Project

General

Profile

Activity

From 09/24/2012 to 10/23/2012

10/23/2012

08:00 PM rbd Bug #2933: rbd: bio_pair leak in bio_chain_clone()
I have posted a patch for review that adjusts rbd_rq_fn() so it
uses a new function, bio_chain_clone_range(). The n...
Alex Elder
07:56 PM Linux kernel client Bug #3385 (Resolved): krbd: running simple fsstress produces corrupt XFS file system
The failure this was opened for was due to a bug in the
implementation of bio_chain_clone_range(), which was written...
Alex Elder
03:58 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
I think that the bio_pair leak won't be reproduced in
the current environment. The reason is that bio_split()
and ...
Alex Elder
12:38 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
This whole episode started after multiple attempts to
fix bio_chain_clone() so it wouldn't leak bio_pair
structures...
Alex Elder
11:21 AM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
I updated my code to reflect that bio's in a chain might
not be contiguous. It wasn't quite working, though, and
I...
Alex Elder
05:20 PM Documentation #3391: doc: add instructions on snapshot reversion
Is this a duplicate of #2979? Ross Turk
03:43 PM Documentation #3391: doc: add instructions on snapshot reversion
Also note that pool-wide snapshots can't be used at the same time as object-based (such as rbd) snapshots in a given ... Josh Durgin
03:40 PM Documentation #3391 (Closed): doc: add instructions on snapshot reversion
There is documentation around how to make a snapshot of a pool and remove one, but nothing on how to revert to a snap... Ross Turk
05:18 PM Documentation #3394 (Resolved): doc: create documentation for ceph-deploy
Ross Turk
03:46 PM Documentation #3393 (Resolved): doc: clarify process for setting replicas
Under Set Pool Values, it suggests that you can use "ceph osd set {pool-name} size {value}". Under Set the Number of... Ross Turk
03:43 PM Documentation #3392 (Resolved): doc: set/get pool values seems incomplete?
The list of pool values you can get is different from the list of pool values you can set. Shouldn't they match?
...
Ross Turk
03:33 PM Documentation #3390 (Resolved): doc: add detail on different bucket algorithms
There are three bucket algorithms (uniform, list, tree, straw), but they aren't explained
http://ceph.com/docs/mas...
Ross Turk
03:32 PM Documentation #3389 (Resolved): doc: crush docs could use a full example crushmap
Wido suggests that this page could use a full crush map as an example:
http://ceph.com/docs/master/cluster-ops/crush...
Ross Turk
02:34 PM Documentation #3388 (Resolved): doc: create documentation for juju installation
Ross Turk
12:42 PM rgw Bug #3381 (Resolved): rgw: processing continued after client shut down
Fixed, commit:3cc39f558d88c87122d011cea243637c6dc10cb4. Yehuda Sadeh
11:23 AM devops Feature #3158 (Closed): downburst: add --wait based on cd-rom eject detection kludge
Implemented in https://github.com/ceph/downburst/commit/481af791d02dabeb3c615e329d359bbed2deef2b Anonymous
10:18 AM rbd Subtask #3387 (Rejected): librbd: expose changed objects since a given snapshot
This could be used by #3272 and/or an rbd 'diff' command to show the changes between two snapshots or since a particu... Josh Durgin
10:17 AM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
ooh, captured libceph logging leading up to this!
Sage Weil
05:39 AM rbd Feature #3262 (Resolved): librbd: use 'assert exists' operation to guard cloned writes
Sage Weil

10/22/2012

08:34 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
> That's surprising to me that a bio might represent a discontiguous IO...
> is there someone we can bounce that the...
Alex Elder
08:28 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
I wanted to add this detail but forgot... These are the
debug statements that I thought told the story.
[10758.8...
Alex Elder
05:55 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
That's surprising to me that a bio might represent a discontiguous IO... is there someone we can bounce that theory o... Sage Weil
05:44 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
I think I may know what's going on, at least with my particular
fix for the bio_chain_clone(). If I'm right the pro...
Alex Elder
04:49 PM Linux kernel client Bug #3385: krbd: running simple fsstress produces corrupt XFS file system
OK, I've simplified it further. The dependence on name sizes
makes me think it has to do with log blocks. We'll se...
Alex Elder
03:49 PM Linux kernel client Bug #3385 (Resolved): krbd: running simple fsstress produces corrupt XFS file system
This does not occur with the current ceph-client/master branch:
35152979 rbd: activate v2 image support
Howe...
Alex Elder
08:33 PM rbd Bug #3383 (Resolved): rbd copy fails in the nightlies
Dan Mick
08:30 PM rbd Bug #3383: rbd copy fails in the nightlies
Ah, I was just assuming copy was old; it was changed in 62420599006691d70a1634223bd0d1a3dc10e9ee, and the new size c... Dan Mick
08:26 PM rbd Bug #3383: rbd copy fails in the nightlies
So I understand the problem; the check for 'size is the same' was checking
the size of the source image at the reque...
Dan Mick
01:55 PM rbd Bug #3383: rbd copy fails in the nightlies
More logs:
ubuntu@teuthology:/a/teuthology-2012-10-21_19:00:07-regression-master-testing-gcov/5324
ubuntu@teutholog...
Tamilarasi muthamizhan
01:53 PM rbd Bug #3383 (Resolved): rbd copy fails in the nightlies
Logs: ubuntu@teuthology:/a/teuthology-2012-10-21_19:00:07-regression-master-testing-gcov/5323... Tamilarasi muthamizhan
04:33 PM Bug #3386 (Resolved): #7 ReplicatedPG::add_object_context_to_pg_stat (this=0x8654000, obc=0xe34...
segfault in #7 ReplicatedPG::add_object_context_to_pg_stat (this=0x8654000, obc=0xe342000, pgstat=0x7f9a8bf97c10) a... Samuel Just
03:24 PM Documentation #3099 (Resolved): doc: write install ceph+openstack with crowbar
Documented and reviewed. John Wilkins
03:23 PM Bug #2947 (In Progress): osd: out of order reply
Recent logs: ubuntu@teuthology:/a/teuthology-2012-10-20_19:00:11-regression-master-testing-gcov/4810... Tamilarasi muthamizhan
02:28 PM CephFS Bug #1947: mds: SIGBUS during _mark_dirty
Recent logs: ubuntu@teuthology:/a/teuthology-2012-10-22_00:00:20-regression-next-testing-basic/5426... Tamilarasi muthamizhan
01:39 PM Bug #3270 (Duplicate): osd crash during rbd test run
dup #3142 Sage Weil
12:36 PM Documentation #3382: doc: create page for OS requirements, detailing compatibility for the packag...
I was able to install Ceph .52 and mount a cfuse directory. iogen test worked fine. That's as far as I have gotten w... Ken Franklin
12:14 PM Documentation #3382: doc: create page for OS requirements, detailing compatibility for the packag...
Gary says: "We currently build centos6 and fedora17. The fedora18 build is working, but we've not pushed packages ... Ross Turk
11:59 AM Documentation #3382 (Resolved): doc: create page for OS requirements, detailing compatibility for...
Ross Turk
11:29 AM rgw Bug #3365: Broken metadata (duplicated as CSV)
Not sure exactly where the csv is coming from, may be the client library doing something. I'll need to see radosgw lo... Yehuda Sadeh
11:10 AM rgw Bug #3365: Broken metadata (duplicated as CSV)
According to https://bugs.launchpad.net/swift/+bug/939982 , clients should not rely on casing. I guess MediaWiki can ... Aaron Schulz
10:55 AM rgw Bug #3365: Broken metadata (duplicated as CSV)
I made a small logging change to our tests (https://gerrit.wikimedia.org/r/#/c/29346/1) which causes a large number o... Aaron Schulz
10:57 AM rgw Bug #3381 (Resolved): rgw: processing continued after client shut down
We continue processing requests long after the client has closed the connection. Yehuda Sadeh
10:51 AM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
wip-msgr didn't help
this appears to be triggered by osd reset, which isn't exercised by the socket errors. going...
Sage Weil
09:42 AM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
crossing my fingers that
rbd: reset BACKOFF if unable to re-queue
will fix it
Sage Weil
10:45 AM Bug #3142: osd: crash induced by fsx workload
fix for the watcher thing merged to next branch, yay! hopefully that was the root cause for the mysterious nightly f... Sage Weil
10:25 AM Documentation #3173 (Resolved): Cephx Overview
John Wilkins
10:04 AM Bug #3271 (Resolved): kernel BUG() on unmount after export/unexport of ceph using kclient
pushed to testing branch Sage Weil
04:58 AM rbd Feature #3380 (Resolved): rbd: export and import functions should support stdout
Currently these functions do not allow import/export.
It would be very useful if these functions were supported fo...
Wido den Hollander

10/21/2012

07:05 PM Bug #3379 (Won't Fix): rbd map + rbd mv: couldn't remove old source object
ceph @7477a081, linux-3.4.12
Using 'rbd mv' on a mapped object creates a "ghost" object that prevents
further use...
Chris Dunlop
06:14 PM Bug #3378 (Can't reproduce): common/HeartbeatMap.cc: 78: FAILED assert(0 == "hit suicide timeout")
This is a cluster of 2 OSDs that is generally unhappy with life. After deleting the cephfs pools the new pool creatio... Matthew Roy
03:35 PM Feature #3324 (Resolved): mon: coverity clean
Sage Weil
02:53 PM CephFS Bug #3301 (Resolved): cfuse: chmod not getting respected
Sage Weil
02:27 PM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
also reproduced on testing after 10 iterations. Sage Weil
08:45 AM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
was able to reproduce on kernel master after 7 iterations of hammer.sh. Sage Weil
02:25 PM CephFS Bug #1527 (Fix Under Review): mds: assert in handle_cache_expire
reproduced this with a log. the problem is the preceding conditional not including the EXPORTING stage; testing the ... Sage Weil
09:25 AM CephFS Bug #3371: java: ldout after ceph_shutdown causes segfault
Noah Watkins
09:25 AM CephFS Feature #3368: java: add O_WRONLY open flag
Noah Watkins
09:24 AM CephFS Feature #3374: java: add unit tests for CephMount use after shutdown
Noah Watkins
09:24 AM CephFS Bug #3375: java: shutdown() doesn't clean-up if unmounted
Noah Watkins
09:21 AM CephFS Feature #3372: java: add finalize() for users that don't do explicit clean-up
Noah Watkins
09:19 AM CephFS Bug #3373: java: after shutdown() access ceph_* in JNI is not prevented
Noah Watkins
09:07 AM Feature #3377 (Resolved): cephx: sign messages
Sage Weil
09:03 AM Feature #3215 (Resolved): qa: fix up system tests and add them to regression suite
Sage Weil
08:52 AM rbd Bug #2937 (Duplicate): btrfs filesystem on rbd device kernel BUG writing large file
Sage Weil
08:50 AM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
this was a btrfs bug; Josef has a fix pending. Sage Weil
08:47 AM Feature #3376 (Duplicate): use external leveldb package for default builds
Sage Weil

