Project

General

Profile

Activity

From 01/01/2018 to 01/30/2018

01/30/2018

11:31 PM Feature #16775 (Resolved): MDS command for listing open files
Patrick Donnelly
11:30 PM Bug #22610 (Pending Backport): MDS: assert failure when the inode for the cap_export from other M...
Patrick Donnelly
11:29 PM Bug #22734 (Pending Backport): cephfs-journal-tool: may got assertion failure due to not shutdown
Patrick Donnelly
11:27 PM Bug #22730 (Pending Backport): mds: scrub crash
Patrick Donnelly
11:26 PM Bug #22776 (Pending Backport): mds: session count,dns and inos from cli "fs status" is always 0
Patrick Donnelly
11:26 PM Bug #22741 (Pending Backport): osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-...
Patrick Donnelly
10:45 PM Bug #22754 (Fix Under Review): mon: removing tier from an EC base pool is forbidden, even if allo...
https://github.com/ceph/ceph/pull/20190 Patrick Donnelly
10:36 PM Bug #22754 (In Progress): mon: removing tier from an EC base pool is forbidden, even if allow_ec_...
Patrick Donnelly
05:33 PM Bug #21402 (Fix Under Review): mds: move remaining containers in CDentry/CDir/CInode to mempool
https://github.com/ceph/ceph/pull/19954
Also ran two 64-client kernel build tests (one patched, one master) with a...
Patrick Donnelly
03:42 PM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
mds_blacklist_interval = 1440
We found that that arguments is too little for the HA testing, it should be adjusted l...
Yong Wang
02:01 PM Feature #12107: mds: use versioned wire protocol; obviate CEPH_MDS_PROTOCOL
have added the support for MDentryLink and MDentryUnlink,
next step is CInode::encode_lock_state/CInode::decode_lock...
dongdong tao
03:32 AM Bug #22788: ceph-fuse performance issues with rsync
with -H option, rsync does 1k write. without -H option, rsync does 4k write. ceph-fuse does not enable kernel writeba... Zheng Yan
02:26 AM Bug #22829 (Resolved): ceph-fuse: uses up all snap tags
got following crash during snap tests... Zheng Yan

01/29/2018

03:23 PM Bug #22788: ceph-fuse performance issues with rsync
It seems that the -H option causes low performance when destination is in cephfs. I still don't figure out why Zheng Yan
03:20 PM Bug #22754: mon: removing tier from an EC base pool is forbidden, even if allow_ec_overwrites is set
As far as I'm aware, nobody has worked on it, so that would be a no. John Spray
03:13 PM Bug #22754: mon: removing tier from an EC base pool is forbidden, even if allow_ec_overwrites is set
Is this going to make it into 12.2.3? David Turner
02:53 PM Feature #21995 (Fix Under Review): ceph-fuse: support nfs export
https://github.com/ceph/ceph/pull/20168 Jos Collin
02:41 PM Bug #22801: client: Client::flush_snaps still uses obsolete Client::user_id/group_id
CapSnap is for flushing snapshoted metadata (the metadata that were dirty at the time of mksnap), nothing do with set... Zheng Yan
02:27 PM Bug #22776 (Fix Under Review): mds: session count,dns and inos from cli "fs status" is always 0
Patrick Donnelly
02:26 PM Bug #22734 (Fix Under Review): cephfs-journal-tool: may got assertion failure due to not shutdown
Patrick Donnelly
12:59 PM Bug #22802: libcephfs: allow setting default perms
Part of the problem here is that there are really two sets of default permissions in this code. There is one in the C... Jeff Layton
03:48 AM Bug #22824 (Fix Under Review): Journaler::flush() may flush less data than expected, which causes...
https://github.com/ceph/ceph/pull/20155 Zheng Yan
03:02 AM Bug #22824 (Resolved): Journaler::flush() may flush less data than expected, which causes flush w...
Zheng Yan

01/28/2018

11:27 AM Bug #22821 (Fix Under Review): mds: session reference leak
https://github.com/ceph/ceph/pull/20148 Zheng Yan
08:15 AM Bug #22821 (Resolved): mds: session reference leak
there are several places get session by:
"Session *session = static_cast<Session *>(m->get_connection()->get_priv(...
Zheng Yan

01/26/2018

08:16 PM Bug #22802: libcephfs: allow setting default perms
What I think I'm going to do is just add a ceph_mount_perms_set() function to the API that will reset it to a UserPer... Jeff Layton
05:49 PM Bug #22802: libcephfs: allow setting default perms
The current default is to set it to -1, so that's probably what we'll do here.
Further down the rabbit hole, we ha...
Jeff Layton
05:36 PM Bug #22802: libcephfs: allow setting default perms
Jeff Layton wrote:
> Serious question: does anyone actually use the SyntheticClient? It's only linked into the ceph-...
Patrick Donnelly
04:52 PM Bug #22802: libcephfs: allow setting default perms
Serious question: does anyone actually use the SyntheticClient? It's only linked into the ceph-syn binary, and I don'... Jeff Layton
06:16 PM Bug #21091 (Resolved): StrayManager::truncate is broken
Nathan Cutler
06:16 PM Backport #21657 (Resolved): luminous: StrayManager::truncate is broken
Nathan Cutler
02:35 PM Feature #21156 (Fix Under Review): mds: speed up recovery with many open inodes
https://github.com/ceph/ceph/pull/20132 Zheng Yan
01:48 PM Bug #22801: client: Client::flush_snaps still uses obsolete Client::user_id/group_id
Hmm...cap_dirtier_uid is only set to anything non-default in _do_setattr(). That function does this very early:
<p...
Jeff Layton
01:43 PM Bug #22801: client: Client::flush_snaps still uses obsolete Client::user_id/group_id
I think we should cap_dirtier_uid/gid into CapSnap Zheng Yan
01:06 PM Bug #22801: client: Client::flush_snaps still uses obsolete Client::user_id/group_id
I don't have the best grasp of how snapshots work in cephfs, so I'm a little confused as to what should be done here:... Jeff Layton

