Project

General

Profile

Activity

From 02/24/2015 to 03/25/2015

03/25/2015

10:39 AM Bug #5488: librbd: deadlock in image refresh
* *invalid and closed* firefly backport https://github.com/ceph/ceph/pull/4176 Loïc Dachary

03/24/2015

08:44 PM Bug #6299 (Resolved): Dumpling Creates Extra Log Files
this isn't important enough to backport to dumpling at this point Josh Durgin
08:19 PM Bug #11075 (Rejected): librbd: fsx failure - read zeroes when data was expected
the branch that caused this isn't worth merging Josh Durgin
08:18 PM Bug #11179 (Won't Fix): "rbd ls" hang forever when cluster is damaged
The best way to script around things like this is to use the timeout command, e.g. "timeout 30 rbd ls". There are "ra... Josh Durgin

03/23/2015

02:26 PM Bug #11179: "rbd ls" hang forever when cluster is damaged
I'm not sure. Like I said, we usually don't want to timeout. In some user-facing tools we've added specific timeouts ... Greg Farnum

03/21/2015

05:20 AM Bug #11179: "rbd ls" hang forever when cluster is damaged
Greg Farnum wrote:
> In general, all of the commands which try to access data from RADOS will block until the data i...
science luo

03/20/2015

05:36 PM Bug #11179: "rbd ls" hang forever when cluster is damaged
In general, all of the commands which try to access data from RADOS will block until the data is accessible, on the t... Greg Farnum
09:22 AM Bug #11179 (Won't Fix): "rbd ls" hang forever when cluster is damaged
Hi there,
There is something wrong with my OSD,but I did not realize this,when I run "rbd ls",this command hangs lik...
science luo

03/19/2015

07:53 PM Bug #11113 (Pending Backport): RBD snapshot, created while flattening image, is unremovable
Josh Durgin

03/18/2015

07:35 PM Bug #5488 (Pending Backport): librbd: deadlock in image refresh
Josh Durgin

03/17/2015

04:17 PM Feature #3283 (Duplicate): rbd: calc-utilization or similar
Josh Durgin
04:09 PM Feature #11133: Add support for updating image features
https://github.com/ceph/ceph/pull/3980 Jason Dillaman
04:09 PM Feature #11133 (Resolved): Add support for updating image features
This will allow the exclusive lock, object map, and future journaling feature to be enabled/disabled on existing images. Jason Dillaman
04:07 PM Feature #8901 (Resolved): rbd mirroring: librbd: add functional tests for librbd lock breaks
Josh Durgin
04:02 PM Feature #11132 (Resolved): rbd CLI should be able to rebuild an invalid object map
This tool can also be used after enabling the object map on an existing image. Jason Dillaman
04:00 PM Feature #2557 (Rejected): QEMU support for image locking
automatic exclusive locking handles this Josh Durgin
01:47 PM Bug #11113 (Fix Under Review): RBD snapshot, created while flattening image, is unremovable
https://github.com/ceph/ceph/pull/4035 Jason Dillaman
01:17 PM Bug #11113 (In Progress): RBD snapshot, created while flattening image, is unremovable
Jason Dillaman

03/16/2015

03:17 PM Bug #5488 (Fix Under Review): librbd: deadlock in image refresh
https://github.com/ceph/ceph/pull/4013 Jason Dillaman
03:43 AM Bug #11113 (Resolved): RBD snapshot, created while flattening image, is unremovable
The following actions will create an unremovable RBD snapshot:
# Clone an image from an origin snapshot
# Flatten...
Benjamin Gilbert

03/12/2015

05:46 PM Bug #11074 (Duplicate): "FAILED assert(m_image_ctx.snap_lock.is_wlocked())" in rbd-hammer-distro-...
http://tracker.ceph.com/issues/11067 Jason Dillaman
05:45 PM Bug #5488 (In Progress): librbd: deadlock in image refresh
Jason Dillaman
03:57 PM Bug #11075: librbd: fsx failure - read zeroes when data was expected
does not affect hammer or master, just present when other cleanups are there Josh Durgin

03/11/2015

09:58 PM Bug #10956 (Resolved): upgrade from 0.80.7 to 0.80.8 make read request become slow
Josh Durgin
12:45 PM Bug #11067 (Resolved): FAILED assert(m_image_ctx.snap_lock.is_wlocked())
Jason Dillaman
11:15 AM Bug #8184: krbd: make sure we have latest osdmap on 'rbd map'
No, I don't think so. Can you do... Ilya Dryomov