10/20/2012

06:48 PM Linux kernel client Bug #3342: con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
again on master branch, ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2012-10-19_02:00:08-regression-test... Sage Weil
05:38 PM CephFS Bug #3375 (Won't Fix): java: shutdown() doesn't clean-up if unmounted
shutdown() throws CephNotMountedException if CephMount is not mounted. This means that the native memory allocated by... Noah Watkins
01:32 PM Bug #3142: osd: crash induced by fsx workload
i got a log for... Sage Weil
08:14 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
2) e.g. if `ctx->read` needs an empty list, it should throw from `ctx->read`, where the application can catch excepti... Artem Grinblat
08:01 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
1) yes, some documentation in librados.hpp would be nice
2) that `ctx->read (id, bl, size, 0)` might append into `bl...
Artem Grinblat

10/19/2012

05:09 PM CephFS Feature #3374 (Rejected): java: add unit tests for CephMount use after shutdown
There is one unit test per function in CephMount. Pre-test setup is a new CephMount object followed by shutdown(). Ca... Noah Watkins
04:56 PM rbd Documentation #3171 (Resolved): doc: how to setup OpenStack (Glance and Cinder|Nova) to use rbd c...
Josh Durgin
04:56 PM CephFS Bug #3373: java: after shutdown() access ceph_* in JNI is not prevented
An alternative semantics which is more natural is to return CephMount to its initial state by performing a shutdown o... Noah Watkins
04:28 PM CephFS Bug #3373 (Won't Fix): java: after shutdown() access ceph_* in JNI is not prevented
When CephMount is created, a mount context is created to back it. Checks in libcephfs return errors for calls when us... Noah Watkins
03:53 PM CephFS Feature #3372 (Closed): java: add finalize() for users that don't do explicit clean-up
The JVM will run finalize() before GC, but is not guaranteed to ever do so. Still, it is good to do this in case user... Noah Watkins
03:37 PM CephFS Bug #3371 (Won't Fix): java: ldout after ceph_shutdown causes segfault
#define dout_subsys ceph_subsys_javaclient
...
ceph_shutdown
ldout(cct, 10) <- blows up
Noah Watkins
03:28 PM Bug #3300: ceph::buffer::end_of_buffer isn't caught
So I think there are two things here:

1) your return code check for getxattr will never pass - getxattr() returns...
Josh Durgin
11:17 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Modifying the "extracting" part not to reuse the bufferlist seems to workaround the problem:... Artem Grinblat
09:33 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Attached is the log (with "log file = /tmp/librados.log\n debug ms = 1\n debug rados = 20" in ceph.conf). Artem Grinblat
09:24 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
I was thinking a log of the client side. I'm not sure if the log from an osd will be necessary. You can set the debug... Josh Durgin
09:22 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
E.g. both ceph and librados are from:... Artem Grinblat
09:21 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Version is the already mentioned "librados-dev 0.53-1~bpo70+1".
There are four OSD servers, all track the same Debia...
Artem Grinblat
09:17 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Also curious what version of librados is running vs what version of ceph-osd it is talking to. Sage Weil
09:12 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Here's the code that's causing this:... Artem Grinblat
09:06 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
What are you doing with librados that's causing this? A log with 'debug ms = 1' and 'debug rados = 20' would help, as... Josh Durgin
05:09 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
GDB stack trace:... Artem Grinblat
04:14 AM Bug #3300: ceph::buffer::end_of_buffer isn't caught
Same problem with @librados-dev 0.53-1~bpo70+1@. Artem Grinblat
03:25 PM CephFS Bug #3370 (Resolved): All nfsd hung trying to lock page(s) on export of kclient ceph
Workunit bonnie hung over NFS client with retransmitted NFS read:
ubuntu 2667 2572 0 Oct18 ? 00:00:00 ...
David Zafman
03:13 PM CephFS Bug #3369: journaled two client session close events
3369-mds-session-workaround for a bypass. Not sure if we want to merge this or not since it's just a bandaid!
It's ba...
Greg Farnum
02:05 PM CephFS Bug #3369: journaled two client session close events
Dur, forgot the file. Greg Farnum
02:05 PM CephFS Bug #3369 (Resolved): journaled two client session close events
Unfortunately there are no logs of the event occurring, but an MDS crash on startup was reported on the mailing list.... Greg Farnum
01:54 PM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
I just tested a btrfs patch provided by Josef Bacik and
I no longer am hitting this problem. I will attach the
pat...
Alex Elder
01:50 PM CephFS Bug #3346 (Resolved): client: we broken open permissions
Sage Weil
12:16 PM CephFS Feature #3368 (Closed): java: add O_WRONLY open flag
Noah Watkins
10:25 AM CephFS Bug #3348 (Resolved): java: use libcephfs naming rather than generic ceph
Noah Watkins
10:24 AM CephFS Bug #3348: java: use libcephfs naming rather than generic ceph
Now builds packages:
* libcephfs-java
* libcephfs-jni
Noah Watkins
10:04 AM rgw Bug #3365: Broken metadata (duplicated as CSV)
Is that reproducible? If so, a gateway log could help (with debug ms = 1, debug rgw = 20) Yehuda Sadeh
09:36 AM CephFS Bug #3367 (Resolved): client: hardlink without original file breaks on unmount
I'm not able to reproduce this on fuse, but with the libcephfs client, the following:
mkdir dir1
cd dir1
create ...
Sam Lang
09:13 AM rgw Feature #3201 (Resolved): rgw: coverity clean
Sage Weil

10/18/2012

09:33 PM rgw Feature #3366 (Resolved): rgw: dr: define management api
Yehuda Sadeh
09:26 PM rgw Feature #1717 (Fix Under Review): rgw: support json input
Yehuda Sadeh
08:25 PM Feature #3355: make Ceph build faster, take less RAM to compile
These are good links. The low hanging fruit here seems to be precompiled headers (its hard to justify restructuring ... Sam Lang
01:14 PM Feature #3355 (New): make Ceph build faster, take less RAM to compile
Linkdump:
http://stackoverflow.com/questions/373142/what-techniques-can-be-used-to-speed-up-c-compilation-times
htt...
Anonymous
05:35 PM rgw Bug #3365 (Can't reproduce): Broken metadata (duplicated as CSV)
I am having a problem running MediaWiki with rados+rgw. The code has worked fine with Swift for months (and in produc... Aaron Schulz
05:28 PM Bug #3361: don't log secrets
I will take care of this in the morning, and will make sure nothing else of value is leaked onto the logs. Joao Eduardo Luis
03:22 PM Bug #3361: don't log secrets
there's a larger audit that needs to be done here. soon! Sage Weil
02:15 PM Bug #3361 (Resolved): don't log secrets
... Anonymous
05:14 PM Bug #3270: osd crash during rbd test run
Latest logs: ubuntu@teuthology:/a/teuthology-2012-10-17_19:00:10-regression-master-testing-gcov/3136... Tamilarasi muthamizhan
04:44 PM rbd Feature #3239 (Resolved): librbd: make striping fancier
Sage Weil
04:10 PM rgw Feature #3364 (New): rgw: cache objects
We can cache objects, using their tags for validation.
Yehuda Sadeh
03:21 PM RADOS Feature #3362 (Resolved): Warn users before allowing pools to be created with more than N*<num_os...
Samuel Just
02:56 PM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
... Alex Elder
02:52 PM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
... Alex Elder
09:10 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
... Alex Elder
08:16 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
Whoops, should have previewed that. Here's the same thing,
formatted as I intended....
Alex Elder
08:13 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
I can easily reproduce this problem by doing this in my UML
environment:
rbd create image1 --size=1024
r...
Alex Elder
01:53 PM RADOS Feature #3360 (New): osd: Prevent primary from choosing a pg_temp that violates the crush rule sp...
Not sure how to actually do this. There is currently no guarrantee that the primary while trolling the prior_set for... Samuel Just
01:51 PM RADOS Feature #3359 (New): Add perf counter tracking for major memory affecting structures
We want to count:
1) number of pgs
2) total number of past_intervals
3) total number of log_entries
4) total numb...
Samuel Just
01:50 PM Feature #3358 (Resolved): Add pool option for min size to prevent a pg from going active with exc...
Otherwise, the fate of a pg might end up resting on a single osd. Samuel Just
01:31 PM devops Feature #3357 (Rejected): How to revoke OSD for good?
Use case: As an admin, a machine I had running 8 OSDs was compromised. The attacker got hold of all the secrets the O... Anonymous
01:07 PM devops Feature #3354 (New): good default answer for how you should do monitoring, log analysis, metrics
Anonymous
01:05 PM devops Feature #3353 (Rejected): commands for moving journals
As an admin, I want to move an OSD data disk from one physical host (chassis) to another (e.g. for maintenance of non... Anonymous
12:58 PM devops Feature #3351: ceph-deploy: Support upgrades
Note: beware indirectly installed packages. E.g. "ceph-deploy install" does "apt-get install ceph", that perhaps upgr... Anonymous
12:55 PM devops Feature #3351 (Resolved): ceph-deploy: Support upgrades
"ceph-deploy install HOST" should also do upgrade/downgrades, if that's what's needed to satisfy the --dev= etc optio... Anonymous
12:51 PM devops Feature #3350 (Rejected): ceph-deploy: add "ceph-deploy refresh" that just pushed out changes to ...
Anonymous
10:03 AM CephFS Bug #3348 (Resolved): java: use libcephfs naming rather than generic ceph
Noah Watkins
09:21 AM devops Feature #3255: ceph-disk: allow prepare without activate (for spares)
For a while I thought that a ceph.conf entry like osd_activate_on_hotplug=false would help, but that conflicts with t... Anonymous
09:16 AM devops Feature #3347 (Resolved): ceph-deploy: allow setting ssh user
Anonymous
08:09 AM CephFS Bug #3346: client: we broken open permissions
I didn't fully understand the (subtle) semantics of open() with O_CREAT and the mode. It expects the file to first b... Sam Lang
01:13 AM CephFS Bug #3346: client: we broken open permissions
first two:... Sage Weil
12:50 AM CephFS Bug #3346 (Resolved): client: we broken open permissions
... Sage Weil

