Project

General

Profile

Activity

From 04/04/2016 to 05/03/2016

05/03/2016

02:33 PM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
I don't think that's acceptable unless we have a way of verifying nobody else has modified (or maybe even just touche... Greg Farnum
03:01 AM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
Greg Farnum wrote:
> Zheng, does that patch mean that the client no longer drops any dirty state? Or does it keep th...
Zheng Yan
05:37 AM Bug #15694 (Closed): process enter D status when running on rbd
Hi,
Seldom, some kind of process running with data locate on rbd(formated with ext4) may enter D status(stuck).
I c...
Benetnash Alkaid
04:08 AM Bug #12159 (Need More Info): setfilelock requests stuck
Zheng Yan
03:59 AM Bug #14487 (Resolved): ctime may go backward in cephfs kernel client
Zheng Yan
03:58 AM Bug #14232 (Resolved): Kernel NULL pointer dereference in __dcache_readdir
Zheng Yan
03:58 AM Bug #14485 (Resolved): cephfs kernel client does not flush ctime
Zheng Yan
03:57 AM Bug #15552 (Closed): cephfs kernel client timeout error 110
Zheng Yan

05/02/2016

04:46 PM Bug #15683: RBD image features settings are not copied onto the clone.
tnx Rohith Radhakrishnan
04:41 PM Bug #15683: RBD image features settings are not copied onto the clone.
See ticket #15388 Jason Dillaman
09:09 AM Bug #15683: RBD image features settings are not copied onto the clone.
Attaching the output from all commands:-
root@host1:~$ ceph osd pool create pool4 1024
pool 'pool4' created
ro...
Rohith Radhakrishnan
09:06 AM Bug #15683 (Closed): RBD image features settings are not copied onto the clone.
Steps to reproduce:-
1) Create rbd image
2) Disable some features like layering, exclusive-lock, object-map etc....
Rohith Radhakrishnan

04/29/2016

08:44 PM Bug #14360: Cephfs kernel client ceph_send_cap_releases hung task
I'm not seeing those warnings. strace shows read() and write() immediately returning EIO. I'll see if I can try a 4.6... Matthew Garrett
12:00 PM Bug #14360: Cephfs kernel client ceph_send_cap_releases hung task
Matthew Garrett wrote:
> I've pulled the 4.6 ceph code back to 4.5, but any attempt to read or write to a file is no...
Zheng Yan
01:55 PM Bug #13712 (Resolved): Page allocation failure in Kernel 4.2.3.
Ah, this was a GFP_ATOMIC allocation (which is not allowed to block and therefore more likely to fail). We used to r... Ilya Dryomov
12:18 PM Bug #15448 (Resolved): Jewel : KRBD : rbd map is failing, we should log correct error message eit...
"rbd: report unsupported features to syslog" in 4.6-rc6. Ilya Dryomov
12:17 PM Bug #15490 (Resolved): rbd map vs notify race
"rbd: fix rbd map vs notify races" in 4.6-rc6. Ilya Dryomov
12:16 PM Bug #15447 (Resolved): use-after-free on ceph_auth_none_info / ceph_none_authorizer
"libceph: make authorizer destruction independent of ceph_auth_client" in 4.6-rc6. Ilya Dryomov

04/28/2016

07:44 PM Bug #14360: Cephfs kernel client ceph_send_cap_releases hung task
I've pulled the 4.6 ceph code back to 4.5, but any attempt to read or write to a file is now giving me EIO. Is there ... Matthew Garrett

04/24/2016

04:12 AM Bug #15552: cephfs kernel client timeout error 110
I found a solution I can live with.
I set:
@ceph osd crush tunables default@
And this works without generati...
Heath Jepson
03:17 AM Bug #15552: cephfs kernel client timeout error 110
OK, so my last update led me to trying a shot in the dark.
running:
@ceph osd crush tunables legacy@
Allowed...
Heath Jepson
03:08 AM Bug #15552: cephfs kernel client timeout error 110
more info, I see this pop up on the mon when running in debug mode:
@2016-04-23 21:03:28.685457 7f2eead54700 1 --...
Heath Jepson

04/21/2016

06:58 AM Bug #15552: cephfs kernel client timeout error 110
4.3.0-0.bpo.1-amd64 Heath Jepson
06:44 AM Bug #15552: cephfs kernel client timeout error 110
which version of kernel do you use Zheng Yan

04/20/2016

11:03 PM Bug #15552 (Closed): cephfs kernel client timeout error 110
After upgrading my test environment from infernalis to the latest jewel release candidate last night, I cannot mount ... Heath Jepson

