Project

General

Profile

Activity

From 06/29/2015 to 07/28/2015

07/28/2015

03:18 PM Backport #12500 (In Progress): segfault launching ceph-fuse with bad --name
Loïc Dachary
10:37 AM Backport #12500 (Resolved): segfault launching ceph-fuse with bad --name
https://github.com/ceph/ceph/pull/5382 Loïc Dachary
03:18 PM Backport #12499 (In Progress): ceph-fuse 0.94.2-1trusty segfaults / aborts
Loïc Dachary
10:37 AM Backport #12499 (New): ceph-fuse 0.94.2-1trusty segfaults / aborts
Loïc Dachary
10:36 AM Backport #12499 (Resolved): ceph-fuse 0.94.2-1trusty segfaults / aborts
https://github.com/ceph/ceph/pull/5381 Loïc Dachary
03:03 PM Bug #12506 (Resolved): "Fuse mount failed to populate" error
Run: http://pulpito.ceph.com/teuthology-2015-07-27_16:05:09-upgrade:hammer-hammer-distro-basic-vps/
Jobs: ['988514',...
Yuri Weinstein
01:41 PM Bug #11255 (Fix Under Review): nfs: mount failures on ceph-backed NFS share
Greg Farnum
01:40 PM Bug #12426 (Pending Backport): "ceph mds add_data_pool" check for EC pool is wrong
John Spray

07/23/2015

01:57 PM Feature #12445 (New): improved path enforcement for hard links
The initial path restriction will not work properly for remote links inside the security domain that link to a primar... Sage Weil
01:05 PM Bug #12355 (Resolved): MDS assertion during shutdown (MDLog !capped), in TestStrays.test_migratio...
John Spray
09:50 AM Bug #12355 (Fix Under Review): MDS assertion during shutdown (MDLog !capped), in TestStrays.test_...
https://github.com/ceph/ceph/pull/5324 John Spray
09:46 AM Bug #12426 (Fix Under Review): "ceph mds add_data_pool" check for EC pool is wrong
https://github.com/ceph/ceph/pull/5323 John Spray
08:18 AM Bug #12209: CephFS should have a complete timeout mechanism to avoid endless waiting or unpredict...
i think you can send a Pull request on github huang jun
02:02 AM Feature #12334: nfs-ganesha: handle client cache pressure in NFS Ganesha FSAL
I finished the final test using 1 Ceph file system client and the fuse interface. I ran the create 5 million file te... Eric Eastman

07/22/2015

03:38 PM Bug #12417 (Pending Backport): segfault launching ceph-fuse with bad --name
Sage Weil
12:59 PM Bug #12355 (In Progress): MDS assertion during shutdown (MDLog !capped), in TestStrays.test_migra...
Reproduced this locally. Fixing... John Spray
03:22 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
Zheng Yan wrote:
> https://github.com/ceph/ceph/pull/5222
Hi Yan,
We found a serious problem with this patch w...
Zhi Zhang
02:00 AM Bug #12354 (Resolved): ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
fuse fsyncdir callback was added recently Zheng Yan

07/21/2015

09:43 PM Bug #12426 (Resolved): "ceph mds add_data_pool" check for EC pool is wrong