10/17/2012

11:06 PM Bug #3135: mon_host without ports leads to failure
Sage Weil
11:04 PM Bug #3145 (Resolved): ceph-mon: mon_host insists on commas, does not understand spaces
commit:1c382869ba6aff0c159103a8267045c6eb600114
Sage Weil
06:10 PM CephFS Bug #3339 (Resolved): libcephfs: failure on Dir_ls test
test was broken. Sage Weil
10:34 AM CephFS Bug #3339 (Resolved): libcephfs: failure on Dir_ls test
... Sage Weil
05:48 PM Feature #3345 (Resolved): support multiple clusters with sysvinit
e.g., 'service ceph start --cluster foo'. Sage Weil
03:29 PM devops Feature #3344 (Resolved): ceph-disk-prepare?: provide some sort of tool to clear out a disk, with...
"sgdisk --zap-all --clear" just doesn't clear out all the possible funky states a disk may be in.
A simple "dd if=...
Anonymous
03:16 PM rgw Feature #3343 (Resolved): rgw: stripe objects
By default, objects that are uploaded in a plain upload (not multipart) span over a max of two objects. That means th... Yehuda Sadeh
03:13 PM Linux kernel client Bug #3342 (Resolved): con_work crash on bad con pointer or mutex (kclient ffsb + thrashing)
... Sage Weil
01:20 PM devops Feature #3341 (Rejected): ceph-disk-activate: Make --mount the default
Cautious route: introduce --no-mount, make --mount default, etc.
Probably good enough: remove the --mount from src...
Anonymous
12:04 PM Bug #3233 (Resolved): osd: caps in master are incompatible with argonaut caps
This was merged a couple weeks ago. Josh Durgin
11:26 AM Feature #3340 (Rejected): refuse to accept "cluster=foo" in ceph.conf
I've seen some confusion with people trying to have /etc/ceph/ceph.conf with cluster=foo inside it (as opposed to how... Anonymous
09:20 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
The way things are supposed to work for bio_split() is that
any bio that will be subject to a split will have alread...
Alex Elder
08:55 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
I'm think I might know what's happening here.
Although rbd_merge_bvec() is a little odd in how it works, I
believ...
Alex Elder
08:21 AM Bug #2828: osd: assign_bid was allowed to mutate and return data
this block in ReplicatedPG.cc should *also* blow away the outdata on the osd_ops:... Sage Weil

10/16/2012

09:28 PM Bug #3213: PG: we do not persist missing set elements not in the log
this is fixed now, right? Sage Weil
09:28 PM Bug #3252 (Resolved): mon: mixed mon cluster of argonaut + master eats cpu on new code
Sage Weil
09:28 PM rbd Bug #3237 (Resolved): rbd_header_race failes to reproduce a race
Sage Weil
05:02 PM Feature #3324 (In Progress): mon: coverity clean
Sage Weil
01:54 PM Feature #3324 (Resolved): mon: coverity clean
Sage Weil
04:57 PM Bug #2339: osd: EBUSY on object delete when watchers present
when this is fixed, testrados_watch_notify needs to get its last test reenabled! it is currently disabled due to thi... Sage Weil
04:57 PM Feature #3215 (Fix Under Review): qa: fix up system tests and add them to regression suite
wip-testrados Sage Weil
04:20 PM CephFS Feature #3338 (Resolved): java: add unit tests for link/unlink/readlink/symlink
Noah Watkins
04:14 PM CephFS Bug #3337 (Rejected): java: unit tests don't remove container directory after a failure
Java unit tests construct a container directory in which lots of temporary files are created and destroyed. The conta... Noah Watkins
03:55 PM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
Just adding a little debug notes for my own reference.
Here is a dump of the bio involved in a crash using the bas...
Alex Elder
01:45 PM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
After poking around with kdb this afternoon I have narrowed it down to
bio_split() being called with something other...
Alex Elder
10:37 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
Reproduced it using stock kernel v3.6.1 (which was what was
reported used by tziOm).
Alex Elder
09:44 AM Linux kernel client Bug #3291 (In Progress): rbd: BUG in bio from rbd_rq_fn
Additional notes on the configuration that reproduced this:
- 4 OSD's, 3 monitors, (1 MDS)
- 1 GB RBD image
Alex Elder
03:54 PM Documentation #3173: Cephx Overview
Added a new section with help from Peter Reiher. There is some additional compatibility information that may be added... John Wilkins
03:45 PM Feature #3336 (Closed): java: create teuthology task for running unit test suite
We'd like to integrate the Java unit tests into the normal set of automated tests. Noah Watkins
02:56 PM CephFS Documentation #3335 (Rejected): doc: Explain kernel dynamic printk debugging
Under doc/dev/
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/1809/focus=1810
Anonymous
02:55 PM Documentation #3334 (Duplicate): doc: Explain how to choose journal size, when does the journal g...
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/5210/focus=5269 Anonymous
02:51 PM Documentation #3333 (Resolved): doc: Explain "degraded" more
Especially how an OSD can be "active" for a PG but still not have all the objects, and how that can be monitored.
...
Anonymous
02:48 PM rbd Documentation #3332 (Resolved): doc: Explain how qemu rbd caching defaults come from qemu config
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/7227/focus=7228 Anonymous
02:45 PM Documentation #3331 (Resolved): doc: Where is my data placed?
..this is often useful for learning the system, and ultimately troubleshooting; caution reader that placement is not ... Anonymous
02:42 PM Documentation #3330 (Closed): doc: How to troubleshoot unbalanced CRUSH
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/8904
Also note little gem of "Note that the ideal situa...
Anonymous
02:40 PM Documentation #3329 (Resolved): doc: What metrics should be used to set node weight
Note: this is more about adjusting the weights dynamically than setting them once in the beginning.
http://thread....
Anonymous
02:38 PM Documentation #3328 (Closed): doc: "Designing a cluster guide" feedback, extra material
Make sure the content from this thread is folded into the current docs:
http://thread.gmane.org/gmane.comp.file-sys...
Anonymous
02:36 PM Documentation #3327 (Resolved): doc: replica size=1 can leave pg stale forever
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/8287/focus=8292 Anonymous
02:29 PM Documentation #3326 (Closed): doc: Workaround needed to put journal on tmpfs
tmpfs doesn't to direct_io, hence journal creation will fail, workaround is journal_dio=false.
http://thread.gmane...
Anonymous
02:25 PM rbd Documentation #3325 (Closed): doc: Show qemu example with space->semicolon workaround
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/6381/focus=6400
Also document the general convention th...
Anonymous
02:10 PM rgw Bug #3297: Rados Gateway does not handle Transfer-Encoding: chunked
Try using the following deb:
[[http://gitbuilder.ceph.com/libapache-mod-fastcgi-deb-precise-x86_64-basic/ref/maste...
Yehuda Sadeh
01:38 PM rgw Bug #3297: Rados Gateway does not handle Transfer-Encoding: chunked
Some version info:
apache2 (2.2.22-1ubuntu1) precise; urgency=low
libapache-mod-fastcgi (2.4.7~0910052141-1) unstab...
Aaron Schulz
01:31 PM Feature #3323 (Resolved): When is it safe to reboot a node?
"Something like:
- make sure losing any local mon won't break quorum
- make sure all PGs touching local osd(s) ...
Anonymous
01:26 PM Documentation #3322 (Resolved): doc: Explain multi-tenant CephFS
We end up explaining this over and over, let's write it up once.
http://thread.gmane.org/gmane.comp.file-systems.c...
Anonymous
01:25 PM Documentation #3321 (Resolved): doc: Explain monitor HA better
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/5730/focus=5789 Anonymous
01:22 PM Documentation #3320 (Resolved): doc: What persistency does Ceph guarantee
(I'm cleaning out my notes, stumbled on this, it's a good topic we don't explain well enough.)
The explanations fr...
Anonymous
01:03 PM CephFS Bug #2681 (Resolved): client: got push without mds session
commit:db976663a50ce7ab2448c42cb0c6bfd4f58d25a5 Sage Weil
12:24 PM rgw Bug #3296 (Resolved): Ceph RADOS gateway does not handle apache listening on ports other than 80
Pushed a fix, commit:8d7c8e3b860f448426ba0220fe98f70e9b91622f. Yehuda Sadeh
10:48 AM CephFS Bug #3301: cfuse: chmod not getting respected

This time the logs are from doing the following:
root@scout:/tmp/cmnt# echo foo > foo
root@scout:/tmp/cmnt# chm...
Sam Lang
10:19 AM CephFS Bug #3301: cfuse: chmod not getting respected
This log is of the client with only the following operations performed:
echo baz > baz
chmod 400 baz
chmod 644 b...
Sam Lang
09:56 AM CephFS Bug #3301: cfuse: chmod not getting respected
Another issue:
chmod 400 a11
echo "foo" >> a11
< now fails correctly >
chmod 644 a11
ls -la a11
-r-------- 1 ...
Sam Lang
07:44 AM Feature #3319 (Resolved): java: add unit test target with check:jni,nabounds options
Run unit tests with check:jni and check:nabounds options that perform various JNI-related sanity checks. Hopefully th... Noah Watkins
07:37 AM CephFS Bug #3318 (Rejected): java: lock access to CephStat, CephStatVFS from native
Native code needs to synchronize updates to Java structures with the runtime. Noah Watkins
07:24 AM Feature #3317 (Rejected): java: add multi-threaded unit tests
Remove @synchronized@ bkl business. Java should be using pthreads, so Ceph locking should be fine. Noah Watkins
07:09 AM Feature #3316 (Rejected): java: add generated JavaDoc pages to ceph.jar
Noah Watkins
06:19 AM CephFS Feature #3315 (Resolved): client: Add acl support

Add acl support to the client. This will probably require a library to parse the acl extended attributes, and mods...
Sam Lang
06:15 AM CephFS Feature #3314 (Resolved): client: client interfaces should take a set of group ids

Right now each of the client interfaces takes only uid and gid parameters. Extend the interface to take a set of g...
Sam Lang