01/25/2018

10:47 PM Bug #22802 (Resolved): libcephfs: allow setting default perms
-These options no longer work as advertised and only effect the SyntheticClient (with the exception of #22801). Best ... Patrick Donnelly
10:45 PM Bug #22801 (Resolved): client: Client::flush_snaps still uses obsolete Client::user_id/group_id
This appears to be a hold-out from the UserPerm work last year and the last remaining user of Client::user_id|user_gi... Patrick Donnelly
08:25 PM Bug #22219 (Resolved): mds: mds should ignore export_pin for deleted directory
Patrick Donnelly
08:25 PM Backport #22385 (Resolved): luminous: mds: mds should ignore export_pin for deleted directory
Patrick Donnelly
08:02 PM Backport #22385: luminous: mds: mds should ignore export_pin for deleted directory
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/19360
merged
Yuri Weinstein
08:25 PM Bug #22357 (Resolved): mds: read hang in multiple mds setup
Patrick Donnelly
08:24 PM Backport #22503 (Resolved): luminous: mds: read hang in multiple mds setup
Patrick Donnelly
08:02 PM Backport #22503: luminous: mds: read hang in multiple mds setup
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/19646
merged
Yuri Weinstein
08:24 PM Bug #21985 (Resolved): mds: definition of MDS_FEATURE_INCOMPAT_FILE_LAYOUT_V2 is wrong
Patrick Donnelly
08:24 PM Backport #22067 (Resolved): luminous: mds: definition of MDS_FEATURE_INCOMPAT_FILE_LAYOUT_V2 is w...
Patrick Donnelly
07:11 PM Backport #22067: luminous: mds: definition of MDS_FEATURE_INCOMPAT_FILE_LAYOUT_V2 is wrong
Patrick Donnelly wrote:
> https://github.com/ceph/ceph/pull/18782
merged
Yuri Weinstein
08:23 PM Bug #21975 (Resolved): MDS: mds gets significantly behind on trimming while creating millions of ...
Patrick Donnelly
08:23 PM Backport #22068 (Resolved): luminous: mds: mds gets significantly behind on trimming while creati...
Patrick Donnelly
07:09 PM Backport #22068: luminous: mds: mds gets significantly behind on trimming while creating millions...
Patrick Donnelly wrote:
> https://github.com/ceph/ceph/pull/18783
merged
Yuri Weinstein
08:22 PM Bug #22009 (Resolved): don't check gid when none specified in auth caps
Patrick Donnelly
08:22 PM Backport #22074 (Resolved): luminous: don't check gid when none specified in auth caps
Patrick Donnelly
07:07 PM Backport #22074: luminous: don't check gid when none specified in auth caps
Shinobu Kinjo wrote:
> https://github.com/ceph/ceph/pull/18835
merged
Yuri Weinstein
08:21 PM Bug #21722 (Resolved): mds: no assertion on inode being purging in find_ino_peers()
Patrick Donnelly
08:21 PM Backport #21952 (Resolved): luminous: mds: no assertion on inode being purging in find_ino_peers()
Patrick Donnelly
07:07 PM Backport #21952: luminous: mds: no assertion on inode being purging in find_ino_peers()
Shinobu Kinjo wrote:
> https://github.com/ceph/ceph/pull/18869
merged
Yuri Weinstein
08:21 PM Bug #21843 (Resolved): mds: preserve order of requests during recovery of multimds cluster
Patrick Donnelly
08:20 PM Backport #21947 (Resolved): luminous: mds: preserve order of requests during recovery of multimds...
Patrick Donnelly
07:06 PM Backport #21947: luminous: mds: preserve order of requests during recovery of multimds cluster
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/18871
merged
Yuri Weinstein
08:20 PM Bug #21928 (Resolved): src/mds/MDCache.cc: 7421: FAILED assert(CInode::count() == inode_map.size(...
Patrick Donnelly
08:19 PM Backport #22077 (Resolved): luminous: src/mds/MDCache.cc: 7421: FAILED assert(CInode::count() == ...
Patrick Donnelly
07:05 PM Backport #22077: luminous: src/mds/MDCache.cc: 7421: FAILED assert(CInode::count() == inode_map.s...
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/18912
merged
Yuri Weinstein
08:19 PM Bug #21959 (Resolved): MDSMonitor: monitor gives constant "is now active in filesystem cephfs as ...
Patrick Donnelly
08:19 PM Backport #22192 (Resolved): luminous: MDSMonitor: monitor gives constant "is now active in filesy...
Patrick Donnelly
07:05 PM Backport #22192: luminous: MDSMonitor: monitor gives constant "is now active in filesystem cephfs...
Shinobu Kinjo wrote:
> https://github.com/ceph/ceph/pull/19055
merged
Yuri Weinstein
08:18 PM Backport #22379 (Resolved): luminous: client reconnect gather race
Patrick Donnelly
08:16 PM Feature #19578 (Resolved): mds: optimize CDir::_omap_commit() and CDir::_committed() for large di...
Patrick Donnelly
08:16 PM Backport #22563 (Resolved): luminous: mds: optimize CDir::_omap_commit() and CDir::_committed() f...
Patrick Donnelly
06:58 PM Backport #22563: luminous: mds: optimize CDir::_omap_commit() and CDir::_committed() for large di...
Zheng Yan wrote:
> https://github.com/ceph/ceph/pull/19775
merged
Yuri Weinstein
08:16 PM Bug #21853 (Resolved): mds: mdsload debug too high
Patrick Donnelly
08:15 PM Backport #22587 (Resolved): luminous: mds: mdsload debug too high
Patrick Donnelly
06:58 PM Backport #22587: luminous: mds: mdsload debug too high
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/19827
merged
Yuri Weinstein
08:15 PM Backport #22763 (Resolved): luminous: mds: crashes because of old pool id in journal header
Patrick Donnelly
06:56 PM Backport #22763: luminous: mds: crashes because of old pool id in journal header
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/20085
merged
Yuri Weinstein
08:15 PM Bug #22629 (Resolved): client: avoid recursive lock in ll_get_vino
Patrick Donnelly
08:14 PM Backport #22765 (Resolved): luminous: client: avoid recursive lock in ll_get_vino
Patrick Donnelly
06:55 PM Backport #22765: luminous: client: avoid recursive lock in ll_get_vino
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/20086
merged
Yuri Weinstein
08:13 PM Bug #22157 (Resolved): client: trim_caps may remove cap iterator points to
Patrick Donnelly
08:04 PM Bug #22157: client: trim_caps may remove cap iterator points to
merged https://github.com/ceph/ceph/pull/19105 Yuri Weinstein
08:13 PM Backport #22228 (Resolved): luminous: client: trim_caps may remove cap iterator points to
Patrick Donnelly
07:46 PM Backport #22004 (Resolved): luminous: FAILED assert(get_version() < pv) in CDir::mark_dirty
Nathan Cutler
07:12 PM Backport #22004: luminous: FAILED assert(get_version() < pv) in CDir::mark_dirty
Kefu Chai wrote:
> https://github.com/ceph/ceph/pull/18008
merged
Yuri Weinstein
07:45 PM Backport #22078 (Resolved): luminous: ceph.in: tell mds does not understand --cluster
Nathan Cutler
07:08 PM Backport #22078: luminous: ceph.in: tell mds does not understand --cluster
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/18831
merged
Yuri Weinstein
07:45 PM Bug #21967 (Resolved): 'ceph tell mds' commands result in 'File exists' errors on client admin so...
Nathan Cutler
07:45 PM Backport #22076 (Resolved): luminous: 'ceph tell mds' commands result in 'File exists' errors on ...
Nathan Cutler
07:08 PM Backport #22076: luminous: 'ceph tell mds' commands result in 'File exists' errors on client admi...
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/18831
merged
Yuri Weinstein
07:11 PM Feature #18490 (Resolved): client: implement delegation support in userland cephfs
Nathan Cutler
07:11 PM Backport #22407 (Resolved): luminous: client: implement delegation support in userland cephfs
Nathan Cutler
07:00 PM Backport #22407: luminous: client: implement delegation support in userland cephfs
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/19480
merged
Yuri Weinstein
07:11 PM Bug #21512 (Resolved): qa: libcephfs_interface_tests: shutdown race failures
Nathan Cutler
07:11 PM Backport #21874 (Resolved): luminous: qa: libcephfs_interface_tests: shutdown race failures
Nathan Cutler
06:57 PM Backport #21874: luminous: qa: libcephfs_interface_tests: shutdown race failures
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/20082
merged
Yuri Weinstein
07:10 PM Backport #21525 (Resolved): luminous: client: dual client segfault with racing ceph_shutdown
Nathan Cutler
06:57 PM Backport #21525: luminous: client: dual client segfault with racing ceph_shutdown
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/20082
merged
Yuri Weinstein
07:03 PM Bug #22263: client reconnect gather race
Zheng Yan wrote:
> https://github.com/ceph/ceph/pull/19326
merged
Yuri Weinstein
05:35 PM Backport #22798 (Resolved): luminous: mds: add success return
No upstream tracker issue for this. It was fixed upstream in https://github.com/ceph/ceph/pull/16778/commits/f519fca9... Patrick Donnelly
05:11 PM Backport #21657: luminous: StrayManager::truncate is broken
Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/18019
merged
Yuri Weinstein
05:44 AM Bug #22741 (Fix Under Review): osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-...
https://github.com/ceph/ceph/pull/20113 Zheng Yan
04:45 AM Backport #22764 (In Progress): jewel: mds: crashes because of old pool id in journal header
https://github.com/ceph/ceph/pull/20111 Prashant D

