Activity
From 12/04/2012 to 01/02/2013
01/02/2013
- 09:45 AM Bug #3700: mds: FAILED assert(!item_session_list.is_on_list())
- fixed by revert of bad fix, see commit:6711a4c4038dbdf843f9dfe42c7809c5c37ae534
- 09:37 AM Bug #3700 (Resolved): mds: FAILED assert(!item_session_list.is_on_list())
12/30/2012
- 06:08 PM Fix #3630: mds: broken closed connection cleanup
- ...
- 06:06 PM Fix #3630: mds: broken closed connection cleanup
- The con re-use looks like this:
- client connects
- mds ms_verify_authorizer creates a new session
- msgr see ex... - 06:04 PM Bug #3696 (Resolved): mds: FAILED assert(session_map.count(s->inst.name) == 0)
- see #3630..let's fix this properly.
12/29/2012
- 02:39 PM Bug #3700 (Resolved): mds: FAILED assert(!item_session_list.is_on_list())
- logs: ubuntu@teuthology:/a/teuthology-2012-12-29_03:00:03-regression-master-testing-gcov/30039...
- 02:32 PM Bug #3696: mds: FAILED assert(session_map.count(s->inst.name) == 0)
- ubuntu@teuthology:/a/teuthology-2012-12-29_03:00:03-regression-master-testing-gcov/30036
- 09:43 AM Bug #3696: mds: FAILED assert(session_map.count(s->inst.name) == 0)
- reverted the broken fix, reproducing the original problem again.
12/28/2012
- 09:11 PM Bug #3696: mds: FAILED assert(session_map.count(s->inst.name) == 0)
- 06:42 PM Bug #3696 (Resolved): mds: FAILED assert(session_map.count(s->inst.name) == 0)
- This occurred shortly after startup when trying to reproduce another bug on the master branch:...
- 06:21 PM Fix #3630: mds: broken closed connection cleanup
12/26/2012
- 09:59 AM Bug #3681 (Resolved): kclient fsx fails nightly
- ...
- 08:39 AM Feature #3679 (Closed): Any API to get metadata?
- Yep! See libcephfs. There is...
- 01:08 AM Feature #3679 (Closed): Any API to get metadata?
- hello,there.
I am wondering if there is any API to get the metadata of a file .
I have the ceph file system run by ... - 01:10 AM Tasks #3680 (Rejected): deduplication in ceph
- I am wondering how to do deduplication in ceph...the big problem is how to get the metadata of a file
and how to mod...
12/24/2012
- 02:58 PM Feature #1448 (In Progress): test hadoop on sepia
- 02:58 PM Cleanup #814 (Resolved): hadoop: refactor hadoop shim in terms of java libceph bindings
12/23/2012
- 09:12 PM Cleanup #3677 (Closed): libcephfs, mds: test creation/addition of data pools, create policy
- the create data pool argument is tested only with the default pools. once an lib is in place for the unit/functional...
- 09:06 PM Bug #3663 (Rejected): ceph kernel client is getting stuck on xstat* operations
- No worries. Let us know if you do come across behavior that looks like a bug!
- 08:59 PM Bug #3663: ceph kernel client is getting stuck on xstat* operations
- Hi Sage,
i am very sorry for taking your time with this issue, I feel like an idiot :(
The buggy client is runnin...
12/21/2012
- 02:39 PM Documentation #3672 (Resolved): doc: how to mount ceph-fuse from fstab
- There's a new mount helper in bobtail for this. It contains these comments:...
- 10:20 AM Bug #3666 (Resolved): Segfault running test_libcephfs
- ...
- 08:36 AM Bug #3655 (Can't reproduce): client: hang in fsstress
- I ran this test throughout the day yesterday and couldn't reproduce it, with message delays enabled. Marking as can'...
- 07:52 AM Bug #3663: ceph kernel client is getting stuck on xstat* operations
- Hi Roman-
The logging levels are right, but in both mds logs neither mds was ever active; both were in the up:stan...