We have a generic MDSMonitor::_check_pool that's used in fs new, but add_data_pool is just checking if something is...
John Spray
03:14 PM Bug #12417 (Fix Under Review): segfault launching ceph-fuse with bad --name
https://github.com/ceph/ceph/pull/5305 John Spray
10:09 AM Bug #12417 (Resolved): segfault launching ceph-fuse with bad --name
This is just in a vstart environment -- running ceph-fuse with no --name arg is fine.... John Spray
09:54 AM Bug #11255: nfs: mount failures on ceph-backed NFS share
Did you mean to reference https://github.com/ceph/teuthology/pull/554, Zheng? Greg Farnum
09:35 AM Bug #11255: nfs: mount failures on ceph-backed NFS share
maybe https://github.com/ceph/teuthology/pull/342 can fix this Zheng Yan
08:37 AM Bug #11255: nfs: mount failures on ceph-backed NFS share
http://pulpito.ceph.redhat.com/teuthology-2015-07-15_18:10:02-knfs-hammer-testing-basic-magna/153971/ John Spray
09:51 AM Bug #12357: kclient: some kind of hang
Sorry I missed that. I believe the missing bzip2 is being handled elsewhere by the teuthology guys. Greg Farnum
02:37 AM Bug #12357 (Won't Fix): kclient: some kind of hang
... Zheng Yan
09:48 AM Bug #12354 (Pending Backport): ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
I guess this is a long-standing issue, but given CephFS' support state just backporting to hammer should be fine. Greg Farnum
02:38 AM Bug #12354 (Fix Under Review): ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
https://github.com/ceph/ceph/pull/5294 Zheng Yan
02:27 AM Bug #12365: kcephfs: hang on umount
by commit https://github.com/ceph/ceph-client/commit/d2ca589aa92410b6760047dea877ca6edd0b15c7 Zheng Yan

07/20/2015

07:33 PM Bug #12297 (Pending Backport): ceph-fuse 0.94.2-1trusty segfaults / aborts
Excellent! This is merged to master now. Greg Farnum
06:45 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Cluster has been processing data without any problem during the weekend. I consider the bug fixed. Burkhard Linke
07:08 PM Feature #12334: nfs-ganesha: handle client cache pressure in NFS Ganesha FSAL
I am seeing the "Client <foo> failing to respond to cache pressure" warning using the Ceph Kernel driver after creat... Eric Eastman

07/19/2015

03:52 AM Feature #12334: nfs-ganesha: handle client cache pressure in NFS Ganesha FSAL
I ran the same 5 million file create test using a cifs mount instead of a NFS mount and did not see the "Client <foo>... Eric Eastman

07/17/2015

04:26 PM Bug #12354: ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
I've got the core file and appropriate packages on vpm119 if you want an environment to look at it with. Greg Farnum
04:25 PM Bug #12354: ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
I'm not finding where it's gone wrong, but I think this must be an issue with the new refcounting. Please take a look... Greg Farnum
03:40 PM Bug #12354: ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
Or more likely we're freeing the Fh inappropriately and reusing the memory for a dir_result_t, since we control the l... Greg Farnum
03:36 PM Bug #12354: ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
Looking at the core dump, it's crashing on dereferencing a NULL Inode* contained in the passed-in Fh structure. But t... Greg Farnum
01:30 PM Bug #12354 (In Progress): ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync
Greg Farnum
01:07 PM Bug #12344: libcephfs-java/test.sh: com.ceph.fs.CephMountTest fails
We haven't seen this in any of our regular runs, which makes me think it's a config issue in your OpenStack setup. Se... Greg Farnum
11:42 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Branch is deployed to the compute cluster host and the application causing the problem is started.
Debug output a...
Burkhard Linke
10:00 AM Bug #12365 (In Progress): kcephfs: hang on umount
Zheng Yan

07/16/2015

03:57 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
https://github.com/ceph/ceph/pull/5260 Greg Farnum
02:42 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Burkhard, if you can test out the hammer-12297 branch on one of the boxes seeing this bug and find out if that fixes ... Greg Farnum
12:07 PM Bug #12297 (Fix Under Review): ceph-fuse 0.94.2-1trusty segfaults / aborts
I think I have a fix for this issue but I want to get a few eyes on it before handing it off for user testing. Greg Farnum
03:51 PM Bug #12365 (Resolved): kcephfs: hang on umount
http://pulpito.ceph.com/teuthology-2015-07-13_23:08:02-kcephfs-next-testing-basic-multi/972963/
There was an error...
Greg Farnum
02:37 PM Bug #12357 (Won't Fix): kclient: some kind of hang
http://pulpito.ceph.redhat.com/teuthology-2015-07-12_18:08:02-kcephfs-hammer-testing-basic-magna/151448/
It looks ...
Greg Farnum
02:16 PM Bug #12355: MDS assertion during shutdown (MDLog !capped), in TestStrays.test_migration_on_shutdown
See also http://pulpito.ceph.com/teuthology-2015-07-03_23:04:02-fs-master---basic-multi/959994/ Greg Farnum
01:56 PM Bug #12355 (Resolved): MDS assertion during shutdown (MDLog !capped), in TestStrays.test_migratio...