01/24/2018

04:13 PM Bug #22788 (Won't Fix): ceph-fuse performance issues with rsync
Hi,
I have a performance issue when running rsync on a FUSE-mounted CephFS.
dd runs on "line speed" on my test ...
Robert Sander
02:02 PM Bug #22249: Need to restart MDS to release cephfs space
The logs show that the client held caps on stray inodes. which is root cause of the issue.
Did you try -client_try...
Zheng Yan
09:08 AM Bug #22249: Need to restart MDS to release cephfs space
Zheng Yan wrote:
> junming rao wrote:
> > Zheng Yan wrote:
> > > please try remounting all cephfs with ceph-fuse o...
junming rao
09:31 AM Bug #22741: osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-yuri-master-1.19.18...
... Zheng Yan
02:44 AM Backport #22765 (In Progress): luminous: client: avoid recursive lock in ll_get_vino
https://github.com/ceph/ceph/pull/20086 Prashant D
01:48 AM Backport #22763 (In Progress): luminous: mds: crashes because of old pool id in journal header
https://github.com/ceph/ceph/pull/20085 Prashant D

01/23/2018

06:06 PM Bug #21393 (Resolved): MDSMonitor: inconsistent role/who usage in command help
Patrick Donnelly
06:05 PM Backport #22508 (Closed): luminous: MDSMonitor: inconsistent role/who usage in command help
Yes, let's forgo the luminous backport. Thanks for pointing that out Nathan! Patrick Donnelly
12:41 PM Bug #22776: mds: session count,dns and inos from cli "fs status" is always 0
*PR*: https://github.com/ceph/ceph/pull/20079 shangzhong zhu
12:07 PM Bug #22776 (Resolved): mds: session count,dns and inos from cli "fs status" is always 0
... shangzhong zhu
09:48 AM Backport #22762 (In Progress): jewel: mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
Nathan Cutler
09:40 AM Backport #22762 (Resolved): jewel: mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
https://github.com/ceph/ceph/pull/20067 Nathan Cutler
09:40 AM Backport #22765 (Resolved): luminous: client: avoid recursive lock in ll_get_vino
https://github.com/ceph/ceph/pull/20086 Nathan Cutler
09:40 AM Backport #22764 (Resolved): jewel: mds: crashes because of old pool id in journal header
https://github.com/ceph/ceph/pull/20111 Nathan Cutler
09:40 AM Backport #22763 (Resolved): luminous: mds: crashes because of old pool id in journal header
https://github.com/ceph/ceph/pull/20085 Nathan Cutler