12/20/2012
- 10:19 PM Bug #3663: ceph kernel client is getting stuck on xstat* operations
- Hello Sage,
added 4 logs:
screen output from console of the laggy client. it ends up on 'jroger@pr02:~/data$ cp... - 09:07 PM Bug #3663 (Need More Info): ceph kernel client is getting stuck on xstat* operations
- Hmm. It's actually just saying its the oldest client; it's not actually too old (yet). The looping connect attempts...
- 08:48 PM Bug #3663 (Rejected): ceph kernel client is getting stuck on xstat* operations
- there are 2 kernel clients happily working with ceph. as soon as I try mounting ceph from the third client, it's gett...
12/19/2012
- 11:19 PM Bug #3655 (Can't reproduce): client: hang in fsstress
- fsstress stuck in _read_sync()
#0 pthread_cond_wait@@GLIBC_2.3.2 ()
at ../nptl/sysdeps/unix/sysv/linux/x86_6... - 04:03 PM Bug #3637: client: not issuing caps for with clients doing shared writes
- Proposed fix in wip-3637. The client's max size request in MClientCaps gets dropped if the file lock is in a non-sta...
- 12:30 PM Bug #3625: client: EEXIST error on multiple clients to create
- Pushed fixes to wip-3625 (ceph and ceph-client repos) that implement #3 (mds sends back the created flag in reply to ...
- 12:29 PM Bug #3625: client: EEXIST error on multiple clients to create
- David and I have posted comments on github about the fix to allow multiple
clients opening the same file to get a va...
12/18/2012
- 08:59 PM Bug #3637: client: not issuing caps for with clients doing shared writes
- The hang occurs because a client requests a max size increase, but doesn't have write caps, so the mds puts it on the...
- 07:53 AM Bug #3637 (Resolved): client: not issuing caps for with clients doing shared writes
- With 3 clients running ceph-fuse, running the ior command:
/tmp/cephtest/binary/usr/local/bin/ior -e -w -r -W -b 1... - 04:23 PM Feature #3645 (Resolved): Requesting the ability to rename CephFS snapshots inside the ".snap"-di...
- I believe the ability to rename CephFS snapshots can come in handy in many cases. For example, if one wants to imple...
- 12:01 PM Bug #3639 (Duplicate): kclient: hit EOF prematurely
Moved to #3641- 10:56 AM Bug #3639 (Duplicate): kclient: hit EOF prematurely
- Failures seen when running IOR on the kernel client:
WARNING: Task 1 requested transfer of 1048576 bytes,
... - 12:00 PM Bug #3641 (Resolved): kclient: hit EOF prematurely
Failures seen when running IOR on the kernel client:
WARNING: Task 1 requested transfer of 1048576 bytes,
...- 11:57 AM Bug #3640 (Duplicate): kclient: hang and kernel panic
Creating a placeholder for the following issue reported by Eric Renfro on the mailing list:
http://thread.gmane....
12/16/2012
- 09:48 AM Fix #3630 (Resolved): mds: broken closed connection cleanup
- Consider:
- client->mds REQUEST_CLOSE
- mds->client CLOSE
- client closes con
- mds see fault, goes to stan...
12/14/2012
- 06:31 PM Bug #3625: client: EEXIST error on multiple clients to create
- I made some commits to wip-3625, which resolve the EEXIST, but now the test returns an EIO...
- 03:31 PM Bug #3625 (Resolved): client: EEXIST error on multiple clients to create
- Discovered with IOR shared file test on ceph-fuse, if multiple clients attempt to create a file at the same time (do ...
- 03:51 PM Feature #3626 (Resolved): mds: debug mode to generate traceless replies to clients
- 11:13 AM Bug #3610 (Resolved): client: Possible lock cycle in client/objectcacher
- 11:13 AM Bug #3610: client: Possible lock cycle in client/objectcacher
- Merged wip-3610 to next.
12/13/2012
- 07:49 PM Bug #3610 (Fix Under Review): client: Possible lock cycle in client/objectcacher
- This appears to be related to the fork() in the caps.cc test messing up data structures in the lockdep code when its ...
