Activity
From 08/25/2012 to 09/23/2012
09/23/2012
- 11:58 PM Revision 153fb3bd (ceph): doc: Adds package deps to list in README
- The README provides a list of packages needed to successfully
configure ceph. uuid-dev and libatomic-ops-dev are mis... - 10:07 PM Documentation #2968 (In Progress): doc: complete architecture section
- 10:03 PM Documentation #2998 (Can't reproduce): doc: validate install docs on ubuntu server
- I could use additional information. I have tried installing and re-installing Ubuntu 12.04 followed by Ceph. I did no...
- 10:00 PM Documentation #155 (Resolved): document ceph auth
- This can be closed now, as we have had several iterations. There will be some new additions with a high level overvie...
- 09:58 PM Documentation #3185 (Resolved): doc: Differences from POSIX is useful content
- Added it back in, but to the dev section as suggested.
- 05:47 PM Feature #3203: Add support for Ubuntu 12.10 (Quantal Quetzal) in deb repository
- Gitbuilder and release scripts
- 03:58 PM Feature #3203 (Resolved): Add support for Ubuntu 12.10 (Quantal Quetzal) in deb repository
- Quantal Quetzal is in beta now, it can't be used with the current repository.
Kernel 3.5 will be great for btrfs.
... - 03:03 PM Revision 30748f36 (ceph): fix lock held when returning to user space typo
- 09:42 AM Feature #3202 (Resolved): tools: coverity clean
- 09:42 AM rgw Feature #3201 (Resolved): rgw: coverity clean
- 09:42 AM Feature #3200 (Resolved): os: coverity clean
- 09:41 AM Feature #3199 (Resolved): msg: coverity clean
- 09:39 AM rbd Feature #3198 (Resolved): librbd: coverity clean
- 09:39 AM Feature #3197 (Resolved): librados: coverity clean
- 09:39 AM Feature #3196 (Resolved): include: coverity clean
- 09:39 AM Feature #3195 (Resolved): crush: coverity clean
- 09:38 AM Feature #3194 (Resolved): common: coverity clean
- 09:38 AM Feature #3193 (Resolved): cls: coverity clean
- 09:26 AM Feature #3192 (Resolved): osd: coverity clean
- 09:25 AM Feature #3191 (Resolved): client: coverity clean
- 09:23 AM rbd Feature #3190 (Resolved): qa: test fencing
- 09:18 AM Feature #3053 (Duplicate): osd: (de)prioritize recovery
- 09:18 AM Feature #3189 (Resolved): osd: use reservations for non-backfill recovery too
- 09:11 AM Fix #3188 (New): osd: close read hole
- client and now-marked-down osd with old map may continue to read.
solution probably goes something like this:
*... - 08:50 AM Bug #3142: osd: crash induced by fsx workload
- heap corruption? this hardly narrows it down, but from ubuntu@teuthology:/a/teuthology-2012-09-22_19:00:05-regressio...
09/22/2012
- 09:16 AM Bug #3142: osd: crash induced by fsx workload
- ubuntu@teuthology:/a/teuthology-2012-09-21_19:00:08-regression-master-testing-gcov/27383
- 05:36 AM Revision e32ee375 (ceph): librados: cleanup: use Mutex::Locker
- Signed-off-by: Sage Weil <sage@inktank.com>
- 05:36 AM Revision d1de1f12 (ceph): librados: protect lookup_pool, get_pool_name with lock
- Fixes: #2525
Signed-off-by: Sage Weil <sage@inktank.com> - 04:05 AM Revision 720a3017 (ceph): mon: tolerate no session on no_reply()
- We remove the Session* in ms_handle_reset().
Signed-off-by: Sage Weil <sage@inktank.com> - 01:11 AM Revision a794c936 (ceph): ReplicatedPG: set op return version to pg version on ENOENT
- This will be useful for detecting read-delete races in
radosgw.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 12:55 AM Revision 53b18e3c (ceph): osd: catch decoding errors from client ops
- Check for decode errors in client ops and return -EINVAL.
Fixes: #2500
Signed-off-by: Sage Weil <sage@inktank.com> - 12:39 AM Revision ef29d90e (ceph): osd: some whitespace
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:38 AM Revision a73777a9 (ceph): osd: return -EPERM on insufficient caps
- Send a failure to the client instead of dropping the request on the floor.
Fixes: #3066
Signed-off-by: Sage Weil <sa... - 12:16 AM Revision a09153b6 (ceph): Allow scheduled jobs to use different teuthology branches
- teuthology-[schedule|suite] get a parameter to specify the branch,
to put the job in a branch-specific queue. Workers...
09/21/2012
- 11:46 PM Revision 57bb434d (ceph): Fix errors found by pyflakes
- A bunch of unused imports and variables.