/a/teuthology-2015-07-13_23:04:03-fs-master---basic-multi/972783
mds.a crashed...
John Spray
01:53 PM Bug #12354 (Resolved): ceph-fuse crash in ll_fsync, during TestClientRecovery.test_fsync

teuthology-2015-07-13_23:04:03-fs-master---basic-multi/972688/
ceph-client.0.4375.log...
John Spray
08:59 AM Backport #12350 (Resolved): Provided logrotate setup does not handle ceph-fuse correctly
https://github.com/ceph/ceph/pull/7110
Burkhard Linke
02:22 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
Hi, Yan. I think it is also unsafe in the _read_async code when this patch applied. When thread1 wait on read, threa... Yan Shen

07/15/2015

09:47 PM Bug #12344 (Can't reproduce): libcephfs-java/test.sh: com.ceph.fs.CephMountTest fails
This is on an OpenStack virtual machine with 8GB RAM, 40GB disk, 1CPU.... Loïc Dachary
02:43 PM Bug #12321: MDS crash when try to connect clients
After you've destroyed the MDS's record of client sessions, it's important to make sure that those clients don't exis... John Spray
05:18 AM Bug #12321: MDS crash when try to connect clients
zcc icy wrote:
> John Spray wrote:
> > This indicates an inconsistency between the inodes apparently allocated to t...
zcc icy
04:38 AM Bug #12321: MDS crash when try to connect clients
John Spray wrote:
> This indicates an inconsistency between the inodes apparently allocated to the session, and thos...
zcc icy
04:34 AM Bug #12321: MDS crash when try to connect clients
Greg Farnum wrote:
> What's the disaster that happened? Did you lose some of your RADOS objects?
I deploy 3 mon &...
zcc icy
11:21 AM Feature #12334: nfs-ganesha: handle client cache pressure in NFS Ganesha FSAL
Note that NFS-Ganesha created significantly more inodes than the cache size limit before it had too many pinned. So i... Greg Farnum
07:31 AM Feature #12334 (Resolved): nfs-ganesha: handle client cache pressure in NFS Ganesha FSAL
Reported by Eric Eastman on ceph-users: http://lists.ceph.com/pipermail/ceph-users-ceph.com/2015-July/003000.html
...
John Spray

07/14/2015

04:53 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Okay, I dug into this today and I think our readdir is a bit broken (obviously). In particular, we're dropping the cl... Greg Farnum
09:48 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
We've got logs and core file via ceph-post-file
c00fb309-e2a1-482c-b093-1733affb2901 gzipped core dump
60fbd6a7...
Greg Farnum
08:03 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
I can't figure out what's wrong neither Zheng Yan
02:43 PM Bug #12321: MDS crash when try to connect clients
This indicates an inconsistency between the inodes apparently allocated to the session, and those marked as free in t... John Spray
02:19 PM Bug #12321: MDS crash when try to connect clients
What's the disaster that happened? Did you lose some of your RADOS objects? Greg Farnum
08:23 AM Bug #12321: MDS crash when try to connect clients
I have 3 mon 9 osd .
when I recovery after the disaster happened. I found that mds can't run.
log is in the file.
zcc icy
08:19 AM Bug #12321 (Can't reproduce): MDS crash when try to connect clients
I have 3 mon 9 osd .
when I recovery after the disaster happened. I found that mds can't run.
log :
2015-...
zcc icy
10:45 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
Backports team: Zheng created a backport PR at https://github.com/ceph/ceph/pull/5222, but let's let it bake a bit in... Greg Farnum
06:45 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
https://github.com/ceph/ceph/pull/5222 Zheng Yan
07:39 AM Backport #12319 (In Progress): cephfs client crash after enable readahead mechanism through setti...
Nathan Cutler
07:38 AM Backport #12319 (Resolved): cephfs client crash after enable readahead mechanism through setting ...
https://github.com/ceph/ceph/pull/5222 and https://github.com/ceph/ceph/pull/5427 Nathan Cutler
03:52 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Hi Yan,
We have tested both patches for few days and no above issue happened again. We will perform some other tes...
Zhi Zhang