03/10/2015

11:03 PM Bug #5488: librbd: deadlock in image refresh
Encountered this during a very long fsx run, deadlock after a flatten due to ictx_refresh()->refresh_parent()->clear_... Josh Durgin
07:55 PM Bug #11067 (Fix Under Review): FAILED assert(m_image_ctx.snap_lock.is_wlocked())
https://github.com/ceph/ceph/pull/3926 Jason Dillaman
05:43 PM Bug #8184: krbd: make sure we have latest osdmap on 'rbd map'
I've been experiencing this problem recently.
root@coordinator:~# cat /proc/version
Linux version 3.18.4pl2 (roo...
Jeff Epstein
01:43 AM Bug #11075 (In Progress): librbd: fsx failure - read zeroes when data was expected
Josh Durgin

03/09/2015

09:51 PM Bug #11075: librbd: fsx failure - read zeroes when data was expected
Client-side log with debugging: https://objects.dreamhost.com/joshd-public/client.admin.18275.log.xz Josh Durgin
09:40 PM Bug #11075 (Rejected): librbd: fsx failure - read zeroes when data was expected
fsx command was:... Josh Durgin
09:18 PM Bug #11074 (Duplicate): "FAILED assert(m_image_ctx.snap_lock.is_wlocked())" in rbd-hammer-distro-...
Run: http://pulpito.ceph.com/teuthology-2015-03-08_23:00:02-rbd-hammer-distro-basic-multi
Job: ['796738']
Logs: htt...
Yuri Weinstein
01:18 PM Bug #11067 (Resolved): FAILED assert(m_image_ctx.snap_lock.is_wlocked())
http://qa-proxy.ceph.com/teuthology/teuthology-2015-03-08_23:00:02-rbd-hammer-distro-basic-multi/796738/teuthology.lo... Jason Dillaman

03/06/2015

11:44 PM Fix #11056 (Resolved): librbd: aio calls may block
QEMU runs these from its threads, so blocking here causes the vcpu to lock up from the guest's perspective, resulting... Josh Durgin

03/04/2015

01:13 PM Bug #10968 (Resolved): librbd: copy-on-read request outstanding while closing image
Jason Dillaman

03/03/2015

06:02 PM Bug #10956 (Pending Backport): upgrade from 0.80.7 to 0.80.8 make read request become slow
Backport is including in the forthcoming v0.80.9 release. Jason Dillaman
01:19 PM Bug #10968 (Fix Under Review): librbd: copy-on-read request outstanding while closing image
https://github.com/ceph/ceph/pull/3864 Jason Dillaman
02:21 AM Bug #10962 (Resolved): Failed cram test cli-integration/rbd/formatted-output.t
Jason Dillaman

03/02/2015

10:54 PM Bug #10962 (Fix Under Review): Failed cram test cli-integration/rbd/formatted-output.t
https://github.com/ceph/ceph/pull/3850 Jason Dillaman
10:08 PM Bug #10962 (In Progress): Failed cram test cli-integration/rbd/formatted-output.t
Jason Dillaman
10:50 PM Bug #10990 (Resolved): test: copy.sh fails to remove locked image w/ exclusive lock feature enabled
https://github.com/ceph/ceph/pull/3848 Jason Dillaman
10:08 PM Bug #10990 (Fix Under Review): test: copy.sh fails to remove locked image w/ exclusive lock featu...
Jason Dillaman
08:24 PM Bug #10990 (Resolved): test: copy.sh fails to remove locked image w/ exclusive lock feature enabled
http://qa-proxy.ceph.com/teuthology/teuthology-2015-03-01_23:00:02-rbd-hammer-distro-basic-multi/784223/teuthology.lo... Jason Dillaman
07:11 PM Bug #9854 (Resolved): librbd: reads contending for cache space can cause livelock
Josh Durgin
04:37 PM Bug #10958 (Resolved): librbd: fsx detected invalid object map after resize without object map en...
Jason Dillaman