- 11:21 PM Bug #2536 (Can't reproduce): librados crashed while getting stat of an object
- 11:20 PM Bug #2525 (Fix Under Review): librados: some functions are not thread-safe
- 11:08 PM CephFS Bug #1677 (Fix Under Review): mds interval_set.h: 385: FAILED assert(p->first <= start)
- wip-mds
- 10:39 PM CephFS Bug #3111 (Need More Info): Multiple MDS inevitable appear assert
- Can you describe how you're causing the crash in more detail so we can reproduce it here? What commands are you runn...
- 10:37 PM CephFS Bug #1775 (Resolved): mds startup: _replay journaler got error -22, aborting, possible regresion?
- chalking this up to a msgr failure due to one of the zillions of bugs we've fixed in the last few months.
- 10:36 PM CephFS Bug #2596 (Can't reproduce): mds: spinning on restart
- 10:35 PM CephFS Bug #2288: libcephfs: setxattr returns EEXIST following removexattr
- the problem is that it's not looking at the projected xattrs when it should. the logically simplest thing would be t...
- 10:32 PM CephFS Bug #1748 (Can't reproduce): mds segfault CDir::project_fnode
- 09:44 PM Feature #3046 (Resolved): mon: factor osd failure reporters into heartbeat grace
- 09:44 PM Feature #3047 (Resolved): mon: apply heartbeat grace adjustment to down_out_interval
- 09:44 PM Feature #3043 (Resolved): mon: track osd laggy rate/interval
- 09:44 PM Feature #3045 (Resolved): mon: factor osd laggy interval into heartbeat grace
- 06:46 PM CephFS Bug #2778: uclient: failed asserts if mounting with a non-existent directory
- As of last week, I cannot reproduce either, but I didn't look for a specific commit that fixed it.
- 04:49 PM CephFS Bug #2778 (Can't reproduce): uclient: failed asserts if mounting with a non-existent directory
- Was the error code from ceph_mount() getting checked? I'm not able to reproduce this with either fuse or the followi...
- 05:45 PM Bug #1975 (Won't Fix): btrfs: EINVAL on snap create
- 05:44 PM Bug #2673: ReplicatedPG::prepare_transaction: don't crash on empty ops
- this should be part of the librados api tests
- 05:43 PM Bug #2687 (Resolved): FileStore crashes when "osd_journal_size" is larger than the filesystem
- this has called fallocate(2) for a while now.
- 05:17 PM Revision f13eaf4c (ceph): msg/Accepter: fix race in accepter shutdown
- We want to avoid a race like:
- entry() starts, populates pfd with listen_sd, gets past !done check
- stop() does sh... - 05:17 PM Revision 3b4f754a (ceph): msg/Accepter: assert that listen_sd always >= 0
- The listen_sd is set by bind(); we should never call stop() when it is
is still negative.
Signed-off-by: Sage Weil <... - 04:53 PM Revision 0c7637d6 (ceph): rgw: prepare_update_index should not error on system bucket
- Should just return true. This way we don't need higher level
functions to be aware of system buckets. Also, don't use... - 04:29 PM CephFS Bug #2285: libcephfs: failure with empty name components
- I tested this by working around #3186, first creating the file and then in a separate test running ceph_open as the o...
- 08:02 AM CephFS Bug #2285: libcephfs: failure with empty name components
- I suspect that differs from his failure because the paths are already cached on the client. Try creating the file, u...
- 07:50 AM CephFS Bug #2285: libcephfs: failure with empty name components
- I noticed that the above test doesn't have O_CREAT, not sure if the file already existed in previous tests. The foll...
- 03:47 PM Linux kernel client Bug #3187: ceph fs: crash/hang on 32-bit architecture
- OK, mounting with the "nocrc" mount option makes the hang
when writing problem go away. The fix is pretty easy, and... - 03:34 PM Linux kernel client Bug #3187: ceph fs: crash/hang on 32-bit architecture
- After some digging, I'm pretty sure the workqueue is from the
ceph messenger, because I have pretty good confidence ... - 02:55 PM Linux kernel client Bug #3187: ceph fs: crash/hang on 32-bit architecture
- Below is the stack dump taken from gdb that I've been seeing
on the 32-bit system running inside a VM. Note this li... - 01:45 PM Linux kernel client Bug #3187 (Resolved): ceph fs: crash/hang on 32-bit architecture
- I was hitting this while attempting to write files on a 32-bit system
running inside a VM, trying to reproduce bug 3... - 03:38 PM Revision 6a2be388 (ceph): smoke: add cls unit tests in validator
- 03:36 PM Revision 8eaaddcf (ceph): rados: run class unit tests through validators
- 03:36 PM Revision 31a2a83c (ceph): move rgw tasks to separate suite
- 11:31 AM CephFS Bug #3186: mds: mds crash on client shutdown
- The getattr segfault is actually from the second mount call in the client test.
- 11:00 AM CephFS Bug #3186: mds: mds crash on client shutdown
- Verified that this is only an issue with an empty component in the path (related to #2285).