01/22/2018

10:11 PM Bug #22754 (Resolved): mon: removing tier from an EC base pool is forbidden, even if allow_ec_ove...
OSDMonitor::_check_remove_tier needs to be made aware that this should be permitted if the base tier is suitable for ... John Spray
08:06 PM Bug #22741: osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-yuri-master-1.19.18...
Core: /ceph/teuthology-archive/yuriw-2018-01-19_18:23:03-powercycle-wip-yuri-master-1.19.18-distro-basic-smithi/20909... Patrick Donnelly
02:14 PM Bug #22741: osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-yuri-master-1.19.18...
Assigned to CephFS because it's crashing in the ceph-fuse process (in the absence of a better home for ObjectCacher i... John Spray
03:03 PM Feature #12107 (In Progress): mds: use versioned wire protocol; obviate CEPH_MDS_PROTOCOL
Patrick Donnelly
07:28 AM Feature #12107: mds: use versioned wire protocol; obviate CEPH_MDS_PROTOCOL
I'm working on this, please assign this to me dongdong tao
12:05 PM Backport #22508 (Need More Info): luminous: MDSMonitor: inconsistent role/who usage in command help
Non-trivial backport - since it's essentially a documentation fix, I'm not sure if it's worth the risk. Nathan Cutler
11:18 AM Backport #22078 (In Progress): luminous: ceph.in: tell mds does not understand --cluster
Nathan Cutler

01/20/2018

05:33 PM Bug #22741 (Resolved): osdc: "FAILED assert(bh->last_write_tid > tid)" in powercycle-wip-yuri-mas...
Run: http://pulpito.ceph.com/yuriw-2018-01-19_18:23:03-powercycle-wip-yuri-master-1.19.18-distro-basic-smithi/
Jobs:...
Yuri Weinstein