07/13/2015

02:40 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Zheng, can you take a look at this and see if you need the log or can identify the issue? I presume something's gone ... Greg Farnum
02:33 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
(gdb) p dir
$2 = (Dir *) 0x7f709000f9a0
(gdb) p dir->dentry_list
$3 = {_front = 0x7f6f8002a260, _back = 0x7f70b83a...
Burkhard Linke
02:25 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Okay, I think I see what happened, one more please.
[run gdb]
f 4
p dir
p dir->dentry_list
And if you could ...
Greg Farnum
02:18 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
(gdb) f 4
#4 Client::_readdir_cache_cb (this=this@entry=0x38a8ab0, dirp=dirp@entry=0x7f6f7c018400, cb=cb@entry=
...
Burkhard Linke
02:14 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
That's pretty bizarre. Signal 7 is SIGBUS "Bus error (bad memory access)", which is separate from SIGSEGV. And given ... Greg Farnum
01:48 PM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
(gdb) bt
#0 0x00007f70d7bf020b in raise (sig=7)
at ../nptl/sysdeps/unix/sysv/linux/pt-raise.c:37
#1 0x000000...
Burkhard Linke
11:12 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
You mean the ulimit for dumping the core file? That will depend on how you're invoking it; if you're just running "ce... Greg Farnum
11:04 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Sorry, no core file available. I have a number of hosts with CephFS running without any problems; some other hosts sh... Burkhard Linke
10:48 AM Bug #12297: ceph-fuse 0.94.2-1trusty segfaults / aborts
Do you have a core file from this that you can look at with gdb, and the ceph debug packages installed? (You can pull... Greg Farnum
10:33 AM Bug #12297 (Resolved): ceph-fuse 0.94.2-1trusty segfaults / aborts
Running CephFS for data volumes on a compute cluster.
ceph-fuse aborts and leaves the mount point unusable (transp...
Burkhard Linke
12:40 PM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
... John Spray
12:40 PM Bug #12088 (Pending Backport): cephfs client crash after enable readahead mechanism through setti...
I imagine we probably want backports to firefly + hammer? John Spray
02:24 AM Fix #12296 (Rejected): cephfs-hadoop: do not stash libcephfs.jar in git repo
The libcephfs.jar artifact should not be stored in the cephfs-hadoop git repository. Ideally it lands in Maven Centra... Noah Watkins

07/10/2015

02:25 PM Feature #12284 (New): Deprioritise forward scrub on highly loaded MDS/cluster

When MDS cpu load is high, slow or pause scrubs.
Do a similar thing if we are already sending many RADOS ops per...
John Spray
02:21 PM Feature #12283 (New): Time of day scheduling for background forward scrub

Probably makes sense to handle this as a blacklist, essentially to let admin define their "business hours" during w...
John Spray
02:17 PM Feature #12282 (Resolved): mds: progress/abort/pause interface for ongoing scrubs

Progress (n dentries scrubbed of m within this subtree) may be calculated based on rstats.
Abort and pause shoul...
John Spray
02:10 PM Feature #11950: Strays enqueued for purge cause MDCache to exceed size limit
Related to FSCK: if adding a persistent structure containing references to inodes to be purged, ensure backward scrub... John Spray
02:02 PM Feature #12281 (New): Respect PIN_SCRUBQUEUE from StrayManager

