage. Actual coverage support comes in the nex... - 12:12 AM Revision f3757f76 (ceph): Let ceph task config specify branch or tag to run.
- 12:10 AM Revision 4903d06a (ceph): Archive everything written to the "archive" directory.
- This avoids listing "log", "profiling-logger", etc individually.
Adding new types of archived files should just be a ... - 12:03 AM Revision 4c798752 (ceph): rgw: can specify alternative remote addr http header param
06/09/2011
- 11:40 PM Revision cf5e70d4 (ceph): rgw: skeleton for list multipart uploads
- 11:21 PM Revision 2c4367f9 (ceph): CephContext: initialize module_type in ctor
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:14 PM Revision 5b304bdd (ceph): mds: make pneed_issue required for eval(SimpleLock*)
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision 87ba0435 (ceph): mds: fix xlock_finish issue flag check
- We were sometimes setting do_issue but not *pneed_issue. Simplify by
setting do_issue internally to the function and... - 11:14 PM Revision 7c0ae247 (ceph): mds: fix eval_gather callers passing pissue_gather as wrong arg
- Whoops!
Signed-off-by: Sage Weil <sage@newdream.net> - 11:14 PM Revision 96f01b03 (ceph): mds: fix xlock_finish do_issue checks
- Should default to false, and only get set to true if there are caps for
this lock. Among other things this means we ... - 11:14 PM Revision 20d11b0b (ceph): mds: be slightly more careful about checking if lock has caps
- issued_caps_need_gather==true implies it, but we can skip the work if the
lock has no caps.
Signed-off-by: Sage Weil... - 11:14 PM Revision 2a45bcad (ceph): mds: issue caps from drop_locks
- In drop_locks, build a set of inodes we need to issue caps on. Then do it
all at once. This does two things:
- it... - 11:14 PM Revision 88b22463 (ceph): mds: make issue_caps from file_update_finish smarter
- We do one funky thing in file_update_finish that only issues caps on a
single cap when max_size changes. This is mor... - 11:14 PM Revision 3e52a33d (ceph): mds: set or issue caps on lock state changes
- Set pneed_issue (or issue ourselves) whenever we jump directly to the
target lock state. Make sure we only do it if ... - 11:14 PM Revision f62f8deb (ceph): mds: efficiently drop_rdlocks and drop_xlocks too
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision cc398ccf (ceph): mds: efficiently issue_caps for scatterlocks unlock during refragment
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision ffeb59d5 (ceph): mds: explicitly issue_caps for slave xlock release
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision a340421a (ceph): mds: explicitly issue_caps for rdlock_finish_set
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision bc69d4e0 (ceph): mds: efficiently issue_caps for OOO or extra locks in acquire_locks
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision b06fa1ce (ceph): mds: make pneed_issue required for *lock_finish()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision bdd94ef1 (ceph): mds: explicitly issue_caps after file_recover
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:14 PM Revision c7863911 (ceph): mds: pass pissue_caps through *lock_finish()
- This allows *lock_finish() callers to handle the issue_caps themselves.
None of them do yet (this arg is still option... - 11:09 PM Revision 6d4b93dd (ceph): CephContext: remove silly debug code
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:08 PM Revision e6349ec2 (ceph): Dispatcher should take a CephContext argument
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:33 PM Revision c1ea3a26 (ceph): mds: replay: save the right dir!
- We were using the wrong one and breaking replay.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 10:33 PM Revision 58250fdd (ceph): mdcache: correct error message.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:33 PM Revision 379dc6fa (ceph): MDiscoverReply: encode/decode wanted_ino.
- This wasn't encoded previously so it was always zero on the
receiver's end, which led to nasty hangs. :(
Signed-off-... - 10:33 PM Revision 63073ab8 (ceph): mds: xlock_finish should only do_issue in certain cases.
- We accidentally (we think) initialized this variable to true when
we want it to be false: we should only do_issue if ... - 10:33 PM Revision 4ef41c5a (ceph): mds: fix tcmalloc check.
- We can run commands to do things like start the profiler even if
it's not already running!