01/19/2018

04:31 AM Bug #22734: cephfs-journal-tool: may got assertion failure due to not shutdown
https://github.com/ceph/ceph/pull/19991 dongdong tao
04:22 AM Bug #22734 (Resolved): cephfs-journal-tool: may got assertion failure due to not shutdown
```
2018-01-14T19:36:56.381 INFO:teuthology.orchestra.run.smithi139.stderr:Error loading journal: (2) No such file o...
dongdong tao

01/18/2018

11:02 PM Bug #22733 (Duplicate): ceph-fuse: failed assert in xlist<T>::iterator& xlist<T>::iterator::opera...
Thanks for the report anyway! Patrick Donnelly
10:09 PM Bug #22733 (Duplicate): ceph-fuse: failed assert in xlist<T>::iterator& xlist<T>::iterator::opera...
We had several ceph-fuse crashes with errors like... Andras Pataki
08:02 PM Bug #22730 (Fix Under Review): mds: scrub crash
https://github.com/ceph/ceph/pull/20012 Patrick Donnelly
05:38 PM Bug #22730: mds: scrub crash
Doug, please take a look at this one. Patrick Donnelly
04:17 PM Bug #22730 (Resolved): mds: scrub crash
this crash can be reproduced by 2 steps
1 ceph daemon mds.a scrub_path <dir> recursive
2 ceph daemon mds.a scrub_...
dongdong tao
12:43 AM Backport #22700 (In Progress): jewel: client:_rmdir() uses a deleted memory structure(Dentry) lea...
https://github.com/ceph/ceph/pull/19993 Prashant D
12:27 AM Backport #22700: jewel: client:_rmdir() uses a deleted memory structure(Dentry) leading a core
I'm on it. Prashant D

01/17/2018

10:07 PM Bug #22683 (Fix Under Review): client: coredump when nfs-ganesha use ceph_ll_get_inode()
Patrick Donnelly
03:34 PM Feature #4208: Add more replication pool tests for Hadoop / Ceph bindings
Bulk move of hadoop category into FS project. John Spray
03:34 PM Feature #4361: Setup another gitbuilder VM for building external Hadoop git repo(s)
Bulk move of hadoop category into FS project. John Spray
03:34 PM Bug #3544: ./configure checks CFLAGS for jni.h if --with-hadoop is specified but also needs to ch...
Bulk move of hadoop category into FS project. John Spray
03:34 PM Bug #1661: Hadoop: expected system directories not present
Bulk move of hadoop category into FS project. John Spray
03:34 PM Bug #1663: Hadoop: file ownership/permission not available in hadoop
Bulk move of hadoop category into FS project. John Spray
03:26 PM Bug #21748 (Can't reproduce): client assertions tripped during some workloads
No response in several months, and I've never seen this trip in my own testing. Closing for now. Please reopen if you... Jeff Layton
03:24 PM Bug #22003 (Resolved): [CephFS-Ganesha]MDS migrate will affect Ganesha service?
No response in two months. Closing bug.
Please reopen or comment if you've been able to test with that patch and i...
Jeff Layton
03:23 PM Bug #21419 (Rejected): client: is ceph_caps_for_mode correct for r/o opens?
Ok, I think you're right. may_open happens at a higher level and we will simply request the caps at that point. False... Jeff Layton
10:50 AM Bug #21734: mount client shows total capacity of cluster but not of a pool
(Just moving this closed ticket because I'm deleting the bogus "cephfs" category in the toplevel Ceph project) John Spray
07:05 AM Backport #22719 (In Progress): luminous: mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
https://github.com/ceph/ceph/pull/19982 Zheng Yan
06:57 AM Backport #22719 (Resolved): luminous: mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
https://github.com/ceph/ceph/pull/19982 Zheng Yan
05:43 AM Backport #22590 (In Progress): jewel: ceph.in: tell mds does not understand --cluster
Prashant D
04:12 AM Bug #22629 (Pending Backport): client: avoid recursive lock in ll_get_vino
Patrick Donnelly
04:12 AM Bug #22631 (Pending Backport): mds: crashes because of old pool id in journal header
Patrick Donnelly
04:11 AM Backport #22690 (In Progress): luminous: qa: qa/cephfs/clusters/fixed-2-ucephfs.yaml has insuffic...
https://github.com/ceph/ceph/pull/19976 Prashant D
04:10 AM Bug #22647 (Pending Backport): mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
Patrick Donnelly
02:38 AM Backport #22689 (In Progress): jewel: client: fails to release to revoking Fc
Prashant D
02:38 AM Backport #22689: jewel: client: fails to release to revoking Fc
https://github.com/ceph/ceph/pull/19975 Prashant D

01/16/2018

07:28 PM Bug #22428: mds: don't report slow request for blocked filelock request
Here's a recent example from someone in #ceph:... John Spray
02:13 PM Backport #22688 (In Progress): luminous: client: fails to release to revoking Fc
https://github.com/ceph/ceph/pull/19970 Zheng Yan
08:16 AM Backport #22688 (Resolved): luminous: client: fails to release to revoking Fc
https://github.com/ceph/ceph/pull/20342 Nathan Cutler
01:57 PM Backport #22699 (In Progress): luminous: client:_rmdir() uses a deleted memory structure(Dentry) ...
Zheng Yan
01:57 PM Backport #22699 (Fix Under Review): luminous: client:_rmdir() uses a deleted memory structure(Den...
https://github.com/ceph/ceph/pull/19968 Zheng Yan
08:17 AM Backport #22699 (Resolved): luminous: client:_rmdir() uses a deleted memory structure(Dentry) lea...
https://github.com/ceph/ceph/pull/19968 Nathan Cutler
08:34 AM Backport #22579 (In Progress): luminous: mds: check for CEPH_OSDMAP_FULL is now wrong; cluster fu...
Nathan Cutler
08:31 AM Backport #22580 (In Progress): luminous: qa: full flag not set on osdmap for tasks.cephfs.test_fu...
Nathan Cutler
08:23 AM Backport #22695 (In Progress): jewel: mds: fix dump last_sent
Nathan Cutler
08:17 AM Backport #22695 (Resolved): jewel: mds: fix dump last_sent
https://github.com/ceph/ceph/pull/19961 Nathan Cutler
08:22 AM Backport #22694 (In Progress): luminous: mds: fix dump last_sent
Nathan Cutler
08:17 AM Backport #22694 (Resolved): luminous: mds: fix dump last_sent
https://github.com/ceph/ceph/pull/19959 Nathan Cutler
08:17 AM Backport #22700 (Resolved): jewel: client:_rmdir() uses a deleted memory structure(Dentry) leadin...
https://github.com/ceph/ceph/pull/19993 Nathan Cutler
08:17 AM Backport #22697 (Rejected): jewel: client: dirty caps may never get the chance to flush
Nathan Cutler
08:17 AM Backport #22696 (Resolved): luminous: client: dirty caps may never get the chance to flush