10/15/2012

08:59 PM Feature #3191 (Resolved): client: coverity clean
Sage Weil
08:59 PM Documentation #2996 (Resolved): doc: write install Ceph with RPMs doc
Sage Weil
05:54 PM rgw Feature #771 (Fix Under Review): rgw: POST
caleb miles
02:16 PM Feature #3235 (Resolved): doc: procedure for collecting data from failed node/whatever
Sage Weil
02:15 PM Feature #3167 (Resolved): auth: separate read/write capabilities for classes
Sage Weil
02:14 PM Feature #3193 (Resolved): cls: coverity clean
Sage Weil
01:35 PM devops Feature #3119 (Duplicate): chef: support destroying an osd
Duplicate of #2939 Anonymous
01:34 PM devops Feature #2939: chef: Write up how cluster shrinking should work
Moving content from duplicate #3119:
DH cookbooks do this by setting a node attribute that maps osd.id -> desired ...
Anonymous
01:27 PM devops Documentation #2976 (Closed): doc: update chef doc to git clone with http, not ssh
Anonymous
01:22 PM Bug #3287: OSD dies when using zfs
Clean new unused zfs filesystems are mounted at /data/osd.N, /data is in the root ext4 filesystem. The ceph filesyste... Mike Lowe
01:16 PM rbd Feature #3100 (In Progress): rbd-fuse prototype
Sage Weil
12:39 PM devops Feature #3313 (Resolved): ceph-deploy: some way to initialize GPT on SSD storing multiple journals?
Anonymous
12:39 PM devops Feature #2398: chef: external osd journal support
ceph-disk-prepare now has external journal support, much of this doesn't have to live in the cookbook.
If admin wa...
Anonymous
11:25 AM devops Feature #3312 (Resolved): ceph-deploy: pushy uses pickle, that's a security problem
That means "ceph-deploy osd myhost:sdb" could allow myhost to attack the admin workstation. Deserializing pickles fro... Anonymous
11:25 AM devops Feature #3311: ceph-deploy: pushy bug triggers on interpreter shutdown
Probably a threading bug inside the pushy library, perhaps in a __del__ method. At the time it triggers, all actions ... Anonymous
11:24 AM devops Feature #3311 (Resolved): ceph-deploy: pushy bug triggers on interpreter shutdown
Anonymous
11:19 AM devops Feature #3310 (Resolved): ceph-deploy: handle stderr from commands better
Currently, we use pushy (http://pypi.python.org/pypi/pushy/ ) for RPC-over-SSH to the target node, and don't really d... Anonymous
11:18 AM devops Feature #3019: juju: modernize ceph charm, mon & osd bootstrap
James Page from Canonical has been making nice progress with this. Anonymous
10:50 AM Feature #3305 (Closed): have way of separating auth keys per host
It would be nice to separate hosts from each other.. myapp on host1 is not the same as myapp on host2, even if both r... Anonymous
10:45 AM devops Feature #3304 (Resolved): ceph-deploy: add "ceph-deploy client HOST", that adds ceph.conf and cli...
Anonymous
10:32 AM devops Feature #3303: ceph-deploy: support external journals
Using ceph-disk-prepare external journal support from #3078, #3079, support
ceph-deploy osd HOST:DATADISK[:JOURN...
Anonymous
10:30 AM devops Feature #3303 (Resolved): ceph-deploy: support external journals
Anonymous
10:29 AM devops Feature #3082 (Closed): ceph-deploy: Support deployments without a framework like Chef, Puppet, J...
http://github.com/ceph/ceph-deploy is plenty good enough to serve as a base for further tickets. Anonymous
09:55 AM CephFS Bug #3301: cfuse: chmod not getting respected
yeah, apparently fuse isn't doing the check itself... :/ we should enforce this in Client::_open(), methinks. after... Sage Weil
07:17 AM CephFS Bug #3301 (Resolved): cfuse: chmod not getting respected
echo "foo" > foo
chmod 400 foo
echo "bar" >> foo
Should fail with EPERM, but it succeeds.
Attached is a clien...
Sam Lang
09:54 AM devops Feature #3302: ceph-disk: activate-journal, and matching udev rule
Summary: try starting the OSD again when the journal disk appears.
Currently, osd data disk add is seen in /etc/in...
Anonymous
09:47 AM devops Feature #3302 (Resolved): ceph-disk: activate-journal, and matching udev rule
Anonymous
09:42 AM devops Feature #3078 (Resolved): support journal in 2nd partition on same disk as osd
In master as commit 662c69e525fe8c88c65e8d5c597e29634ea3d521, pushed to branch for-stable-ext-journal as commit 68fe1... Anonymous
09:42 AM devops Feature #3079 (Resolved): support journal on any accessible block device
In master as commit 662c69e525fe8c88c65e8d5c597e29634ea3d521, pushed to branch for-stable-ext-journal as commit 68fe1... Anonymous
04:44 AM rgw Bug #3294: Ceph S3 API test
I have figured it out that the error is due to lacking lib???when I try to compile the ceph,I forget the "--with-rado... lollipop king

10/14/2012

11:10 AM Bug #3300 (Resolved): ceph::buffer::end_of_buffer isn't caught
A program using librados terminates with the following uncaught exception:... Artem Grinblat

10/13/2012

11:17 PM Bug #3284 (Resolved): logrotate script broken
thanks, applied this (with your signed-off-by; assuming that's ok). Sage Weil
09:17 PM Feature #1720 (Duplicate): qa: rpm autobuilders
Sage Weil
09:16 PM Feature #2807 (Resolved): RPM git-builder
Sage Weil
08:14 AM rgw Documentation #3299 (Closed): rgw: docs should be clearer about fastcgi module
Docs don't really mention where to get the fastcgi module. Also, the fastcgi should also mention that it fixes chunke... Yehuda Sadeh
08:13 AM rgw Bug #3297: Rados Gateway does not handle Transfer-Encoding: chunked
Yeah, looking at the docs now I can't find a concrete reference to the fastcgi module. So the issue is that the fastc... Yehuda Sadeh

10/12/2012

09:23 PM rgw Bug #3294: Ceph S3 API test
Yehuda Sadeh wrote:
> Is there anything on the radosgw/apache logs?
Absolutely?there is nothing wrong reported ...
lollipop king
10:50 AM rgw Bug #3294: Ceph S3 API test
Is there anything on the radosgw/apache logs? Yehuda Sadeh
06:31 AM rgw Bug #3294 (Rejected): Ceph S3 API test
hello,everyone...
nowadays i try to test Ceph s3 API. I have configure the ceph to run and can use it's POSIX API
a...
lollipop king
08:12 PM rgw Bug #3297: Rados Gateway does not handle Transfer-Encoding: chunked
Yehuda Sadeh wrote:
> What version are you using? It actually is supported, but it might be that your fastcgi module...
Aaron Schulz
04:55 PM rgw Bug #3297: Rados Gateway does not handle Transfer-Encoding: chunked
What version are you using? It actually is supported, but it might be that your fastcgi module swallows it. What fast... Yehuda Sadeh
03:39 PM rgw Bug #3297 (Rejected): Rados Gateway does not handle Transfer-Encoding: chunked
You can test this using curl & bash to PUT and object using Transfer-Encoding: chunked and thus no Content-Length hea... Aaron Schulz
03:55 PM Bug #3287: OSD dies when using zfs
For reference, osd.0 is failing with this backtrace:... Josh Durgin
03:37 PM rgw Bug #3296 (Resolved): Ceph RADOS gateway does not handle apache listening on ports other than 80
See http://ceph.com/docs/master/radosgw/config/
Anything other than <VirtualHost *:80> is broken. You will be able...
Aaron Schulz
09:50 AM Bug #3285 (Closed): rbd map w/ cephx
Cool, glad that fixed it. Probably the best way to get the solution indexed is to email ceph-devel. The list archives... Josh Durgin
09:18 AM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
Possibly related: http://tracker.newdream.net/issues/3291
Alex Elder
09:18 AM Linux kernel client Bug #3291: rbd: BUG in bio from rbd_rq_fn
This may be a duplicate of http://tracker.newdream.net/issues/2937
and even if not, it's probably related.
Alex Elder
08:39 AM devops Feature #3295 (Rejected): UML: configure boot image for Ubuntu
I'm not sure how to classify or assign this.
I use UML a lot for efficient development of the kernel code. This
...
Alex Elder
12:30 AM Bug #3286: librbd, kvm, async io hang
FYI, 8 rounds of the full size=4g test passed without a hitch. Chris Dunlop

10/11/2012

07:39 PM Bug #3286: librbd, kvm, async io hang
Attached, output of:... Chris Dunlop
06:26 PM Bug #3286: librbd, kvm, async io hang
Thanks for investigating so much. It certainly sounds like the issue is with the cache. Could you get a log with debu... Josh Durgin
05:41 PM Bug #3286: librbd, kvm, async io hang
Reducing the amount of data fio is testing with produces different behaviour. The default is size=4g. Below about siz... Chris Dunlop
03:01 PM Bug #3286: librbd, kvm, async io hang
Err, I read that second part backwards - the hanging case is skipping the page cache, so an strace of the fio in the ... Josh Durgin
02:54 PM Bug #3286: librbd, kvm, async io hang
So we'll need to dig deeper to figure out exactly what is hanging. There's a couple general possibilities - a deadloc... Josh Durgin
06:41 PM Bug #3285: rbd map w/ cephx
Aha! That did it, thanks.
After installing cbc.ko (already had aes):...
Chris Dunlop
05:44 PM Bug #3285: rbd map w/ cephx
The only path I'm seeing that could lead to that error is if the kernel wasn't compiled with support for "cbc(aes)" w... Josh Durgin
03:57 PM Bug #3285: rbd map w/ cephx
Augh. Sorry, I should have noticed that before:... Chris Dunlop
02:37 PM Bug #3285: rbd map w/ cephx
After you the map fails, is there anything added to dmesg? Josh Durgin
04:32 PM rgw Bug #2527: RGW may return 409 Conflict when deleting a bucket
Recent log: ubuntu@teuthology:/a/teuthology-2012-10-10_02:00:02-regression-testing-master-basic/5604 Tamilarasi muthamizhan
04:29 PM Bug #3275: Monitors unable to recover after network line card replacement
Their cluster is currently running:
ceph version 0.48.1argonaut-9-g657ca11 (commit:657ca118a7658617b9117311d9ee1cb...
JuanJose Galvez
04:20 PM Bug #3292: osd crash in handle_osd_ping
Logs: ubuntu@teuthology:/a/teuthology-2012-10-10_00:00:05-regression-next-testing-basic/5460... Tamilarasi muthamizhan
04:14 PM Bug #3292 (Resolved): osd crash in handle_osd_ping
Logs:ubuntu@teuthology:/a/teuthology-2012-10-10_00:00:05-regression-next-testing-basic/5458
2012-10-10 01:54:05.66...
Tamilarasi muthamizhan
02:07 PM Linux kernel client Bug #3291 (Resolved): rbd: BUG in bio from rbd_rq_fn
From #ceph, with linux 3.6.1:... Josh Durgin
01:47 PM Bug #3290 (Resolved): libceph: unable to handle osdmap
tziOm on IRC reported getting this 'corrupt' osdmap in dmesg when trying to map an rbd image with vanilla linux 3.6.1... Josh Durgin
01:21 PM CephFS Feature #3289 (Resolved): ceph-fuse: somehow exert pressure on the VFS to remove dentries from th...
We think this is the cause of some ballooning caches in ceph-fuse and the MDS that have been reported. The VFS will h... Greg Farnum
01:12 PM rbd Bug #2933: rbd: bio_pair leak in bio_chain_clone()
I'm sorry I haven't been updating this. Zhao has provided
several patches that aim to solve this problem. The late...
Alex Elder
11:49 AM Feature #3288 (Resolved): docs: document the chooseleaf command in crush
choose is ok, but can cause issues with the default crushmap. chooseleaf should usually be used instead. Josh Durgin
11:14 AM Bug #3271: kernel BUG() on unmount after export/unexport of ceph using kclient