04/18/2016

07:48 AM Bug #14360 (Resolved): Cephfs kernel client ceph_send_cap_releases hung task
upstream commit 315f24088048a51eed341c53be66ea477a3c7d16 Zheng Yan
07:47 AM Bug #14736: Kernel cephfs bind-mounts break if parent dir modified elsewhere
upstream commit 200fd27c8fa2ba8bb4529033967b69a7cbfa2c2e Zheng Yan
07:42 AM Bug #14736 (Resolved): Kernel cephfs bind-mounts break if parent dir modified elsewhere
Zheng Yan

04/15/2016

07:26 PM Bug #15448 (Fix Under Review): Jewel : KRBD : rbd map is failing, we should log correct error mes...
Ilya Dryomov
05:28 PM Bug #15490 (Fix Under Review): rbd map vs notify race
Ilya Dryomov
12:29 PM Bug #15432: kcephfs: umount -f can fail after mds reconnect failure
find a bug that may cause forced umount hang
https://github.com/ceph/ceph-client/commit/e9344de458cd61efac6cccb98c78...
Zheng Yan

04/14/2016

09:35 PM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
Sage, I think we discussed this a day or two ago in standup but I don't think you've looked at it yet. Please comment... Greg Farnum
09:33 PM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
Zheng, does that patch mean that the client no longer drops any dirty state? Or does it keep the dirty page cache and... Greg Farnum
09:51 AM Bug #15432: kcephfs: umount -f can fail after mds reconnect failure
should be fixed by https://github.com/ceph/ceph-client/commit/11ff967a17ea6ac638f3c154b737a094f29d30bb Zheng Yan
09:48 AM Bug #15462: ceph: build_path did not end path lookup where expected, namelen is 146, pos is 0
"ceph: build_path did not end path lookup where expected" does not indicate there is an issue.... Zheng Yan

04/13/2016

08:01 PM Bug #15490 (Resolved): rbd map vs notify race
... Ilya Dryomov

04/11/2016

08:17 PM Bug #15462 (Won't Fix): ceph: build_path did not end path lookup where expected, namelen is 146, ...
... Sage Weil
05:46 PM Bug #15447 (Fix Under Review): use-after-free on ceph_auth_none_info / ceph_none_authorizer
Ilya Dryomov
04:36 PM Bug #15448: Jewel : KRBD : rbd map is failing, we should log correct error message either in syst...
Thanks Ilya! Vikhyat Umrao
03:16 PM Bug #15448 (In Progress): Jewel : KRBD : rbd map is failing, we should log correct error message ...
It can't pretty-print which features are missing, in general. But we do have to make it nicer. Ilya Dryomov

04/10/2016

06:36 PM Bug #15448 (Resolved): Jewel : KRBD : rbd map is failing, we should log correct error message eit...
Jewel : KRBD : rbd map is failing , we should log correct error message either in system logs or in standard error ou... Vikhyat Umrao
01:15 PM Support #15302: Kernel panic when using CEPHFS (Hammer) /w kernel vivid-lts on Ubuntu 14.04
v3.18.[26-30] are also affected. I've asked Sasha to queue up "ceph: fix request timestamp encoding". Ilya Dryomov
01:01 PM Bug #15447 (Resolved): use-after-free on ceph_auth_none_info / ceph_none_authorizer
... Ilya Dryomov

04/08/2016

03:46 PM Support #15302: Kernel panic when using CEPHFS (Hammer) /w kernel vivid-lts on Ubuntu 14.04
This fix is still queued and pending in the Ubuntu repo's, but has not yet been released (as of 3.19.0-58.x).
In t...
Kamal Mostafa
10:54 AM Support #15302: Kernel panic when using CEPHFS (Hammer) /w kernel vivid-lts on Ubuntu 14.04
Sorry, i was wrong. I'm still getting the same panic with 3.19.0.58.41 Dennis Kramer
10:12 AM Support #15302: Kernel panic when using CEPHFS (Hammer) /w kernel vivid-lts on Ubuntu 14.04
Thank you.
I can also confirm that it's fixed in the latest linux-generic-lts-vivid (3.19.0.58.41) which i installe...
Dennis Kramer
02:22 PM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
This happened again yesterday. It kind of seems like there's more going on that leads to the failed reconnect. :/
...
Greg Farnum
09:49 AM Bug #15255: kclient: lost filesystem operations (on mds disconnect?)
https://github.com/ceph/ceph-client/commit/11ff967a17ea6ac638f3c154b737a094f29d30bb Zheng Yan
01:20 PM Bug #15432 (New): kcephfs: umount -f can fail after mds reconnect failure
We've seen this several times on the lab cluster and teuthology:
- fail to reconnect to mds for some reason
- umo...
Sage Weil

04/05/2016

01:55 AM Support #6455 (Closed): dumpling cephFS on 3.12-rc1 qemu hung tasks
Zheng Yan
 

Also available in: Atom