- 03:53 PM Feature #3568 (Resolved): client: Allow hold_caps_until to be configured
- 03:49 PM Feature #3621 (Closed): qa: add knfsd reexport tests to qa suite
- 03:46 PM Cleanup #3423 (Resolved): Install java libraries into the correct directory
- 01:28 PM Bug #3559 (Resolved): mds: not issuing RDCACHE to exclusive client for some files
- Thought more about it, and I think it's right. Committed something to master that describes the logic in a big comment.
- 10:11 AM Bug #3559 (In Progress): mds: not issuing RDCACHE to exclusive client for some files
- After discussion this apparently needs a bit more thought.
- 10:04 AM Bug #3559 (Resolved): mds: not issuing RDCACHE to exclusive client for some files
12/12/2012
- 09:00 PM Bug #2288: libcephfs: setxattr returns EEXIST following removexattr
- Ah, I wasn't aware of this bug. The commit you mentioned is 323a52ee909621ed0169b86e158370394ba36f62. It makes remo...
- 06:40 PM Bug #2288: libcephfs: setxattr returns EEXIST following removexattr
- SamL did some stuff involving projected xattrs; was this problem included in that set of changes or is it more compli...
- 06:38 PM Bug #3369 (Resolved): journaled two client session close events
- Sage merged that workaround a long time ago, and I think Zheng's recent patches might have fixed some potential root ...
- 09:20 AM Bug #3610 (Resolved): client: Possible lock cycle in client/objectcacher
- Teuthology reported this lock cycle while running test_libcephfs. It was triggered by the Caps.ReadZero test.
For...
12/11/2012
- 02:10 PM Bug #3370 (Resolved): All nfsd hung trying to lock page(s) on export of kclient ceph
- commit: 2978257c56935878f8a756c6cb169b569e99bb91
- 09:28 AM Bug #3597 (Can't reproduce): ceph-fuse: denying root access
- I don't see this behavior with fuse 2.9.0 and latest ceph. Does it happen only on some files? What are the permissi...
12/10/2012
- 01:14 PM Bug #3601 (Resolved): client: With multiple clients, file remove doesn't free up space
- If two or more clients have ceph mounted and one client removes a file, the space for that file doesn't get freed on ...
- 01:02 PM Bug #3572 (Resolved): High CPU after equivalent to node very busy
- commit:ed75ec2cd19b47efcd292b6e23f58e56f4c5bc34
- 11:04 AM Bug #3598 (Resolved): MDS should shut down cleanly on EBLACKLIST
- From #3593, suicide() apparently doesn't turn the MDS off so it asserts and core dumps.
- 09:32 AM Bug #3597: ceph-fuse: denying root access
- "denying root access"? You mean root can't read the files, but other people can? Or nobody can?
Either way this is... - 08:47 AM Bug #3597 (Resolved): ceph-fuse: denying root access
- lxo: ceph-fuse also recently started denying root access to files that shouldn't be readable except for root superpo...
- 08:46 AM Bug #3596 (Can't reproduce): ceph-fuse: crash in mds rejoin
- lxo: ceph-fuse crashes in the destructor of a non-empty xlist of a snap_realm upon mds rejoin. no quick reproducer ...
12/07/2012
- 12:27 PM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- I changed those calls to ceph_debug_get_* and update the wip-3559 branch.
12/06/2012
- 01:26 PM Bug #3553: MDS core dumped running 0.48.2argonaut
- Verified just means we agree it's a bug. New->Verified->In Progress->Need Review->Testing->Resolved is the simple pa...
- 10:17 AM Bug #3553: MDS core dumped running 0.48.2argonaut
- How was this bug verified? Would you provide the test and its description in this bug, please.
- 06:47 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- libcephtools is wrapping the monitor interaction stuff; i think this belongs in libcephfs, we just need to advertise ...
- 06:41 AM Feature #3575: ceph-fuse: Add support for forget_multi
- yay! i think this is the answer to #3289.
actually, i think the normal invalidate will work too, one inode at a t... - 05:47 AM Fix #2215: ceph-fuse does not invalidate page cache
- wip-2215 looks good to me!