02/27/2015

07:41 PM Bug #10968 (In Progress): librbd: copy-on-read request outstanding while closing image
Jason Dillaman
06:33 AM Bug #10968 (Resolved): librbd: copy-on-read request outstanding while closing image
This caused a crash in ceph_test_librbd_fsx:... Josh Durgin
04:55 AM Bug #10958: librbd: fsx detected invalid object map after resize without object map enabled
https://github.com/ceph/ceph/pull/3812 Jason Dillaman
04:49 AM Bug #10958: librbd: fsx detected invalid object map after resize without object map enabled
Josh Durgin wrote:
> Seems like we still want it before invalidating the cache though?
Agreed -- I moved both flu...
Jason Dillaman
04:48 AM Bug #10958 (Fix Under Review): librbd: fsx detected invalid object map after resize without objec...
Jason Dillaman

02/26/2015

10:55 PM Bug #10958: librbd: fsx detected invalid object map after resize without object map enabled
Seems like we still want it before invalidating the cache though? Josh Durgin
05:37 PM Bug #10958: librbd: fsx detected invalid object map after resize without object map enabled
The AIO flush was removed during the lock updates. However, it really should have been in a AsyncResizeRequest state... Jason Dillaman
03:58 PM Bug #10958 (In Progress): librbd: fsx detected invalid object map after resize without object map...
Jason Dillaman
07:48 AM Bug #10958 (Resolved): librbd: fsx detected invalid object map after resize without object map en...
From http://qa-proxy.ceph.com/teuthology/joshd-2015-02-25_21:47:05-rbd-master---basic-multi/779213/teuthology.log
...
Josh Durgin
10:34 PM Bug #7142 (Resolved): list_lockers() never returns after cluster restart and health_ok (librbdpy)
commit:609f4c56718d8279895b02b8163bbe1976c02bfb Josh Durgin
10:29 PM Bug #10122 (Resolved): "LibRBD.TestClone" FAILED in upgrade:dumpling-x-firefly-distro-basic-vps run
Josh Durgin
10:22 PM Bug #10270 (Resolved): "[ FAILED ] LibRBD.ListChildren" in upgrade:firefly-x-giant-distro-basic...
Josh Durgin
10:21 PM Bug #10299 (Resolved): "ObjectCacher.cc: 530: FAILED assert(i->empty())" in rbd-firefly-distro-ba...
Josh Durgin
10:20 PM Bug #9513: rbd_cache=true default setting is degading librbd performance ~10X in Giant
To answer my question from above: this got resolved for firefly in #9854 (PR https://github.com/ceph/ceph/pull/3410) Ken Dreyer
10:16 PM Bug #9854: librbd: reads contending for cache space can cause livelock
PR (now merged) for firefly: https://github.com/ceph/ceph/pull/3410 Ken Dreyer
09:47 PM Bug #9854: librbd: reads contending for cache space can cause livelock
Josh Durgin
09:34 PM Bug #10961 (Resolved): Failed test copy.sh in rbd QA run
Jason Dillaman
05:09 PM Bug #10961: Failed test copy.sh in rbd QA run
https://github.com/ceph/ceph/pull/3806 Jason Dillaman
05:08 PM Bug #10961 (Fix Under Review): Failed test copy.sh in rbd QA run
Jason Dillaman
05:06 PM Bug #10961 (Resolved): Failed test copy.sh in rbd QA run
... Jason Dillaman
05:27 PM Bug #10962 (Resolved): Failed cram test cli-integration/rbd/formatted-output.t
http://qa-proxy.ceph.com/teuthology/joshd-2015-02-25_21:47:05-rbd-master---basic-multi/779289/teuthology.log Jason Dillaman
03:28 PM Bug #10956: upgrade from 0.80.7 to 0.80.8 make read request become slow
Fix is included in https://github.com/ceph/ceph/pull/3410 Jason Dillaman
04:57 AM Bug #10956 (Resolved): upgrade from 0.80.7 to 0.80.8 make read request become slow
I use Ceph+Openstack in our private cloud. I upgrade our centos6.5 based cluster from Ceph Emperor to Ceph Firefly. w... Wanyuan Yang

02/24/2015

05:33 PM Bug #10270: "[ FAILED ] LibRBD.ListChildren" in upgrade:firefly-x-giant-distro-basic-multi run
Run: http://pulpito.ceph.com/teuthology-2015-02-23_17:13:02-upgrade:firefly-x-hammer-distro-basic-multi/
Job: 776388...
Yuri Weinstein
 

Also available in: Atom