Signed-off-by: Greg Farnu... - 10:32 PM Feature #1154 (In Progress): Make tcmalloc profiler commands work with monitor
- I missed some when setting up the "tell *" option, need to look at it a bit more. Initial fixes failed and something ...
- 04:37 PM Feature #1154 (Resolved): Make tcmalloc profiler commands work with monitor
- Pushed to master in commit:3a582f994998fa925a6f2b619326bfc24616306b
- 11:12 AM Feature #1154 (Resolved): Make tcmalloc profiler commands work with monitor
- This will require setting up a tell interface or similar for the monitors. Whee!
- 10:15 PM rgw Bug #1163 (Resolved): anonymous access to objects broken
- Fixed, commit:c863f52245372820ab04814a4957963fb3467032.
- 05:10 PM rgw Bug #1163 (Resolved): anonymous access to objects broken
- After setting bucket and/or object as public-read, can't access them without being authenticated first.
- 09:34 PM Revision f69bc301 (ceph): debian: move gceph into a separate package
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:03 PM Revision 0b40f480 (ceph): tools/common: fix C_ObserverRefresh constructor
- Fixes #1159
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 08:57 PM Revision 3308aa67 (ceph): rgw: implement abort multipart upload
- (not tested yet)
- 08:25 PM Revision 802e9e5a (ceph): rgw: implement namespace, use it in complete multipart
- 06:42 PM Subtask #1164 (Resolved): initialize g_ceph_context in common_preinit
- It will be a major milestone when we can initialize g_ceph_context in common_preinit. Basically, in order to do this,...
- 06:38 PM Subtask #844 (Resolved): g_conf should become a pointer
- implemented
- 06:31 PM Revision 01df7c66 (ceph): ReplicatedPG: make_writeable, use correct size for clone_size entry
- Previously, we used obs.oi.size for the clone size, but obs refers to
new_obs. The clone's size should be the old si... - 06:22 PM Revision 1376a5ab (ceph): gtest: add -lpthread to AM_LDFLAGS
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:18 PM Revision e8abcf0e (ceph): readme: need libevent-dev
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:10 PM Revision b2becefa (ceph): man: update cosd man page to include info on flush-journal option.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 05:36 PM Revision d224b3e2 (ceph): Makefile.am: Add -lpthread to UNITTEST_LDADD
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:09 PM rgw Feature #1161 (Resolved): configurable client ip header variable
- 05:09 PM rgw Feature #1161: configurable client ip header variable
- Fixed, commit:4c798752c6d2805e74464588cae1d67da69e8a4a.
The RGW_REMOTE_ADDR_PARAM can be set now in the configurat... - 04:06 PM rgw Feature #1161 (Resolved): configurable client ip header variable
- When going through a load balancer that rewrites the header, the client ip address (that we want to write into the lo...
- 05:02 PM CephFS Bug #1084: blogbench won't finish: waiting for Fr cap forever
- The mds_issue_caps branch has my current work on cleaning the issue_caps stuff up. I'm pretty happy with what the MD...
- 04:38 PM Bug #1152: Mon getting killed by OOM killer
- Pushed it, let me know if it doesn't work for you.
- 11:14 AM Bug #1152: Mon getting killed by OOM killer
- I've created #1154 to make those commands work and hope to get it done today, but if you're feeling dedicated you sho...
- 10:29 AM Bug #1152 (Closed): Mon getting killed by OOM killer
- I've been seeing this for the last few weeks, my single mon keeps eating more and more memory until it reaches 4G Res...
- 04:30 PM rgw Bug #1162 (Resolved): rgw_admin user modify --email
- We need to make sure that when modifying email address, the old email object under .users.email is removed.
- 04:11 PM Revision 5922de29 (ceph): mds: only target XSYN state if we have a valid target loner
- If we are in EXCL and really shouldn't be, no need to go to XSYN; go to
SYNC instead. This keeps the assert
asser... - 04:09 PM CephFS Bug #1158: Unfinished freeze hangs fsstress
- Although actually based on how long fsstress is taking on this disk maybe nothing was blocked and it was just going s...