- 05:33 AM Bug #3572: High CPU after equivalent to node very busy
- Ah, nice catch! We just need to cycle through the list once to requeue the requests for the MDS. If Yan's patch loo...
12/05/2012
- 09:23 PM Fix #2215: ceph-fuse does not invalidate page cache
- Running in teuthology with the multiclient test, I see a segfault in one of the fuse clients on the plana nodes. Wit...
- 05:17 PM Bug #3579 (Resolved): kclient: Use less secure random number generator so we don't consume entropy
The routines __choose_mds() and ceph_mdsmap_get_random_mds() call the kernel get_random_bytes() which gets random n...- 05:05 PM Bug #3572: High CPU after equivalent to node very busy
- There is already a fix for this in wip-fs branch. I'm not clear if we need to be processing through the list once or...
- 04:42 PM Bug #3572: High CPU after equivalent to node very busy
- There is a bug in this routine:...
- 06:14 AM Bug #3572: High CPU after equivalent to node very busy
- The good news is that we don't need high quality entropy here; the crappy stuff will do...
- 04:24 PM Bug #3554 (Resolved): Logs broken after a rotation
- commit:dc93132d8cc1edf83b7e0a603b366cde6750f4d6
- 12:36 PM Bug #3554 (Fix Under Review): Logs broken after a rotation
- 01:24 PM Bug #3404 (Resolved): oops in strlen() from set_request_path_attr()
- Commit: f18b417f1e80f3d9053ea07958d87f87bd20c138
https://github.com/ceph/ceph-client/commit/f18b417f1e80f3d905... - 12:35 PM Bug #3370 (Fix Under Review): All nfsd hung trying to lock page(s) on export of kclient ceph
- 11:14 AM Feature #3575 (New): ceph-fuse: Add support for forget_multi
- A new function in fuse (as of 2.9) supports dropping multiple inodes from the cache in one upcall.
- 11:12 AM Feature #3574 (New): ceph-fuse: Add support for write_buf
A new function (as of fuse 2.9) 'write_buf' in the fuse lowlevel interface supports doing I/O through a pipe allowi...- 11:10 AM Feature #3573 (Rejected): ceph-fuse: Add ioctl support
- 07:47 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- Noah has created a libcephtools library for the admin socket stuff. Should we add it there, or a separate libcephadm...
- 06:46 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- Sweet.
I'm a bit leary about exposing the caps via libcephfs.. that's an implementation detail that users shouldn'... - 06:30 AM Bug #3559 (Fix Under Review): mds: not issuing RDCACHE to exclusive client for some files
- That fixed it. With your fix it now passes my test. I cherry-picked your fix into wip-3559.
- 06:12 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- Yeah your right - the fix I proposed doesn't work because it ends up in mix state eventually. Testing your patch.
- 05:58 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- I don't think we can alloc CACHE caps in the excl->mix state because caching isn't allowed in mix. Generally speakin...
12/04/2012
- 05:59 PM Bug #3572 (Resolved): High CPU after equivalent to node very busy
- I dropped a ceph kernel client node into the debugger and left it there a while. After resuming execution the machin...
- 03:31 PM Feature #3570 (Resolved): teuthology: mds thrasher
- 02:08 PM Feature #3568 (Fix Under Review): client: Allow hold_caps_until to be configured
- Made this change to the wip-3568 branch. Needs review.
- 11:26 AM Feature #3568 (Resolved): client: Allow hold_caps_until to be configured
- Client::cap_delay_requeue has a hardcoded timeout of 5 secs. Make this configurable.
- 01:24 PM Bug #1047: mds: crash on anchor table query
- commit:a3aad3c3c58ffd7cd556275cd1c7dbc69b3eb4dd, which is in v0.55.
- 11:24 AM Bug #3559: mds: not issuing RDCACHE to exclusive client for some files
- I was able to reproduce this with a libcephfs test that creates, opens and closes files, suspends the client for 20 s...
Also available in: Atom