https://github.com/ceph/ceph/pull/21278 Nathan Cutler
08:16 AM Backport #22690 (Resolved): luminous: qa: qa/cephfs/clusters/fixed-2-ucephfs.yaml has insufficien...
https://github.com/ceph/ceph/pull/19976 Nathan Cutler
08:16 AM Backport #22689 (Resolved): jewel: client: fails to release to revoking Fc
https://github.com/ceph/ceph/pull/19975 Nathan Cutler
06:38 AM Bug #22683: client: coredump when nfs-ganesha use ceph_ll_get_inode()
https://github.com/ceph/ceph/pull/19957 huanwen ren
02:47 AM Bug #22683 (Resolved): client: coredump when nfs-ganesha use ceph_ll_get_inode()
Environment:
nfs : nfs-ganehsa2.5.4 + https://github.com/nfs-ganesha/nfs-ganesha/commit/476c2068bd4a3fd22f0d...
huanwen ren

01/15/2018

02:36 PM Bug #22610 (Fix Under Review): MDS: assert failure when the inode for the cap_export from other M...
Zheng Yan

01/13/2018

01:43 AM Bug #21402 (In Progress): mds: move remaining containers in CDentry/CDir/CInode to mempool
Patrick Donnelly

01/12/2018

10:42 PM Bug #22652 (Pending Backport): client: fails to release to revoking Fc
Patrick Donnelly
10:39 PM Bug #22646 (Pending Backport): qa: qa/cephfs/clusters/fixed-2-ucephfs.yaml has insufficient osds
Patrick Donnelly
03:49 PM Feature #21995 (In Progress): ceph-fuse: support nfs export
Jos Collin
11:07 AM Feature #21156 (In Progress): mds: speed up recovery with many open inodes
Zheng Yan

01/11/2018

10:50 PM Backport #22508: luminous: MDSMonitor: inconsistent role/who usage in command help
See also: https://github.com/ceph/ceph/pull/19926 Patrick Donnelly
10:29 PM Bug #21393: MDSMonitor: inconsistent role/who usage in command help
The fix for this causes upgrade tests to fail: http://tracker.ceph.com/issues/22527#note-9
We will probably need t...
Patrick Donnelly
08:39 AM Bug #22652 (Fix Under Review): client: fails to release to revoking Fc
https://github.com/ceph/ceph/pull/19920 Zheng Yan
08:37 AM Bug #22652: client: fails to release to revoking Fc
hang fuse_reverse_inval_inode() was caused by hang page writeback. Zheng Yan

01/10/2018

11:24 PM Backport #22590: jewel: ceph.in: tell mds does not understand --cluster
https://github.com/ceph/ceph/pull/19907 Prashant D
10:44 PM Backport #22590: jewel: ceph.in: tell mds does not understand --cluster
I'm on it. Prashant D
04:41 PM Bug #22631 (Fix Under Review): mds: crashes because of old pool id in journal header
Jos Collin
03:41 PM Backport #22076 (In Progress): luminous: 'ceph tell mds' commands result in 'File exists' errors ...
Nathan Cutler
03:17 PM Backport #22076 (Fix Under Review): luminous: 'ceph tell mds' commands result in 'File exists' er...
Jos Collin
02:45 PM Bug #22652: client: fails to release to revoking Fc
Sage Weil
01:29 PM Bug #22652: client: fails to release to revoking Fc
I reproduced it locally. it seems like kernel issue. The issue happens only when fuse_use_invalidate_cb is true. Zheng Yan
11:02 AM Bug #22652 (Resolved): client: fails to release to revoking Fc
http://pulpito.ceph.com/pdonnell-2018-01-09_21:14:38-multimds-wip-pdonnell-testing-20180109.193634-testing-basic-smit... Zheng Yan
05:54 AM Bug #22647 (Fix Under Review): mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
https://github.com/ceph/ceph/pull/19891 Zheng Yan
02:34 AM Bug #22647 (Resolved): mds: unbalanced auth_pin/auth_unpin in RecoveryQueue code
... Zheng Yan
01:08 AM Bug #22629 (Fix Under Review): client: avoid recursive lock in ll_get_vino
Patrick Donnelly
01:05 AM Bug #22562 (Pending Backport): mds: fix dump last_sent
Patrick Donnelly
01:05 AM Bug #22546 (Pending Backport): client: dirty caps may never get the chance to flush
Patrick Donnelly
01:04 AM Bug #22536 (Pending Backport): client:_rmdir() uses a deleted memory structure(Dentry) leading a ...
Patrick Donnelly
12:44 AM Bug #22646: qa: qa/cephfs/clusters/fixed-2-ucephfs.yaml has insufficient osds
https://github.com/ceph/ceph/pull/19885 Patrick Donnelly
12:40 AM Bug #22646 (Resolved): qa: qa/cephfs/clusters/fixed-2-ucephfs.yaml has insufficient osds
This causes startup to fail for ec pool configurations.
(This was included in my fix for #22627 but I'm breaking i...
Patrick Donnelly

01/09/2018

04:03 PM Bug #22631: mds: crashes because of old pool id in journal header
https://github.com/ceph/ceph/pull/19860 dongdong tao
08:38 AM Bug #22631: mds: crashes because of old pool id in journal header
through the code, we found it is because of the old pool id in the journal header.
my solution is
add "set pool_id"...
dongdong tao
08:35 AM Bug #22631 (Resolved): mds: crashes because of old pool id in journal header
we have use rados cppool command to copy the cephfs metadata pool
but,after copy done, mds would keep crashing when ...
dongdong tao
02:53 PM Backport #21948 (In Progress): luminous: MDSMonitor: mons should reject misconfigured mds_blackli...
Nathan Cutler
02:43 PM Backport #21874 (In Progress): luminous: qa: libcephfs_interface_tests: shutdown race failures
Nathan Cutler
02:43 PM Backport #21870 (In Progress): luminous: Assertion in EImportStart::replay should be a damaged()
Nathan Cutler
01:02 PM Feature #22545 (Fix Under Review): add dump inode command to mds
Nathan Cutler
09:18 AM Backport #22630 (Resolved): doc: misc fixes for CephFS best practices
This is a backport of: https://github.com/ceph/ceph/pull/19791 Jos Collin
08:13 AM Backport #22630 (Resolved): doc: misc fixes for CephFS best practices
https://github.com/ceph/ceph/pull/19858 Jos Collin
07:47 AM Bug #22629: client: avoid recursive lock in ll_get_vino
https://github.com/ceph/ceph/pull/19837 dongdong tao
07:47 AM Bug #22629 (Resolved): client: avoid recursive lock in ll_get_vino
ll_get_vino would lock the client_lock.
the caller must not have it.
dongdong tao
04:54 AM Bug #21991 (Resolved): mds: tell session ls returns vanila EINVAL when MDS is not active
Jos Collin
04:19 AM Bug #22627 (Fix Under Review): qa: kcephfs lacks many configurations in the fs/multimds suites
https://github.com/ceph/ceph/pull/19856 Patrick Donnelly
04:17 AM Bug #22627 (Resolved): qa: kcephfs lacks many configurations in the fs/multimds suites
In particular:
o Not using the common overrides/
o Not using 8 OSDs for EC configurations
o Not using openstack ...