- 01:21 PM CephFS Bug #1158 (Can't reproduce): Unfinished freeze hangs fsstress
- I've got a freeze that doesn't finish blocking fsstress. Logs in kai:~gregf/logs/fsstress/freeze_not_finishing.
Ha... - 04:02 PM Revision 7a5201e4 (ceph): mkcephfs: fix ceph.conf reference
- It's $conf (which is $dir/conf).
Reported-by: biyan chen <riby.chen@gmail.com>
Signed-off-by: Sage Weil <sage@newdre... - 02:52 PM Subtask #1160 (Resolved): introduce CephContext to some structures in common/
- Some data structures in common like Thread and Context need to use CephContext to do their logging and configuration....
- 02:46 PM Cleanup #1149 (Resolved): debian: separate gceph into a separate package
- 02:24 PM Bug #1145 (Won't Fix): monitor assert fails due to ENOSPC
- Hi Sam-
This is done deliberately so that the system doesn't continue thinking it wrote something when it didn't. ... - 02:09 PM Bug #1159 (Resolved): ceph -w segfault
- fixed by commit:0b40f48073649c74f8f29148cdc84b997a1666e8
- 01:49 PM Bug #1159: ceph -w segfault
- This is something with CephToolCtx, giving to Colin.
- 01:28 PM Bug #1159 (Resolved): ceph -w segfault
- On latest master with a few extra MDS patches, I got a segfault. Core and binary in kai:~gregf/logs/ceph_core
- 12:41 PM CephFS Bug #1153 (Resolved): Hang on 3-MDS fsstress
- Okay, this one was hanging because the wanted_ino member of MDiscoverReply wasn't encoded and decoded. Duh.
- 11:10 AM CephFS Bug #1153 (Resolved): Hang on 3-MDS fsstress
- While testing our rename changes I managed to get a hang in a 3-MDS setup of fsstress. I'm looking through the logs n...
- 11:27 AM Bug #1144: snap_trimmer
- wip_snap_trimmer has the current incarnation of the changes
- 11:25 AM Tasks #1157 (Closed): create teuthology tests for updated snap_trimmer
- 11:24 AM Tasks #1156 (Closed): assert failure in _scrub, assert(p->second.size == snapset.clone_size[*curc...
- In make_write, we set the clone size to the new object size obs.oi.size rather than the cloned size ctx->obc->obs.oi....
- 11:15 AM Tasks #1155 (Resolved): Replica does not remove snap collections
- Caused by replica not updating snap_collections.
Have fix in wip_snap_trimmer, won't push until the rest of the br... - 10:34 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- Sage Weil wrote:
> I pushed commit:5922de29e7414be3aaa6eb04be8bb0ae5c943ccf, which should fix that crash. Can you r... - 09:08 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- I pushed commit:5922de29e7414be3aaa6eb04be8bb0ae5c943ccf, which should fix that crash. Can you retest?
Still work... - 05:01 AM CephFS Bug #1110: mds: ls -l hangs on concurrent writer
- First of all, sorry for the delay, I haven been distracted by other work.
Sage Weil wrote:
> Okay, there was an i... - 09:56 AM Bug #1151: OSD: CephxClientHandler::handle_response
- Also, tell us more about this cmon memory thing! :)
- 09:01 AM Bug #1151: OSD: CephxClientHandler::handle_response
- Can you attach the full osd log? It looks like there are two threads doing the auth handshake.
- 03:06 AM Bug #1151 (Resolved): OSD: CephxClientHandler::handle_response
- I just saw a bunch of OSD's going down with:...
06/08/2011
- 11:35 PM Revision 6f6ac75d (ceph): Makefile: remove ancient comment
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:23 PM Revision 9202a4b0 (ceph): ceph tools: de-globalize CephToolContext
- De-globalize CephToolContext. It's important to do this now because the
constructor for CephToolContext references th... - 10:57 PM Revision 92ee0de8 (ceph): mds: drop workaround for embedded null decoding
- We removed the embedded null check recently, 9b37f4f, so this workaround
isn't needed.
Signed-off-by: Sage Weil <sag... - 09:55 PM Revision 6069fa6a (ceph): mon: weaken pool creation caps check
- Allow a pool creation if we specify an auid but are allowed to create
buckets as anybody.