I've pushed code to Teuthology repo implementing Feature #3245. So now this can be reproduced using the following ...
David Zafman
10:50 AM Bug #3287 (Resolved): OSD dies when using zfs
OSD dies during startup for a new ceph fs when backed by zfs Mike Lowe
02:05 AM Linux kernel client Bug #2573: libceph: many "socket closed" messages
Hi Alex,
looking for annoying messages I got aware of this ticket... we are currently running:
Linux fcmsnode1 3....
Oliver Francke

10/10/2012

10:41 PM CephFS Feature #3245 (Resolved): qa: test knfsd reexport in teuthology
David Zafman
08:20 PM Bug #3286: librbd, kvm, async io hang
Yes, it works when using a raw image file. Chris Dunlop
08:14 PM Bug #3286: librbd, kvm, async io hang
All the IOs in the log completed successfully, so I'm not sure what's causing these hangs. Does this work with the sa... Josh Durgin
06:55 PM Bug #3286: librbd, kvm, async io hang
... Chris Dunlop
06:14 PM Bug #3286: librbd, kvm, async io hang
Is your cluster entirely healthy?
If so, could you add :debug_rbd=20:debug_ms=1:log_file=/path/writeable/by/qemu.l...
Josh Durgin
06:06 PM Bug #3286 (Resolved): librbd, kvm, async io hang
Fio hangs in a linux-2.6.32 vm on librbd when using direct and libaio, with ceph at commit:c8721b956.
Libvirt conf...
Chris Dunlop
06:48 PM Bug #3285: rbd map w/ cephx
self-compiled 3.4.12 Chris Dunlop
06:28 PM Bug #3285: rbd map w/ cephx
What kernel version are you using? That error is coming directly from the kernel rbd module. Josh Durgin
06:17 PM Bug #3285: rbd map w/ cephx
Attached output of "strace -o /tmp/trace -s 60 -f rbd map test1" Chris Dunlop
06:07 PM Bug #3285: rbd map w/ cephx
Can you post the output of 'strace -f rbd map test1'? Josh Durgin
05:53 PM Bug #3285: rbd map w/ cephx
I just tested again, ensuring a complete ceph restart on all devices after setting auth cephx. The mapping failed as ... Chris Dunlop
05:34 PM Bug #3285: rbd map w/ cephx
Did you perhaps add the cephx option after enabling your monitors? If you tell the client to use cephx but it's not e... Greg Farnum
05:30 PM Bug #3285 (Closed): rbd map w/ cephx
"rbd map" with "auth supported = cephx" doesn't work (using ceph at commit c8721b956):... Chris Dunlop
05:45 PM rbd Feature #3281 (Resolved): rbd: dont do usage on error
Dan Mick
05:42 PM rbd Feature #3281: rbd: dont do usage on error
commit:0ec45201f51b44e3d34a04e4680125cdfd1f181e Dan Mick
11:26 AM rbd Feature #3281: rbd: dont do usage on error
Dan Mick
09:12 AM rbd Feature #3281 (Resolved): rbd: dont do usage on error
Sage Weil
03:57 PM Bug #3284 (Resolved): logrotate script broken
Commit 32a6394 introduced shell scripting errors.
The attached patch fixes these.
Chris Dunlop
03:20 PM rbd Bug #2933: rbd: bio_pair leak in bio_chain_clone()
Assigning this to me so I can find it again.
Alex Elder
03:18 PM rgw Documentation #3278 (Resolved): doc: Document the need to run "radosgw temp remove" regularly
Added page: radosgw/purge-temp/index.html John Wilkins
07:59 AM rgw Documentation #3278: doc: Document the need to run "radosgw temp remove" regularly
The docs should note that this is required for argonaut but not 0.52 or later. Sage Weil
02:02 PM rbd Feature #3283 (Duplicate): rbd: calc-utilization or similar
rbd command to scan blocks comprising an image and see how many bytes it's consuming. this could be coarse (count ex... Sage Weil
12:54 PM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
I looked at rbd_merge_bvec() this morning. I believe that,
given how it's used, it should be preventing callers fro...
Alex Elder
11:40 AM Bug #2683 (In Progress): ceph-fuse: crash during fsstress
Recent logs: ubuntu@teuthology:/a/teuthology-2012-10-06_04:00:05-regression-stable-master-basic/3157
2012-10-07T06...
Tamilarasi muthamizhan
09:53 AM Linux kernel client Feature #3282 (Resolved): add new 'performance' kernel gitbuilder
no lockdep
disable most debugging crap
no dynamic_printk (which apparently does not play well with kdb)
Sage Weil

10/09/2012

09:53 PM rbd Bug #3204 (Resolved): rbd client kernel panic when osd connection is lost
This problem has been resolved. The last thing I added
describes the commits that address it, and they're now
in t...
Alex Elder
08:59 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
There are two commits that fixed the problem--one on the read side
and one on the write side. They may require othe...
Alex Elder
09:52 PM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
Also, given what I last wrote, I no longer think it's related
to the bio_chain_clone() problem mentioned earlier.
...
Alex Elder
09:47 PM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
... Alex Elder
06:27 PM CephFS Feature #3245 (Fix Under Review): qa: test knfsd reexport in teuthology
Pushed to the wip-nfs-dzafman branch of the teuthology repo. David Zafman
01:36 PM CephFS Feature #3245 (In Progress): qa: test knfsd reexport in teuthology
Sage Weil
06:24 PM rbd Feature #3280 (Resolved): fsx: test flatten when cloning
Periodically flatten some clones during a run. At the end of the run each clone could be flattened after being verifi... Josh Durgin
02:42 PM rbd Feature #3205: doc: pool permissions for cloning use-case
Sage Weil
02:40 PM rbd Documentation #2992: doc: RBD parent/child snapshot
Sage Weil
02:00 PM devops Feature #2548 (Resolved): ceph-disk-activate: take mount options from ceph.conf
In stable as commit 4aa13c6d228b9256870e02a57496f00a75e17349.
In master as commit ab84949da204da4a94473f353b4b0f3c...
Anonymous
01:59 PM devops Feature #2549 (Resolved): ceph-disk-prepare: take fstype, mkfs and mount options from ceph.conf
In stable as commits cf2e3b15cea99cc100169e7a67103676d8990aee and da86e4e43676ceaf0b39635954489c297de88025.
In mas...
Anonymous
01:37 PM Bug #3279 (Resolved): mon/caps: cap comparison in get-or-create is based on a string literal
If you run ceph auth get-or-create and your caps string differs by a space, then it returns an error:... Greg Farnum
01:36 PM CephFS Feature #3242 (In Progress): samba: push plugin upstream
Sage Weil
01:35 PM Bug #3276 (Resolved): auth: keyring comma-separated locations not being parsed
Sage Weil
08:29 AM Bug #3276 (Fix Under Review): auth: keyring comma-separated locations not being parsed
Fix on branch 'mon-3276-fix' commit '760ad6adf576b72a483d6c161e48c2738bc0ea08' Joao Eduardo Luis
08:25 AM Bug #3276 (Resolved): auth: keyring comma-separated locations not being parsed
The default keyring location is a set of comma-separated files.
The KeyRing class will try to open a single file s...
Joao Eduardo Luis
12:55 PM Bug #2563: leveldb corruption
Just got another report of this on the list.
This user has enabled btrfs' lzo compression, and I believe btrfs compr...
Greg Farnum
12:12 PM Documentation #3264 (Resolved): docs: put osd and client usage restrictions in quick-start docs
John Wilkins
10:06 AM rgw Documentation #3278 (Resolved): doc: Document the need to run "radosgw temp remove" regularly
Anonymous
09:57 AM CephFS Feature #3248 (Resolved): libcephfs: move testceph.cc tests into gtest framework
Sage Weil
09:10 AM CephFS Feature #3248 (Fix Under Review): libcephfs: move testceph.cc tests into gtest framework
The wip-gtest-libcephfs branch contains the libcephfs tests now part of the gtest framework Sam Lang
09:37 AM devops Feature #3277 (Rejected): chef: support multiple clusters on same nodes ("ceph" vs other)
Anonymous
09:13 AM CephFS Bug #2285 (Resolved): libcephfs: failure with empty name components
Fix merged to master Sam Lang
05:44 AM Bug #3275: Monitors unable to recover after network line card replacement
On which version was this triggered on? Joao Eduardo Luis

10/08/2012

07:01 PM Bug #3270: osd crash during rbd test run

Dan Mick
06:59 PM Bug #3270: osd crash during rbd test run
Caught another instance of something similar; attached is the coredump and ceph-osd binary Dan Mick
02:28 PM Bug #3270: osd crash during rbd test run
Recent logs: ubuntu@teuthology:/a/teuthology-2012-10-06_00:00:05-regression-next-testing-basic/2901
0> 2012-1...
Tamilarasi muthamizhan
05:00 PM Bug #3275: Monitors unable to recover after network line card replacement
I'm contacting Kyle about the log and will upload a copy to this bug as soon as I get it. JuanJose Galvez
04:59 PM Bug #3275 (Can't reproduce): Monitors unable to recover after network line card replacement
Roughly around the time that several line cards were replaced the ceph monitors stopped working and were not able to ... JuanJose Galvez
04:15 PM rgw Feature #3274 (Resolved): rgw: RESTful admin api for user admin
Design and implement RESTful admin api. Yehuda Sadeh
04:13 PM rgw Feature #3114 (Resolved): rgw: support cache-control header field
Done, commit:f4a0b2d9260a4523745875e3977a8a1ef9dc5e2e. Yehuda Sadeh
03:45 PM rgw Feature #3225 (Resolved): rgw: stub for get bucket logging
Done, commit:d4725c06ab8fb965b2979a381340d3edbb1ed10d. Yehuda Sadeh
03:44 PM rgw Bug #3234 (Resolved): rgw: missing content-type, etag when listing bucket on some objects
Yehuda Sadeh
03:44 PM rgw Bug #3234: rgw: missing content-type, etag when listing bucket on some objects
Fixed merged to master, commit:c0b537063dda9c2069b280a9d54d2259ff0d62e0. Yehuda Sadeh
03:34 PM Feature #3273 (Need More Info): mon: simple dm-crypt key management
ability to (easily, painlessly) slip dm-crypt underneath ceph-osd file systems.
use monitor or separate keyserver ...
Sage Weil
02:54 PM rgw Feature #3216 (Resolved): rgw: RESTful usage api
Merged at commit:a780d47fbbd8fcb81397ab14e28066fd322e3b12 Yehuda Sadeh
02:16 PM Bug #3041: ceph manager down during osd recovery
ubuntu@teuthology:/a/teuthology-2012-10-06_00:00:05-regression-next-testing-basic/2831 Tamilarasi muthamizhan
10:36 AM Documentation #3264 (In Progress): docs: put osd and client usage restrictions in quick-start docs
John Wilkins
10:32 AM rbd Subtask #3272 (Rejected): send/receive rbd snapshots
It would be great to be able to send and receive incremental RBD snapshots as is done in zfs.
zfs send -i snap1 .....
Pavel Arbuzov