When purging, the StrayManager needs to notice metadata subject to PIN_SCRUBQUEUE and unhook items from the scrub s...
John Spray
01:27 PM Feature #12280 (New): cephfs-data-scan: online injection
This is a third peer to the local filesystem and metadata pool injection: injecting inodes using a running MDS cluste... John Spray
01:26 PM Feature #12279 (New): MDS hooks for online injection from backward scan

- Add hooks to MDS to enable backward scrub to lookup_ino and work out
if an "orphaned ino" is actually orphan...
John Spray
01:25 PM Feature #12278 (New): Recover backtrace hints from backward scrub and move dirs to highest versio...
Exploit the backtraces saved in #12277 -- where we see more recent linkage information for any ancestor, move it to r... John Spray
01:23 PM Feature #12277 (New): cephfs-data-scan: Optionally save latest backtrace when scanning

While we inject metadata with the first backtrace we see (in order to enable parallelism), we may later see more up...
John Spray
01:21 PM Feature #12276 (New): Handle MDS rank shutdown during forward scrub

Forward scrub may be targeting e.g. the stray dirs belong to an MDS, that get torn down during MDS rank shutdown.
...
John Spray
01:18 PM Feature #12275 (Duplicate): Handle metadata migration during forward scrub
Two candidate approaches here:
* Block migration while a (tagging) scrub is ongoing
OR
* When a migration ...
John Spray
01:16 PM Feature #12274 (Resolved): mds: start forward scrubs from all subtree roots, skip non-auth metadata

This is the basic principle that allows scrubs across multiple MDSs: everyone takes responsibility for their own su...
John Spray
01:13 PM Feature #12273 (Resolved): Apply scrub tag to objects during forward scrub
Where a tag was specified (#12258), write this into an xattr of the objects touched by the scrub (inode headers in th... John Spray
09:25 AM Bug #12253 (Resolved): Sometimes mds dump_ops_in_flight will crash mds
Zheng Yan
05:44 AM Bug #12253: Sometimes mds dump_ops_in_flight will crash mds
Please make the code check is_slave() first, then check if slave_request is NULL. If slave_request is NULL, just outp... Zheng Yan
03:20 AM Bug #12253: Sometimes mds dump_ops_in_flight will crash mds
Pls check the mdr->slave_to_mds below:... Zhi Zhang
02:07 AM Bug #12253: Sometimes mds dump_ops_in_flight will crash mds
Yes, we were running multiple mds when hitting this issue. Now we have switched to single mds. Zhi Zhang
05:27 AM Feature #12265 (New): Add Iohint in cephfs
This is the plan of BP "Add Iohint in Cephfs"(http://tracker.ceph.com/projects/ceph/wiki/Add_IOhint_in_CephFS).
By...
jianpeng ma

07/09/2015

11:43 AM Bug #12253: Sometimes mds dump_ops_in_flight will crash mds
are you running multiple mds? could you check value of mdr->slave_to_mds Zheng Yan
05:53 AM Bug #12253: Sometimes mds dump_ops_in_flight will crash mds
https://github.com/ceph/ceph/pull/5175
We've made some minor changes. Can someone help to take a look?
Thanks.
Zhi Zhang
05:52 AM Bug #12253 (Resolved): Sometimes mds dump_ops_in_flight will crash mds
Sometimes there are no suitable ops in mds, then if we run dump_ops_in_flight, mds will crash due to assertion failur... Zhi Zhang
11:41 AM Feature #12258 (Resolved): asok command to initate tagged forward scrub

Taking <path> and <tag> arguments.
Start a forward scrub from the given location, applying this particular tag t...
John Spray
11:09 AM Feature #12257 (New): Scrub recovery&resume on MDS replay