Signed-off-by: Sage Weil <... - 09:54 PM Revision 80c039cd (ceph): rgw_admin: user info dumps auid
- 09:53 PM Revision 825cec3c (ceph): moncaps: whitespace
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:50 PM Revision 4e58308d (ceph): mds: rename: remove illicit assert.
- We actually do want witnesses who aren't auth for anything
to do journaling in some cases, so kill the assert.
That a... - 09:13 PM Revision 40262471 (ceph): rgw_admin: can reset auid
- 09:03 PM CephFS Bug #1084: blogbench won't finish: waiting for Fr cap forever
- I just tested it with the latest ceph stable branch (0.29+) and kclient master branch (backported for 2.6.32 by me). ...
- 09:18 AM CephFS Bug #1084: blogbench won't finish: waiting for Fr cap forever
- Looked at this some more yesterday. Turns out the problem seems to be that the client closes the inode, and all caps ...
- 09:00 PM Revision 8cd949f6 (ceph): Merge remote branch 'origin/mds_rename'
- 08:29 PM Revision 5b43419c (ceph): mds: try_trim_non_auth_subtree if we rename a dir away from a non-auth ...
- It's possible we have non-auth metadata only because we have a subtree
nested beneath. If we rename a directory out o... - 08:27 PM Revision 072e80ed (ceph): mds: use CDIR_AUTH_UNDEF where possible
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:20 PM Revision e19d639f (ceph): Gevent 0.14 switches to libev, avoid it for now.
- This was in requirements.txt earlier, but using this library
from elsewhere does not respect that, only setup.py. - 08:18 PM Revision 811dcae7 (ceph): mds: remove unlinked metadata from cache on replay
- If we replay a metablob that unlinks something, throw it out immediately.
Recursively. This comes up when:
- we ren... - 08:14 PM Revision 23302815 (ceph): rgw: set locator on all required rados calls
- 08:10 PM Revision 5cf6d0e6 (ceph): rgw: put locator key in rgw_obj
- 07:32 PM Revision 8bd984d9 (ceph): rgw: refactor rgw backend interface
- 07:30 PM Revision f3688b71 (ceph): Fix build errors in unit tests
- libcommon uses symbols from the crypto libraries, so they must appear on
the link line whenever libcommon appears. L... - 07:05 PM Revision 24701308 (ceph): monitorstore: return error codes for consistency
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:27 PM Revision cfbcfd7c (ceph): Fix typo in usage output for --num-osds
- This patch fixes the typo in the usage description to match the actual argument string.
Signed-off-by: Sam Lang <sam... - 06:25 PM Revision 5d56dfb7 (ceph): Fix segfault caused by invalid argument string.
- This patchset includes minor fixes to the crushtool utility. If an invalid bucket type is speicifed on the command l...
- 06:22 PM Revision a1a71471 (ceph): qa: test witness subtree adjustment
- This exercises the mds at least; we don't have a way to verify mds
behavior currently.
Signed-off-by: Sage Weil <sag... - 05:34 PM Bug #1150 (Duplicate): kclient: ERESTARTSYS from flock/fcntl locks
- After upgrade from 0.28.2 to 0.29 amanda backup no longer working if amanda files placed on ceph fs and ceph fs mount...
- 02:12 PM CephFS Bug #1146 (Resolved): mds: correctly trim cache after rename witness replay
- 11:25 AM Cleanup #1149 (Resolved): debian: separate gceph into a separate package
- 07:25 AM Bug #1032: osd: Marked down and become zombies after killing
- I'm just upgrading my cluster, I see a Zombie from May 26th, running v0.28.1 and 2.6.38.2.
I'll see if I can repro... - 06:48 AM Linux kernel client Bug #1148: o_direct crash in msgr
- Hi Sage,
I refactored stripe_read a bit and fixed the calculation of page count in ceph_osdc_new_request.