10/05/2012

06:15 PM Bug #3271: kernel BUG() on unmount after export/unexport of ceph using kclient

By default our kernels don't drop into kdb. Here is the logging before and after the BUG() after executing ceph/sr...
David Zafman
03:37 PM Bug #3271 (Resolved): kernel BUG() on unmount after export/unexport of ceph using kclient
plana78:
$ sudo exportfs '*:/tmp/cephtest/mnt.1' -o fsid=1234,rw,no_root_squash
NFS client
$ sudo mount plana78:...
David Zafman
02:01 PM Bug #3061: osd crash during shutdown
Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-30_19:00:06-regression-master-testing-gcov/33616 Tamilarasi muthamizhan
01:56 PM Bug #3209: rbd_fsx_cache_writethrough test failure in the nightlies
Logs: ubuntu@teuthology:/a/teuthology-2012-09-30_19:00:06-regression-master-testing-gcov/33620 Tamilarasi muthamizhan
08:20 AM CephFS Bug #3267: Multiple active MDSes stall when listing freshly created files
Greg Farnum wrote:
> I'll try and take a look at this, but multi-MDS setups are known to be pretty unstable at this ...
Stan Schwertly

10/04/2012

11:16 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
Hi Alex,
Since kernel 3.6 was releases a few days ago (https://lkml.org/lkml/2012/9/30/152) , what would be ...
Christian Yung-Tai Huang
05:11 PM rbd Feature #2788 (Resolved): rbd: add -l to rbd ls, show snapshots
Sage Weil
05:05 PM Bug #3209: rbd_fsx_cache_writethrough test failure in the nightlies
... Tamilarasi muthamizhan
05:04 PM Bug #3209: rbd_fsx_cache_writethrough test failure in the nightlies
recent logs:ubuntu@teuthology:/a/teuthology-2012-10-01_19:00:09-regression-master-testing-gcov/212 Tamilarasi muthamizhan
04:29 PM Bug #3270 (Duplicate): osd crash during rbd test run
Logs: ubuntu@teuthology:/a/teuthology-2012-10-03_19:00:11-regression-master-testing-gcov/1570... Tamilarasi muthamizhan
02:46 PM Bug #3269 (Closed): nightly failure-kclient_workunit_suites_fsstress
Logs: ubuntu@teuthology:/a/teuthology-2012-10-03_19:00:11-regression-master-testing-gcov/1453... Tamilarasi muthamizhan
02:26 PM Bug #3268 (Rejected): osd: localize reads handling is incorrect
Found by code inspection when changing the way MOSDOp->rmw_flags is populated.
The only place that checks for CEPH_O...
Josh Durgin
01:03 PM CephFS Bug #3267: Multiple active MDSes stall when listing freshly created files
I'll try and take a look at this, but multi-MDS setups are known to be pretty unstable at this point. Have you tried ... Greg Farnum
11:54 AM CephFS Bug #3267 (Closed): Multiple active MDSes stall when listing freshly created files
The output from ceph-debugpack can be found at the following location: [[http://cumulonim.biz/mds.tar.gz]] We were ru... Stan Schwertly
10:52 AM CephFS Bug #3266 (Resolved): "ceph mds tell 0 dumpcache /etc/passwd" is not cool
This might have plenty of security implications. Anonymous
10:43 AM rgw Feature #771 (In Progress): rgw: POST
Sage Weil
10:33 AM CephFS Feature #3245: qa: test knfsd reexport in teuthology
* knfsd.py task to export an existing client
* nfs.py task to mount another client
so you can do something like
...
Sage Weil
10:21 AM rgw Tasks #3265: test Amanda against RGW
http://pastebin.com/L005BSQ7 Greg Farnum
09:58 AM rgw Tasks #3265: test Amanda against RGW
... Greg Farnum
09:51 AM rgw Tasks #3265 (Closed): test Amanda against RGW
There were reports in irc that Amanda Backup causes problems with RGW. We should test it against a gateway and see if... Greg Farnum
09:48 AM Documentation #3264: docs: put osd and client usage restrictions in quick-start docs
Tommi Virtanen wrote:
> ... because of #3076
Anonymous
09:47 AM Documentation #3264: docs: put osd and client usage restrictions in quick-start docs
... because of http://tracker.newdream.net/issues/3076 Anonymous
09:45 AM Documentation #3264 (Resolved): docs: put osd and client usage restrictions in quick-start docs
Users following the quick-start guide might naturally try to reduce the number of machines they're using and use a ke... Greg Farnum
09:22 AM Feature #3263 (Resolved): mon: 'mon report' needs to include crush map
commit a6ba95fec06778ad12705fc081c83ebe3ddddbfd on master Joao Eduardo Luis
08:25 AM Feature #3263 (Resolved): mon: 'mon report' needs to include crush map
Sage Weil
04:45 AM Bug #3252: mon: mixed mon cluster of argonaut + master eats cpu on new code
Looked like this was caused by a couple of lingering bugs on mon/Elector.cc.
One would not clean up the list of mo...
Joao Eduardo Luis

10/03/2012

07:57 PM Linux kernel client Bug #2754 (Resolved): cephfs show_location produces kernel "divide error: 0000 [#1]" when run aga...
A suite of tests including these fixes completed without
error. As I said, I did not specifically test this
proble...
Alex Elder
07:53 PM Linux kernel client Bug #3112 (Resolved): ceph 32 bit kernel client issue with file size more than 4GB.
Sage ran a test using the current testing branch, which
includes this fix. The tests he ran completed without
erro...
Alex Elder
05:04 PM rbd Feature #3262 (Resolved): librbd: use 'assert exists' operation to guard cloned writes
Sage Weil
04:56 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
I wonder if there's any chance this also affects http://tracker.newdream.net/issues/2937 ? Dan Mick
04:45 PM rbd Feature #2788 (In Progress): rbd: add -l to rbd ls, show snapshots
Dan Mick
04:24 PM Bug #3260 (Resolved): ceph monitor crashed : In function 'void PaxosService::propose_pending()
Sage Weil
02:48 PM Bug #3260 (Resolved): ceph monitor crashed : In function 'void PaxosService::propose_pending()
Logs: /a/teuthology-2012-10-02_19:00:06-regression-master-testing-gcov/731... Tamilarasi muthamizhan
03:57 PM CephFS Bug #3261: mds crashes in EMetaBlob::replay
aww. i had debug ms = 20 in my ceph.conf. sorry.
the new one is attached
Tobias Florek
03:49 PM CephFS Bug #3261 (Need More Info): mds crashes in EMetaBlob::replay
can you put 'debug mds = 20' in the ceph.conf, restart ceph-mds, and then attach the resulting log (assuming it crash... Sage Weil
03:46 PM CephFS Bug #3261 (Rejected): mds crashes in EMetaBlob::replay
while testing cephfs using the debian wheezy packages on a fairly large volume (2TB) i ran into random unreproduceabl... Tobias Florek
12:59 PM devops Feature #3259 (Rejected): ceph-disk-prepare: don't re-prepare a disk if it has the right partitio...
Anonymous
12:53 PM devops Feature #3258 (Rejected): ceph-disk-prepare: give pretty error if sgdisk not found
Anonymous
12:53 PM devops Feature #3257 (Rejected): ceph-disk-prepare: give pretty error if mkfs.$foo not found
Anonymous
12:52 PM devops Feature #3256 (Resolved): ceph-disk-prepare: abort if target disk has any partition mounted
Anonymous
10:47 AM Feature #3192 (Resolved): osd: coverity clean
Sage Weil
10:47 AM Feature #3194 (Resolved): common: coverity clean
Sage Weil
09:14 AM devops Feature #3255: ceph-disk: allow prepare without activate (for spares)
This is enough to trigger hotplug:
sudo sgdisk --typecode=1:4fbd7e29-9d25-41b8-afd0-062c0ceff05d /dev/vdb
That'...
Anonymous
09:12 AM devops Feature #3255 (Resolved): ceph-disk: allow prepare without activate (for spares)
Anonymous
08:07 AM Bug #3252: mon: mixed mon cluster of argonaut + master eats cpu on new code
it could be that in the burnupi environment one of the clients only has one of the mon addresses in its config, and i... Sage Weil
08:03 AM Bug #3252: mon: mixed mon cluster of argonaut + master eats cpu on new code
Not sure how we reach this state, but with the debug levels cranked up, the leader (say, mon.a), shows the following ... Joao Eduardo Luis
07:33 AM Bug #3252: mon: mixed mon cluster of argonaut + master eats cpu on new code
Was finally able to reproduce it. The trick mainly relies on not specifying the cluster's monmap during --mkfs.
St...
Joao Eduardo Luis
03:03 AM Bug #3252: mon: mixed mon cluster of argonaut + master eats cpu on new code
This hasn't been verified while running all three monitors on the same machine with the default configuration generat... Joao Eduardo Luis
05:09 AM CephFS Bug #3254 (Resolved): mds: Replica inode's parent snaprealms are not open
When an inode with snaprealm get replicated, its parent snaprealms are not open on the dest MDS. This is problematic ... Zheng Yan