Patrick Donnelly
03:48 AM Bug #22626 (Rejected): mds: sessionmap version mismatch when replay esessions
Zhang, we are not accepting bugs for multimds clusters on jewel. You can still seek help/advice on ceph-users if you ... Patrick Donnelly
03:21 AM Bug #22626 (Rejected): mds: sessionmap version mismatch when replay esessions
We used ceph 10.2.10 and backported this PR: https://github.com/ceph/ceph/commit/a49726e10ef23be124d92872470fd258a193... Zhi Zhang
03:46 AM Bug #22551: client: should flush dirty caps on backgroud
that's what i'm concerned about, maybe it's not been flushed periodically, it should be easy to verify, will do it dongdong tao
03:43 AM Bug #22551: client: should flush dirty caps on backgroud
Dirty metadata should be flushed when the cap is released. It may also happen periodically (I'm not certain). Patrick Donnelly
01:59 AM Bug #22551: client: should flush dirty caps on backgroud
i will write a case to verify it. dongdong tao
01:49 AM Bug #22551: client: should flush dirty caps on backgroud
i'm not sure if i'm right, if there is only one client and opened a file write some data and did not close it. i know... dongdong tao
01:09 AM Bug #22607 (Rejected): client: should delete cap in remove_cap
The cap is deleted via "in->caps.erase(mds)". The session xlist entry is deleted in the Cap destructor. Patrick Donnelly

01/08/2018

10:28 PM Documentation #22599 (Resolved): doc: mds memory tracking of cache is imprecise by a constant factor
Patrick Donnelly
05:23 PM Backport #22392 (Resolved): luminous: mds: tell session ls returns vanila EINVAL when MDS is not ...
Jos Collin
02:45 PM Bug #22610 (In Progress): MDS: assert failure when the inode for the cap_export from other MDS ha...
Patrick Donnelly
08:04 AM Bug #22610: MDS: assert failure when the inode for the cap_export from other MDS happened not in ...
Fire a pull request: https://github.com/ceph/ceph/pull/19836 Jianyu Li
07:57 AM Bug #22610 (Resolved): MDS: assert failure when the inode for the cap_export from other MDS happe...
We use two active MDS in our online environment, recently mds.1 restarted and during its rejoin phase, mds.0 met asse... Jianyu Li
02:43 PM Bug #22551 (Need More Info): client: should flush dirty caps on backgroud
Dongdong, can you explain more what the problem is? Do you have an issue you've observed? Patrick Donnelly
02:40 PM Bug #21419: client: is ceph_caps_for_mode correct for r/o opens?
No, I've not had time to look at it. For now, I'll just mark this as low priority until I can revisit ir. Jeff Layton
01:27 PM Backport #22569: jewel: doc: clarify path restriction instructions
Added follow-on cherry-pick https://github.com/ceph/ceph/pull/19840 Nathan Cutler
11:59 AM Backport #22569: jewel: doc: clarify path restriction instructions
Commit 85ac1cd which was a cherry-pick of d1277f1 fixing tracker issue http://tracker.ceph.com/issues/16906 introduce... Jos Collin
11:16 AM Backport #22569 (Resolved): jewel: doc: clarify path restriction instructions
Nathan Cutler
05:16 AM Backport #22569 (In Progress): jewel: doc: clarify path restriction instructions
Jos Collin
04:24 AM Backport #22569 (Resolved): jewel: doc: clarify path restriction instructions
Jos Collin
11:16 AM Documentation #16906 (Resolved): doc: clarify path restriction instructions
Nathan Cutler
04:31 AM Backport #22587 (In Progress): luminous: mds: mdsload debug too high
Prashant D
03:32 AM Backport #22587 (Need More Info): luminous: mds: mdsload debug too high
https://github.com/ceph/ceph/pull/19827 Prashant D
04:12 AM Backport #22579: luminous: mds: check for CEPH_OSDMAP_FULL is now wrong; cluster full flag is obs...
https://github.com/ceph/ceph/pull/19830 Shinobu Kinjo
04:12 AM Backport #22579: luminous: mds: check for CEPH_OSDMAP_FULL is now wrong; cluster full flag is obs...
Shinobu Kinjo wrote:
-> fix already in luminous
-
Shinobu Kinjo
03:58 AM Backport #22579: luminous: mds: check for CEPH_OSDMAP_FULL is now wrong; cluster full flag is obs...
fix already in luminous Shinobu Kinjo
03:37 AM Backport #22573 (In Progress): luminous: AttributeError: 'LocalFilesystem' object has no attribut...
https://github.com/ceph/ceph/pull/19829 Prashant D

01/07/2018

04:48 AM Bug #22607: client: should delete cap in remove_cap
https://github.com/ceph/ceph/pull/19782 dongdong tao
04:48 AM Bug #22607 (Rejected): client: should delete cap in remove_cap
I think the cap should be deleted.
so that the cap can be removed from session->caps
dongdong tao

01/05/2018

09:40 PM Bug #22051 (Need More Info): tests: Health check failed: Reduced data availability: 5 pgs peering...
Patrick Donnelly
09:37 PM Bug #21575 (Resolved): mds: client caps can go below hard-coded default (100)
Patrick Donnelly
09:34 PM Feature #20752 (Resolved): cap message flag which indicates if client still has pending capsnap
Patrick Donnelly
09:32 PM Bug #21419 (Need More Info): client: is ceph_caps_for_mode correct for r/o opens?
Jeff, any update on this? Patrick Donnelly
09:30 PM Documentation #21172: doc: Export over NFS
Ramana, any update on this? Patrick Donnelly
07:25 PM Documentation #22599 (Fix Under Review): doc: mds memory tracking of cache is imprecise by a cons...
https://github.com/ceph/ceph/pull/19807 Patrick Donnelly
07:19 PM Documentation #22599 (In Progress): doc: mds memory tracking of cache is imprecise by a constant ...
Patrick Donnelly
07:19 PM Documentation #22599 (Resolved): doc: mds memory tracking of cache is imprecise by a constant factor
MDS currently can use up much more memory than its mds_cache_memory_limit. This is more noticable in deployments of a... Patrick Donnelly
06:44 PM Bug #22548 (Need More Info): mds: crash during recovery
Patrick Donnelly
05:09 PM Bug #21539 (Resolved): man: missing man page for mount.fuse.ceph
Jos Collin
02:51 PM Bug #21539: man: missing man page for mount.fuse.ceph
follow-on fix: https://github.com/ceph/ceph/pull/19792 Nathan Cutler
05:09 PM Backport #22398 (Resolved): luminous: man: missing man page for mount.fuse.ceph
Jos Collin
04:08 PM Documentation #2206 (Resolved): Need a control command to gracefully shutdown an active MDS prior...
Sage Weil