See:
... - 05:09 AM Revision e2c808ae (ceph): qa: test_sync_io
- Currently only tests the read path for O_DIRECT and sync (if the ioctl is
in place). Also currently crashes the kcli... - 03:49 AM Revision 21795aaf (ceph): qa: add dir rename tests
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 03:48 AM Revision 00ec86a2 (ceph): mds: open renamed import child frags during journal replay
- Open up any child frags of the imported renamed inode that are noted in
the journal event. (Note we blindly open up ... - 03:46 AM Revision 44d19304 (ceph): mds: journal open srci frags on srci import (master)
- If we are importing the renamed inode, and it is a directory, journal a
list of all open dirfrags (currently, this is... - 03:43 AM Revision a7c083ec (ceph): mds: journal renames on witnesses if we have nested subtrees
- If a rename witness has any subtrees that are nested beneath the renamed
directory, we need to journal the rename eve... - 03:40 AM Revision 5c870c60 (ceph): mds: CDir::contains() should use projected hierarchy
- The only current callers are in resolve stage, when nothing is projected.
Signed-off-by: Sage Weil <sage.weil@dreamh...
06/07/2011
- 11:14 PM Revision 1b0fdcae (ceph): re-fix ignore return value change
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:45 PM Linux kernel client Bug #1148 (Resolved): o_direct crash in msgr
- test_sync_io makes us crash on...
- 10:44 PM Feature #1147: qa: tests for O_DIRECT, sync
- beginnings of test with commit:e2c808aea97ced6e9a55b143116b66d344f72c0b
- 10:04 PM Feature #1147 (Resolved): qa: tests for O_DIRECT, sync
- Test correct results with varying buffer alignment, file offset, and either sync or O_DIRECT io.
write with buffer... - 10:13 PM Revision 3a2ce92b (ceph): mds: fix check for no-op rename of two links to the same inode
- touch a
ln a b
mv a b
is a no-op, according to posix. Move the check earlier.
Signed-off-by: Sage Weil <sage.weil... - 09:51 PM Revision fcc73e0d (ceph): Class loading no longer needs a temp dir.
- 09:48 PM Revision 047669e9 (ceph): Cleanup is done, adjust TODO.
- 09:47 PM Revision e6ae6bdd (ceph): Archive logs if given --archive=PATH. Clean up after a test run.
- 09:41 PM Revision 37cbd7ab (ceph): Support PIPE for stdout and stderr, too.
- 09:41 PM Revision f225db7b (ceph): rgw: remove parts after multi-part upload
- 09:27 PM Revision a7012f28 (ceph): cfuse: really ignore write() return value
- Hopefully this will make gcc stop complaining. The void cast no longer
seems to silence unchecked return code errors,... - 09:13 PM Revision 37fd3b58 (ceph): rgw: calc etag for multipart upload
- 09:08 PM CephFS Bug #1128 (Resolved): clustered mds: failed verify_subtree_bounds
- closing this, opening specific bugs for remaining issues
- 09:07 PM CephFS Bug #1146 (Resolved): mds: correctly trim cache after rename witness replay
- 08:57 PM Revision dfa4aeb8 (ceph): Merge remote branch 'origin/wip_gcov'
- 08:34 PM Revision 2969161b (ceph): qa: add rename dir checks
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:29 PM Revision 1f2740a0 (ceph): Makefile.am: install gcno files
- This is needed so that coverage data can be generated from the
gitbuilder tarballs.
Signed-off-by: Josh Durgin <josh... - 08:29 PM Revision 0eda7195 (ceph): Makefile.am: add check-coverage to get coverage of 'make check'
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 08:29 PM Revision b32f00eb (ceph): coverage: add ceph-coverage to run something with the right GCOV_PREFIX...
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 08:28 PM Revision e4f05415 (ceph): Makefile.am: clean gcno and gcda files in "make clean"
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 08:27 PM Revision 2cb0c110 (ceph): qa: rename tests
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:24 PM Revision 3e0352d3 (ceph): mds: rename: add missing pop_and_dirty_projected_inode calls
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:24 PM Revision 569baab6 (ceph): mds: rename: use temp *in in _rename_apply
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:24 PM Revision bdfcc0df (ceph): mds: rename: move unlink out of every possible branch
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 07:58 PM Revision ce10d4f5 (ceph): Add missing trailing commas.