10/02/2012

04:48 PM Bug #3253 (Won't Fix): ceph-disk-activate: support clusters not named "ceph"
When activating an osd data disk, read mount options from ceph.conf, mount the disk, read the cluster uuid, find the ... Anonymous
04:27 PM Messengers Bug #3251 (Resolved): 32-bit Debian: ceph dies in assert(inq == &local_queue)
commit:67a201d64b5b4f1a8effd677dd932680144265d6
Dan Mick
03:00 PM Messengers Bug #3251 (Resolved): 32-bit Debian: ceph dies in assert(inq == &local_queue)
32-bit installation was breaking (while testing with ./vstart.sh, in the ceph
tool itself) with an assert failure on...
Dan Mick
03:48 PM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
Reviewed, committed. Waiting for a test run to complete
before marking this one resolved.
Alex Elder
08:48 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
Patch to fix this is out for review. Alex Elder
08:18 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
(Note--the problem writing files from the 32-bit environment has
been resolved. More info here: http://tracker.newd...
Alex Elder
03:33 PM Bug #3252 (Resolved): mon: mixed mon cluster of argonaut + master eats cpu on new code
3 mon cluster, 2 running argonaut, one upgraded to current next. the new one eats cpu and spams log with... Sage Weil
11:23 AM Bug #3250 (Resolved): debs don't include Contents-<arch>.gz
We're building the debs without Contents-<arch>.gz; this breaks at least
apt-file (which caches package contents for...
Dan Mick
11:09 AM Feature #3249 (New): Efficient access in distributed data centers
As I understand it, in Ceph one can cluster storage nodes, but otherwise every node is essentially identical.
This...
Matthias Urlichs
08:47 AM Linux kernel client Bug #3187 (Resolved): ceph fs: crash/hang on 32-bit architecture
This has been resolved. The problem was exactly the thing I pointed out.
The fix is upstream now:
5ce765a5 ...
Alex Elder
05:49 AM CephFS Feature #3248 (In Progress): libcephfs: move testceph.cc tests into gtest framework
Sam Lang
05:39 AM Linux kernel client Bug #2754: cephfs show_location produces kernel "divide error: 0000 [#1]" when run against a dire...
Reviewed, just waiting for a nightly test to complete before
marking this resolved. I did NOT explicitly test the f...
Alex Elder

10/01/2012

09:51 PM CephFS Feature #3248 (Resolved): libcephfs: move testceph.cc tests into gtest framework
Sage Weil
09:45 PM Feature #3200 (Resolved): os: coverity clean
Sage Weil
06:56 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
Hi Alex,
could you let us know which commits we should apply in order to address this issue? we would like to app...
Christian Yung-Tai Huang
11:49 AM rbd Bug #3204: rbd client kernel panic when osd connection is lost
There already was a fix available that addressed the
write case. I put it in originally as a cleanup but
it turns ...
Alex Elder
05:08 PM rbd Feature #3239 (In Progress): librbd: make striping fancier
Sage Weil
09:58 AM rbd Feature #3239 (Resolved): librbd: make striping fancier
Sage Weil
03:24 PM Linux kernel client Bug #2754: cephfs show_location produces kernel "divide error: 0000 [#1]" when run against a dire...
Sage posted four patches for review. I just reviewed them
and am working on some final testing before committing th...
Alex Elder
01:07 PM Documentation #3247 (Resolved): doc: Move content out of wiki, kill it with fire
Please kill http://ceph.com/wiki with extreme prejudice.
People keep reading the stale content in the wiki. Blogs ...
Anonymous
12:48 PM CephFS Feature #3245 (Resolved): qa: test knfsd reexport in teuthology
Sage Weil
12:48 PM CephFS Feature #3244 (In Progress): qa: integrate Ganesha into teuthology testing to regularly exercise ...
Sage Weil
12:48 PM CephFS Feature #3243 (Resolved): qa: test samba reexport via libcephfs vfs plugin in teuthology
Sage Weil
12:47 PM CephFS Feature #3242 (Resolved): samba: push plugin upstream
Sage Weil
12:08 PM Bug #3233 (Fix Under Review): osd: caps in master are incompatible with argonaut caps
The pool=foo support is added again in wip-osd-caps Josh Durgin
11:28 AM Feature #3241 (Resolved): qa: integration tests for mon, osd, and mds caps
This could be done as a shell script run by the client.admin user, so that it can create users with different caps an... Josh Durgin
11:13 AM Feature #3167 (Fix Under Review): auth: separate read/write capabilities for classes
Josh Durgin
11:11 AM rbd Feature #3198 (Resolved): librbd: coverity clean
Josh Durgin
09:53 AM Bug #3236 (Resolved): logrotate.d script from debian repo does not work
Sage Weil
09:51 AM Bug #3236: logrotate.d script from debian repo does not work
yes. Tobias Florek
09:03 AM Bug #3236 (Fix Under Review): logrotate.d script from debian repo does not work
Tobias Florek wrote:
> i did not read closely. you are, of course, right. the problem is, that on all debian systems...
Sage Weil
01:31 AM Bug #3236: logrotate.d script from debian repo does not work
i did not read closely. you are, of course, right. the problem is, that on all debian systems, initctl won't exist, s... Tobias Florek
05:43 AM Bug #3238 (Resolved): ceph-client: osd BUG_ON() tripped
... Alex Elder

09/30/2012

09:21 PM rgw Feature #3037 (Resolved): rgw: unit test for rgw objclass
Sage Weil
09:17 PM Bug #3236 (Need More Info): logrotate.d script from debian repo does not work
the logrotate script is an 'all of the above':... Sage Weil
06:52 AM Bug #3236 (Resolved): logrotate.d script from debian repo does not work
it uses initctl, which does only work with recentish versions of upstart, which are not in wheezy or sysvinit. Tobias Florek
09:10 PM rbd Bug #3237 (Resolved): rbd_header_race failes to reproduce a race
on current master. see /a/*/33351
Sage Weil

09/29/2012

01:23 AM Feature #3197 (Resolved): librados: coverity clean
Sage Weil

09/28/2012

05:40 PM Feature #3000 (In Progress): osd: balance recovery vs client io
Sage Weil
05:40 PM Feature #3189 (In Progress): osd: use reservations for non-backfill recovery too
Sage Weil
05:39 PM Feature #3199 (Resolved): msg: coverity clean
Sage Weil
10:33 AM Feature #3199 (Fix Under Review): msg: coverity clean
Sage Weil
07:15 AM Feature #3199 (In Progress): msg: coverity clean
Sage Weil
05:39 PM Feature #3195 (Resolved): crush: coverity clean
Sage Weil
10:34 AM Feature #3195 (Fix Under Review): crush: coverity clean
Sage Weil
07:15 AM Feature #3195 (In Progress): crush: coverity clean
Sage Weil
05:39 PM Feature #3196 (Resolved): include: coverity clean
Sage Weil
10:34 AM Feature #3196 (Fix Under Review): include: coverity clean
Sage Weil
05:38 PM Documentation #3223 (Resolved): Doc: ceph osd map not documented
Sage Weil
05:17 PM rgw Feature #3114 (Fix Under Review): rgw: support cache-control header field
Yehuda Sadeh
03:17 PM rgw Bug #2527: RGW may return 409 Conflict when deleting a bucket
Might be related to the messenger issues that were fixed in 0.52. We'll wait a bit, see if it still happens. Yehuda Sadeh
02:51 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
Well, I hit a similar issue with the previous fix in place.
However, this is in the write path. When looking at the...
Alex Elder
02:23 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
OK, I think I found it.
a410702 libceph: (re)initialize bio_iter on start of message receive
I'm going to f...
Alex Elder
07:54 AM rbd Bug #3204: rbd client kernel panic when osd connection is lost
I found another bug fix that I thought might make this go
away. This time it was this one:
3b5ede07 libceph: f...
Alex Elder
05:49 AM rbd Bug #3204: rbd client kernel panic when osd connection is lost
I have a clue to pursue this morning.
I have noticed in two crashes that the state of the connection is
not one t...
Alex Elder
02:14 PM Feature #3235 (Resolved): doc: procedure for collecting data from failed node/whatever
Sage Weil
01:32 PM rgw Bug #3211 (Resolved): rgw: crash on armhf architecture
Fixed, commit:2fddfb0a95288d39cee2e82376297ce43296743b. Yehuda Sadeh
10:39 AM Feature #3167 (In Progress): auth: separate read/write capabilities for classes
Josh Durgin
10:20 AM Bug #3233 (In Progress): osd: caps in master are incompatible with argonaut caps
Josh Durgin
10:11 AM rgw Bug #3234 (Resolved): rgw: missing content-type, etag when listing bucket on some objects
When we list the bucket and see objects that exist, but their index update is not complete, we return results accordi... Yehuda Sadeh

09/27/2012

09:01 PM Feature #3194 (In Progress): common: coverity clean
Sage Weil
08:49 PM rgw Feature #3201 (In Progress): rgw: coverity clean
Sage Weil
08:46 PM Linux kernel client Bug #2754 (Fix Under Review): cephfs show_location produces kernel "divide error: 0000 [#1]" when...
Sage Weil
05:33 PM CephFS Fix #2215: ceph-fuse does not invalidate page cache
Sam Lang
05:04 PM Bug #3233 (Resolved): osd: caps in master are incompatible with argonaut caps
In argonaut, caps could say 'allow rwx pool=blah', while releases since then have a new parser (from commit:c06e40753... Josh Durgin
04:58 PM Bug #2508 (In Progress): osdc/ObjectCacher.cc:761: void ObjectCacher::bh_write_commit(int64_t, so...
Sage Weil
04:54 PM Bug #2508 (Closed): osdc/ObjectCacher.cc:761: void ObjectCacher::bh_write_commit(int64_t, sobject...
Recent logs: ubuntu@teuthology: /a/teuthology-2012-09-24_19:00:29-regression-master-testing-gcov/29465... Tamilarasi muthamizhan
04:55 PM Bug #3226: osd: invalid capability string can allow arbitrary access
Argonaut does not have this bug. It was introduced in a post-argonaut refactoring of OSDCaps. Josh Durgin
04:44 PM Bug #3226 (Resolved): osd: invalid capability string can allow arbitrary access
Josh Durgin
04:43 PM Bug #3226: osd: invalid capability string can allow arbitrary access
Fixed by commit:303f640ce561c22a85a16b49585a22115273f790 Josh Durgin
04:55 PM Bug #3228: osd: capspec parsing is broken
Argonaut does not have this bug. It was introduced in a post-argonaut refactoring of OSDCaps. Josh Durgin
04:43 PM Bug #3228 (Resolved): osd: capspec parsing is broken
Fixed by commit:de6e0d8112d74e8c22281128dfe3085ad1ec01b0 Josh Durgin
04:45 PM Bug #3232 (Resolved): osd: 0.15 scrub stat mismatch, got 25/25 objects, 15/15 clones, 50453469/49...
Logs are in ubuntu@teuthology: /a/teuthology-2012-09-22_00:00:04-regression-next-testing-basic/27498... Tamilarasi muthamizhan
03:46 PM rbd Bug #3204: rbd client kernel panic when osd connection is lost
I am able to reproduce this problem at will using stock
Linux 3.5.4, and running "fio" in a loop, and then making
o...
Alex Elder
02:38 PM CephFS Bug #3184 (Resolved): ceph-fuse should not segfault without a keyring
Sage Weil
02:32 PM Documentation #2996 (Fix Under Review): doc: write install Ceph with RPMs doc
Sage Weil
02:31 PM Feature #2440 (Resolved): osd: understand btrfs performance
Sage Weil
02:15 PM Bug #3227 (Resolved): mon: usage for 'ceph auth' is outdated
Sage Weil
02:10 PM Bug #3230 (Resolved): man page: rados bench has wrong default write size
Sage Weil
12:44 PM Bug #3230 (Resolved): man page: rados bench has wrong default write size
... Greg Farnum
02:02 PM rbd Bug #2534 (Resolved): librbd: make sure watch is established on same header version as initial re...
Fixed by commit: e1776809031c6dad441cfb2b9fac9612720b9083 Josh Durgin
01:55 PM Feature #3203 (Resolved): Add support for Ubuntu 12.10 (Quantal Quetzal) in deb repository
Sage Weil
01:55 PM Feature #2011 (Resolved): osd: do not backfill/recover to full osds
Sage Weil
01:55 PM Feature #2826 (Resolved): osd: backfill target reservations
Sage Weil
10:49 AM rgw Feature #3225 (Fix Under Review): rgw: stub for get bucket logging
Yehuda Sadeh
10:36 AM devops Feature #3229 (Resolved): Support clean ceph-fuse fstab automounting
Right now, you need to do something like the following:... Greg Farnum