03:02 PM Bug #22595 (Resolved): doc: mount.fuse.ceph is missing in index.rst
Nathan Cutler
02:56 PM Bug #22595 (Closed): doc: mount.fuse.ceph is missing in index.rst
Luminous backport handled via #21539 Nathan Cutler
01:57 PM Bug #22595 (Fix Under Review): doc: mount.fuse.ceph is missing in index.rst
Jos Collin
01:57 PM Bug #22595: doc: mount.fuse.ceph is missing in index.rst
https://github.com/ceph/ceph/pull/19792 Jos Collin
01:56 PM Bug #22595 (Resolved): doc: mount.fuse.ceph is missing in index.rst
mount.fuse.ceph is missing in http://docs.ceph.com/docs/master/cephfs/ Jos Collin
12:19 PM Backport #22590 (Resolved): jewel: ceph.in: tell mds does not understand --cluster
https://github.com/ceph/ceph/pull/19907 Nathan Cutler
12:18 PM Backport #22587 (Resolved): luminous: mds: mdsload debug too high
https://github.com/ceph/ceph/pull/19827 Nathan Cutler
12:17 PM Backport #22563 (In Progress): luminous: mds: optimize CDir::_omap_commit() and CDir::_committed(...
Nathan Cutler
12:17 PM Backport #22564 (In Progress): luminous: Locker::calc_new_max_size does not take layout.stripe_co...
Nathan Cutler
12:16 PM Backport #22580 (Resolved): luminous: qa: full flag not set on osdmap for tasks.cephfs.test_full....
https://github.com/ceph/ceph/pull/19962 Nathan Cutler
12:16 PM Backport #22579 (Resolved): luminous: mds: check for CEPH_OSDMAP_FULL is now wrong; cluster full ...
https://github.com/ceph/ceph/pull/19830 Nathan Cutler
12:16 PM Backport #22573 (Resolved): luminous: AttributeError: 'LocalFilesystem' object has no attribute '...
https://github.com/ceph/ceph/pull/19829 Nathan Cutler
10:10 AM Backport #22569 (Fix Under Review): jewel: doc: clarify path restriction instructions
https://github.com/ceph/ceph/pull/19795 Jos Collin
09:39 AM Backport #22569 (Resolved): jewel: doc: clarify path restriction instructions
https://github.com/ceph/ceph/pull/19795 and https://github.com/ceph/ceph/pull/19840 Jos Collin
09:39 AM Documentation #16906 (Pending Backport): doc: clarify path restriction instructions
Jos Collin
12:42 AM Bug #22483 (Pending Backport): mds: check for CEPH_OSDMAP_FULL is now wrong; cluster full flag is...
https://github.com/ceph/ceph/pull/19602 Patrick Donnelly
12:40 AM Bug #22475 (Pending Backport): qa: full flag not set on osdmap for tasks.cephfs.test_full.TestClu...
Patrick Donnelly

01/04/2018

07:32 PM Bug #22562 (Fix Under Review): mds: fix dump last_sent
Patrick Donnelly
03:57 AM Bug #22562: mds: fix dump last_sent
https://github.com/ceph/ceph/pull/19762 dongdong tao
03:57 AM Bug #22562 (Resolved): mds: fix dump last_sent
last_sent in capability is an integer dongdong tao
07:15 AM Backport #22564 (Resolved): luminous: Locker::calc_new_max_size does not take layout.stripe_count...
https://github.com/ceph/ceph/pull/19776 Zheng Yan
07:10 AM Backport #22563 (Resolved): luminous: mds: optimize CDir::_omap_commit() and CDir::_committed() f...
https://github.com/ceph/ceph/pull/19775 Zheng Yan
03:46 AM Bug #22542 (Resolved): doc: epoch barrier mechanism not found
Jos Collin
03:26 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
can't find any 'osd_op ... write' in mds logs. So I can't find any clue how the corruption happened. Zheng Yan
01:48 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
Zheng Yan wrote:
> can't any log for "2017-12-16". next time you do experiment,please set debug_ms=1 for mds
Dear...
鹏 张

01/03/2018

06:11 PM Bug #22536 (Fix Under Review): client:_rmdir() uses a deleted memory structure(Dentry) leading a ...
Patrick Donnelly
05:40 PM Bug #22546 (Fix Under Review): client: dirty caps may never get the chance to flush
Patrick Donnelly
02:42 PM Feature #16775 (Fix Under Review): MDS command for listing open files
https://github.com/ceph/ceph/pull/19760 John Spray
01:04 PM Feature #16775: MDS command for listing open files
could you please have a look at this pr
https://github.com/ceph/ceph/pull/19760
dongdong tao
02:04 PM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
can't any log for "2017-12-16". next time you do experiment,please set debug_ms=1 for mds Zheng Yan
10:10 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
Zheng Yan wrote:
> please upload ceph cluster log. So I can check timestamp of mds failovers
Dear zheng:
I ha...
鹏 张
03:45 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
please upload ceph cluster log. So I can check timestamp of mds failovers Zheng Yan
04:00 AM Bug #22547: active mds session miss for client
Zheng Yan wrote:
> Sorry. the while the process is:
>
> mds close client connection
> client's remote_reset call...
dongdong tao
02:51 AM Bug #22547: active mds session miss for client
Sorry. the while the process is:
mds close client connection
client's remote_reset callback gets called
client s...
Zheng Yan

01/02/2018

03:40 PM Bug #22547: active mds session miss for client
Zheng Yan wrote:
> dongdong tao wrote:
> > zheng, if a client has been evicted by mds, the client should still thin...
dongdong tao
01:47 AM Bug #22547: active mds session miss for client
dongdong tao wrote:
> zheng, if a client has been evicted by mds, the client should still think the connection is av...
Zheng Yan
03:17 PM Backport #22552 (Resolved): luminous: doc: epoch barrier mechanism not found
Sage Weil
11:34 AM Backport #22552 (Fix Under Review): luminous: doc: epoch barrier mechanism not found
Jos Collin
10:57 AM Backport #22552: luminous: doc: epoch barrier mechanism not found
https://github.com/ceph/ceph/pull/19741 Jos Collin
10:43 AM Backport #22552 (Resolved): luminous: doc: epoch barrier mechanism not found
Jos Collin
11:14 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
Jos Collin wrote:
> I don't see anything in the URLs provided. Additionally, this looks like a Support Case.
can ...
Yong Wang
11:10 AM Bug #22523: Jewel10.2.10 cephfs journal corrupt,later event jump into previous position.
wangyong wang wrote:
> Hi all.
> ==============================
> version: jewel 10.2.10 (professional rpms)
> no...
Yong Wang

01/01/2018

11:56 AM Bug #22547 (Need More Info): active mds session miss for client
Jos Collin
06:47 AM Bug #22542 (Pending Backport): doc: epoch barrier mechanism not found
Jos Collin
 

Also available in: Atom