- Lists of strings where last item has no comma are just
traps waiting to hurt you when you add a new item at
the end. - 07:05 PM Revision 12bb3081 (ceph): test/bufferlist: add copy_all test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:04 PM Revision 1cc78b6a (ceph): osd: add command to exit cleanly
- This is required for gcov to work on daemons since the coverage data
is written atexit, and the function that writes ... - 07:04 PM Revision 347dba4e (ceph): mon: ceph tell mds * is a valid command
- Previously this fell through and returned -EINVAL to the user.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com> - 07:04 PM Revision c01961ff (ceph): mds: allow mds to 'exit immediately'
- This is temporary until shutting down cleans up well.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com> - 07:04 PM Revision 69092735 (ceph): mon: add all_exit and exit commands
- all_exit makes each daemon exit(0), for gcov data collection.
exit causes cmon to do this.
Signed-off-by: Josh Durgi... - 07:04 PM Revision eb097649 (ceph): coverage: add helper script to get coverage for a local test
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 07:02 PM Revision 24195b9c (ceph): configure: add option for building with gcov coverage support
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 06:58 PM Revision 7b9fa3c3 (ceph): test/bufferlist.cc: retab for consistency
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:56 PM Revision d9d3c78e (ceph): cfuse.cc: fix warning
- We genuinely don't care about the return code we get back from this call
to write(2). So cast the result to void to t... - 06:40 PM Revision ea9c4fc2 (ceph): bufferlist::iterator: add copy_all, use in KeyRing
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:25 PM Revision 6b2bf234 (ceph): mds: rename: fix oldin pop
- - make the pop match position with the project in prepare
- don't pop on linkmerge, since we don't project in that ca... - 06:24 PM Revision bf7d7fa9 (ceph): mds: rename: some cleanup
- - add oldin temp
- make if conditions match between _prepare and _apply
no functional changes.
Signed-off-by: Sage ... - 06:23 PM Revision b794aeb3 (ceph): client: fix rename request target selection
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 06:11 PM Revision 084a1c74 (ceph): client: fix mds selection for unlink
- It was broken when a remote link target was on another mds.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 06:08 PM Revision a2288574 (ceph): client: make debug prints for choose_target_mds more helpful
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 06:07 PM Revision 5e679d3a (ceph): client: allow caps to be dropped on another inode for requests
- We need the ability to drop caps on another inode that isn't req->inode
or req->old_inode in the request struct.
Sig... - 05:59 PM Revision 0fb79737 (ceph): mds: fix unlink new snaprealm check
- The bool was flipped.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 05:50 PM Revision dbe41a30 (ceph): client: fix rename debug print
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:38 PM Revision 83e49cd8 (ceph): g_conf: change to pointer
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:23 PM Revision e61f5c5c (ceph): simple_spin: fix compiler warnings
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:17 PM Revision 776d3814 (ceph): init: set cct->module_type in common_preinit
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:04 PM Revision e36a4b26 (ceph): mds: rename: fix handling of target that is remote dentry
- If the target is a remote dentry, we need to consider that the destdn
and desti may have different auths.
Signed-off... - 04:58 PM Revision a5662293 (ceph): client: fix mds routing for link requests
- We want to send link requests to the auth for the new name, not the
target inode. We also want to drop FILE_SHARED c... - 04:49 PM Revision 7770e1c6 (ceph): mds: rename: only add target to stray if destdn is a primary
- Only add target into the stray dir if we are renaming over a primary
dentry. (Otherwise we aren't moving the target.... - 04:41 PM Revision df7f8954 (ceph): mds: fix/clean up xlock import/export
- - create xlock import/export helpers
- fix/simplify checks: we want to export/import only xlocks on the inode
that is... - 08:40 AM Bug #1145 (Won't Fix): monitor assert fails due to ENOSPC
- I put the monitor storage on a boot disk that was already 100% full, which is a user error on my part, but I'm not su...
- 08:36 AM CephFS Bug #1084: blogbench won't finish: waiting for Fr cap forever
- The real question here is why the earlier callers of issue_caps think that the client (a loner) can't have the Fr cap...
Also available in: Atom