During replay, load the state we stored when starting scrubs (#12255) and use it to reconstruct the in-memory recor...
John Spray
09:18 AM Feature #12255 (New): Create ScrubMap structure in MDLog

This is where we will record what scrubs are going on, and what tags are associated with them. This is essential s...
John Spray

07/08/2015

01:28 PM Backport #11737 (Resolved): MDS is crashed (mds/CDir.cc: 1391: FAILED assert(!is_complete()))
Loïc Dachary
12:55 PM Backport #12098 (In Progress): kernel_untar_build fails on EL7
Loïc Dachary
12:55 PM Backport #11999 (In Progress): cephfs Dumper tries to load whole journal into memory at once
Loïc Dachary
12:12 PM Support #11923 (Resolved): MDS init script starts multiple instances when MDS is referenced in ce...
Yes, in a default configuration you don't list the daemons and sysvinit starts them based on what folders exist in /v... Greg Farnum
11:57 AM Support #11923: MDS init script starts multiple instances when MDS is referenced in ceph.conf
Greg,
I'd like to put in a slight correction in the 'ps' output. even it may be the same as what Brian said in the...
Vimal A.R
09:43 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
please use the new timeout patch Zheng Yan
08:15 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
please try the attached patches. (the second one is modified version of your timeout patch) Zheng Yan
02:10 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Hi Greg,
Not exactly. We have enabled timeout in 2 ways, one is the patch you saw, the other is in Client::tick()...
Zhi Zhang

07/07/2015

04:25 PM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Do I guess correctly that by the timeout mechanism you're referring to the patch in http://tracker.ceph.com/issues/12... Greg Farnum
12:07 PM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
We met this issue *because we enabled timeout mechanism in tick on cephfs client for all the mds requests*. Once time... Zhi Zhang
04:02 PM Support #11923: MDS init script starts multiple instances when MDS is referenced in ceph.conf
How was this ceph.conf and MDS set up? If you remove the "sysvinit" file from the MDS directory it shouldn't be start... Greg Farnum
03:09 PM Bug #12222 (Resolved): MDSMonitor: set max_mds doesn't respect MAX_MDS
Various allocation ranges are calculated based on MAX_MDS, but we don't actually stop the user from creating more MDS... John Spray

07/06/2015

02:08 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Thanks, guys. I will keep digging to see if we can find something. Will keep you updated and pls stay tuned. Zhi Zhang

07/03/2015

10:49 AM Feature #9755: Fence late clients during reconnect timeout
Nope -- the machinery went in to barrier on OSD epoch after blacklisting a client, but the actual act of blacklisting... John Spray
08:41 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Hi David,
After changing CentOS 7 client OS to Fedora 21 (kernel 3.18) the deadlock problem disappeared.
I did ...
Ilja Slepnev
05:53 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
please dump mds when this hapeens (ceph mds tell \* dumpcache, cache dump file at /cachedump.* on the machine that ru... Zheng Yan
05:35 AM Bug #10944: Deadlock, MDS logs "slow request", getattr pAsLsXsFs failed to rdlock
Hi Greg and Yan,
We are using ceph-dokan on windows and hitting the same problem few times. And this deadlock happ...
Zhi Zhang
05:18 AM Bug #12209 (Won't Fix): CephFS should have a complete timeout mechanism to avoid endless waiting ...
Recently, when we made pressure test on cephfs through ceph-dokan using in windows, there are always some ceph-dokan ... Wenjun Huang