09/26/2012

05:46 PM Bug #3228 (Resolved): osd: capspec parsing is broken
It's always returning ANY, even for simple cases like "allow r". Test cases are in wip-osd-caps. Josh Durgin
05:43 PM Bug #3227 (Resolved): mon: usage for 'ceph auth' is outdated
Currently the output is:... Josh Durgin
05:06 PM CephFS Bug #3186 (Resolved): mds: mds crash on client shutdown
Sage Weil
04:27 PM Bug #3226: osd: invalid capability string can allow arbitrary access
The first commit in wip-osd-caps fixes this. Josh Durgin
04:27 PM Bug #3226 (Resolved): osd: invalid capability string can allow arbitrary access
If you use the cap osd 'allow rwx pool=bar', the parser will add the grant for 'allow rwx', but fail to parse the 'fo... Josh Durgin
04:15 PM rgw Bug #3224 (Resolved): rgw: copy object with alternative attrs does not set ETAG
commit:f031046c4d600ce0a0c2c9ad5856dc100b0901fb fixes that. Yehuda Sadeh
03:47 PM rgw Bug #3224 (Resolved): rgw: copy object with alternative attrs does not set ETAG
When using copy object with alternative attrs, ETAG is not getting set, Yehuda Sadeh
04:03 PM rgw Feature #3225 (Resolved): rgw: stub for get bucket logging
Even if not implementing bucket logging api, we can still have a simple stub that returns that bucket logging is disa... Yehuda Sadeh
01:54 PM Documentation #3084: doc: Why and when should I use latest stable kernel and not the 3.2.0 from U...
Some more details:
The kernel that comes with 12.04 is old. This ticket requests a place in the documentation (FA...
Ross Turk
10:51 AM rgw Bug #2527: RGW may return 409 Conflict when deleting a bucket
The bucket index update when doing object removal is async. That means that when we get to remove the bucket, the ind... Yehuda Sadeh
10:20 AM rgw Feature #3216 (Fix Under Review): rgw: RESTful usage api
Yehuda Sadeh
08:25 AM CephFS Bug #1677 (Resolved): mds interval_set.h: 385: FAILED assert(p->first <= start)
Sage Weil
08:20 AM Bug #2919 (Resolved): ceph kernel module looks for :/ in path, but / stripped by precise mountall
it's fixed in our testing branch, but i didn't send it to linus in for 3.6; it was late in the -rc cycle to be fixing... Sage Weil

09/25/2012

09:53 PM Bug #2919: ceph kernel module looks for :/ in path, but / stripped by precise mountall
Was this not fixed? Dan Mick
09:51 PM Documentation #3223 (Resolved): Doc: ceph osd map not documented
> $ ceph osd getmap -o osdmap
> got osdmap epoch 59
> $ osdmaptool --test-map-object dmick.rbd osdmap
> osdmaptool...
Dan Mick
06:27 PM Feature #3219 (In Progress): Run tests in valgrind on gitbuilder
It turns out this handled by teuthology tests, we just need to update the test suite to include running valgrind on c... Sam Lang
03:31 PM Feature #3219 (Resolved): Run tests in valgrind on gitbuilder
This probably means modifying the scripts to run the test binaries within valgrind and using the --valgrind option in... Sam Lang
06:24 PM CephFS Bug #2285 (Fix Under Review): libcephfs: failure with empty name components
This was probably caused by the memory error in filepath (encoded member not getting initialized), which may explain ... Sam Lang
06:22 PM CephFS Bug #3186 (Fix Under Review): mds: mds crash on client shutdown
Proposed fix in wip-3186, specifically c92e1dd7b0106fa986985d8179a8ed66c61e6709. Needs review. Sam Lang
06:20 PM CephFS Bug #3184 (Fix Under Review): ceph-fuse should not segfault without a keyring
Sam Lang
05:55 PM Documentation #3222 (Resolved): DOC: Get an Object from a Primary OSD
ceph/doc/cluster-ops/control.rst should have usage for `ceph osd map {pool-name} {object-name}`
A knowledgebase ar...
John Wilkins
05:16 PM Bug #3221 (Resolved): disconnect_session_watchers missing pg
/a/teuthology-2012-09-24_04:00:04-regression-stable-master-basic/28971/remote
--- end dump of recent events ---
2...
Samuel Just
05:03 PM rbd Documentation #3220 (Resolved): doc: more detail on QEMU+RBD page
This page could use a bit more detail on the nature of the integration and how to use it. For example, why would som... Ross Turk
03:00 PM Bug #3002: ceph-authtool: --print does not work
commit 4e40a7849442d633c9801639fbdaf27a97f39888
Author: Tommi Virtanen <tv@inktank.com>
Date: 2012-08-02 13:02:04...
Anonymous
02:19 PM RADOS Documentation #3218 (Closed): Doc: osdmaptool manpage out of date with code *and* usage
Dan Mick
01:31 PM Feature #3191 (In Progress): client: coverity clean
Sam Lang
01:05 PM rgw Feature #3216 (In Progress): rgw: RESTful usage api
Yehuda Sadeh
01:03 PM rgw Feature #3216 (Resolved): rgw: RESTful usage api
Yehuda Sadeh
01:03 PM rgw Documentation #3217 (Closed): rgw: document RESTful usage api
Yehuda Sadeh
12:39 PM Feature #3215 (Resolved): qa: fix up system tests and add them to regression suite
These tests produce the testrados_* binaries, and their source is in src/test/system/.
There are a few tests we ha...
Josh Durgin
12:39 PM Bug #2525 (Resolved): librados: some functions are not thread-safe
Sage Weil
11:47 AM Bug #3214 (Closed): osdmaptool's usage is incomplete
--test-map-pg is shown, but not --test-map-object. In fact, there are a pile
of options missing from usage:
--du...
Dan Mick
11:18 AM Bug #3208: osd/PG.cc: 1346: FAILED assert(info.last_complete >= log.tail)
wip_last_complete_assert has a fix/explanation.
#3213 also needs to be handled in this case.
Samuel Just
11:16 AM Bug #3213 (Resolved): PG: we do not persist missing set elements not in the log
Normally the missing set is reconstructed on OSD startup by scanning the log and adding objects which are missing or ... Samuel Just
09:46 AM rbd Bug #3204: rbd client kernel panic when osd connection is lost
Here is an excerpt from the log that Chris originally provided.... Alex Elder
09:42 AM rbd Bug #3204: rbd client kernel panic when osd connection is lost
I believe have reproduced this problem (or perhaps something
like it) using stock stable kernel 3.5.4. I'm not 100% ...
Alex Elder
06:27 AM Bug #2476: osd: watch timeout depends on operations to an object
Have you made any progress with this issue? It is very annoying and breaks my CI. Is there any way to avoid or workar... Maciej Galkiewicz
02:42 AM Bug #3212: librados: failed to decode message of type 59 v1: buffer::end_of_buffer
Ok, with packages 0.51-700-g1a9c8c7-1precise from wip-3212 and back to 0.41-1ubuntu2.1 on the client, it now refuses ... John Leach

09/24/2012

08:30 PM rbd Feature #2788: rbd: add -l to rbd ls, show snapshots
Dan Mick
06:47 PM rbd Feature #3198: librbd: coverity clean
we'll see if coverity detects anything else Josh Durgin
06:12 PM rbd Feature #3198 (Fix Under Review): librbd: coverity clean
wip-rbd-coverity Josh Durgin
03:24 PM rbd Feature #3198 (In Progress): librbd: coverity clean
Josh Durgin
06:47 PM Feature #3197: librados: coverity clean
we'll see if coverity detects anything else Josh Durgin
06:12 PM Feature #3197 (Fix Under Review): librados: coverity clean
wip-rbd-coverity Josh Durgin
03:24 PM Feature #3197 (In Progress): librados: coverity clean
Josh Durgin
06:36 PM Bug #3212: librados: failed to decode message of type 59 v1: buffer::end_of_buffer
see wip-3212
we weren't encoding using the pre-v0.42 pool_stat_t.. or at least that's what it looks like, i wasn't...
Sage Weil
06:35 PM Bug #3212 (Resolved): librados: failed to decode message of type 59 v1: buffer::end_of_buffer
from ML:... Sage Weil
06:08 PM rgw Bug #3211 (Resolved): rgw: crash on armhf architecture
Ubuntu, launchpad issue #149582.... Yehuda Sadeh
05:57 PM CephFS Bug #3210 (Resolved): mds crashed and segfault at unlink_local_finish
Logs: /a/teuthology-2012-09-23_19:00:07-regression-master-testing-gcov/28428... Tamilarasi muthamizhan
05:55 PM Bug #3209 (Resolved): rbd_fsx_cache_writethrough test failure in the nightlies
Logs: ubuntu@teuthology:/a/teuthology-2012-09-23_19:00:07-regression-master-testing-gcov/28530... Tamilarasi muthamizhan
04:29 PM Bug #3208 (Resolved): osd/PG.cc: 1346: FAILED assert(info.last_complete >= log.tail)
... Sage Weil
03:22 PM rgw Feature #3207 (Resolved): qa: swift functional tests in nightly
Sage Weil
03:19 PM rgw Feature #3206 (Resolved): rgw: resolve vanity CNAMEs
Sage Weil
10:58 AM rbd Feature #3205 (New): doc: pool permissions for cloning use-case
Sage Weil
10:50 AM rbd Feature #2326 (Resolved): krbd: use new class interfaces, new image format
Sage Weil
10:00 AM CephFS Bug #3184: ceph-fuse should not segfault without a keyring
client->init() wasn't getting checked for error. Proposed fix in wip-3184 branch. Sam Lang
09:21 AM rbd Bug #3204 (Resolved): rbd client kernel panic when osd connection is lost
From ML:... Sage Weil
 

Also available in: Atom