07/02/2015

09:30 PM Feature #12204 (Resolved): ceph-fuse: warn and shut down when there is no MDS present
Right now if you try to mount ceph-fuse and there's no MDS in the system, it simply hangs. This is confusing for new ... Greg Farnum
09:27 PM Feature #8358: client: opportunistically update backtraces on files
I think the steps to do this are:
1) Have MDS provide a bufferlist to the client whenever an inode is created (as an...
Greg Farnum
08:55 PM Feature #9755: Fence late clients during reconnect timeout
Didn't this get done when the epoch barrier stuff did? (If not, please unassign.) Greg Farnum
11:16 AM Support #11923: MDS init script starts multiple instances when MDS is referenced in ceph.conf
Hello Sage,
The listing of /var/lib/ceph/mds is as following:
~~~
# ls -l /var/lib/ceph/mds/
total 0
drwxr-x...
Vimal A.R
10:48 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
no extra lock is needed. the async readahead context is called while client_lock is locked. Zheng Yan
09:30 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
hi Yan,
Thanks for the patch. We also have a similar patch tested internally, which uses readahead's pending count...
Zhi Zhang
08:47 AM Bug #12088: cephfs client crash after enable readahead mechanism through setting conf option 'cli...
please try the attached patch Zheng Yan
09:01 AM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
no idea what happened, please try using newer kernel (such as 4.0 kernel) on both NFS server and NFS client Zheng Yan
07:59 AM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
The desktop machine do not have access to the ceph network at all. That's why I have to use a NFS gateway.
dd on N...
Burkhard Linke
07:40 AM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
it's likely the client on your desktop machine does not RW permission to the pools. please try doing direct write on ... Zheng Yan
07:02 AM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
Zheng Yan wrote:
> it's likely your client does not have RW permission to the pools
I don't think the problem is ...
Burkhard Linke
01:27 AM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
it's likely your client does not have RW permission to the pools Zheng Yan
02:56 AM Bug #11481: "mds/MDSTable.cc: 146: FAILED assert(is_undef())" on standby->replay transition
fixed in https://github.com/ceph/ceph/pull/4658 Kefu Chai

07/01/2015

05:32 PM Bug #11746: cephfs Dumper tries to load whole journal into memory at once
Hammer backport: https://github.com/ceph/ceph/pull/5120 Abhishek Lekshmanan
03:51 PM Bug #12094 (Duplicate): "Segmentation fault" in smoke-master-distro-basic-multi run
#12123 Greg Farnum

06/30/2015

10:29 PM Cleanup #12191 (Resolved): Remove ceph-mds --journal-check aka ONESHOT_REPLAY

Now that we have separate tools for validating the journal, we should remove MDSMap::STATE_ONESHOT_REPLAY -- it add...
John Spray
02:57 PM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
Replicated pools also seems to be affected:
On client:
:/ceph/test$ ls
:/ceph/test$ touch foo
:/ceph/test$ cp ...
Burkhard Linke
02:36 PM Bug #12189: Editing / Creating files fails for NFS-over-CephFS on EC pool with cache tier
~# ceph df
GLOBAL:
SIZE AVAIL RAW USED %RAW USED
126T 53823G 75664G 58.38 ...
Burkhard Linke
02:33 PM Bug #12189 (Won't Fix): Editing / Creating files fails for NFS-over-CephFS on EC pool with cache ...
Ubuntu 14.04, Kernel 3.13.0-55-generic
Standard kernel-based NFS server
Ceph Hammer release
~# ceph version
ceph...
Burkhard Linke
09:23 AM Bug #12123 (Resolved): testlibcephfs: segfault in preadv/pwritev tests
Kefu Chai
03:40 AM Bug #9994 (Fix Under Review): ceph-qa-suite: nfs mount timeouts
Zheng Yan
03:40 AM Bug #9994: ceph-qa-suite: nfs mount timeouts
https://github.com/ceph/teuthology/pull/554 Zheng Yan

06/29/2015

11:23 AM Bug #12172 (Resolved): tasks.cephfs.test_auto_repair.TestMDSAutoRepair fails
... John Spray
09:55 AM Bug #12172 (Fix Under Review): tasks.cephfs.test_auto_repair.TestMDSAutoRepair fails
Zheng Yan
09:54 AM Bug #12172: tasks.cephfs.test_auto_repair.TestMDSAutoRepair fails
https://github.com/ceph/ceph-qa-suite/pull/473 Zheng Yan
 

Also available in: Atom