- 10:52 AM CephFS Bug #3186: mds: mds crash on client shutdown
- can you reproduce with debug mds = 20 ?
- 09:53 AM CephFS Bug #3186: mds: mds crash on client shutdown
- Just verified the same error occurs with one mds.
- 09:44 AM CephFS Bug #3186 (Resolved): mds: mds crash on client shutdown
- With this libcephfs test: https://gist.github.com/3762534
The first mds crashes during shutdown. The ceph cluster... - 09:03 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- I created a program to write out patterned data to a file,
and in a separate read mode, verify the data in a file
c... - 05:32 AM Documentation #3185 (Resolved): doc: Differences from POSIX is useful content
- removing appendix/differences-from-posix.rst is a bad idea i think. perhaps put it in dev/ or something? people want ...
09/20/2012
- 11:49 PM Revision 9aa467df (ceph): ceph-object-corpus: update with v0.52 objects
- These were generated a few days prior to v0.52 release, but well after
freeze.
Signed-off-by: Sage Weil <sage@inktan... - 11:11 PM Revision a69593b8 (ceph): Merge remote-tracking branch 'upstream/wip-osd-qlock'
- 10:40 PM Revision 675e9f31 (ceph): rados: test all rados classes
- 10:40 PM Revision 58ad4dd2 (ceph): uuid: include unistd to make encode testing work
- The expanded encode/decode macros need getpid(2) and close(2).
Signed-off-by: Sage Weil <sage@inktank.com> - 10:40 PM Revision cbfb0566 (ceph): test_cls_rbd has moved
- 10:12 PM Revision 0c07607f (ceph): qa: move all rados class unit tests into qa/workunits/cls
- This includes the existing rbd one; teh suite is getting fixed to match.
See #3097.
Signed-off-by: Sage Weil <sage@... - 10:11 PM Revision 420baa90 (ceph): Makefile: rename test_rados_api_cls_lock -> test_cls_lock
- Keep this consistent with the other rados classes
Signed-off-by: Sage Weil <sage@inktank.com> - 09:25 PM Revision e975fec6 (ceph): Merge branch 'wip-vstartfixes'
- Reviewed-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage... - 09:22 PM Revision 69743c12 (ceph): vstart.sh: Alternative fix for vstart.sh -n
- The previous fix (0f7c516f3e) breaks osd startup with -k. This one
from dmick just tells the ceph-mon which keyring ... - 08:47 PM Revision c9e81fdb (ceph): Makefile.am: librdb_fsx test needs math library on some platforms
- 06:54 PM Revision 5d2c59e0 (ceph): Merge remote-tracking branch 'gh/wip-1957'
- Reviewed-by: Sage Weil <sage@inktank.com>
- 06:51 PM Revision 93cd73f1 (ceph): doc: link to full 0.48.2 changelog
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:09 PM Revision 8276fa29 (ceph): mds: Clean up finished two phase commits
- When handling master request with slaves, the mds could crash
after receiving all slaves' commit acknowledgement, but... - 05:58 PM Revision 4d395535 (ceph): cfuse: Define CEPH_INO_DOTDOT (3) for top-level parentdir
- Defines the macro CEPH_INO_DOTDOT (to 3) and uses it as the top-level
parent directory (..) inode number. The value ... - 05:12 PM Revision 671c4c32 (ceph): ceph_common.sh: 'hostname -s' (instead of cut ...)
- This will also match up with the instructions in the docs.
Signed-off-by: Sage Weil <sage@inktank.com> - 04:42 PM Revision 51d58159 (ceph): doc: v0.48.2argonaut release notes
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:41 PM Feature #3097 (Resolved): create a teuthology task to run class unitests
- 02:06 PM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- I have set up a VM running a 32-bit kernel. It reports
via arch(1) that it is an i686 architecture. It is not
run... - 12:56 PM CephFS Bug #3184 (Resolved): ceph-fuse should not segfault without a keyring
- Saw this at a customer on-site — it segfaults if it can't find the keyring.
- 11:54 AM CephFS Bug #1957 (Resolved): ceph-fuse: have "." and ".." entries consistently
- 11:16 AM Bug #3182: No JSON object could be decoded - failure in the nightly run
- the rados bench process hasn't started up yet when we try the admin socket... it's a problem with teh test script
- 08:04 AM rbd Bug #3151 (Resolved): krbd: possible circular locking dependency (sysfs_lock and ctl_mutex) (test...
- This has been resolved. The culprit was the committed patch
entitled "rbd: expand lock protection in rbd_add()" whi... - 12:47 AM Revision fd205da1 (ceph): cfuse: Add the parent entry (..) for a top-level readdir
- In the lowlevel fuse api, the current (.) and parent (..) entries
must be added manually in a readdir call. For the ... - 12:05 AM Revision db94347d (ceph): Merge branch 'wip-mon'
- Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
09/19/2012
- 11:42 PM Revision 0395df31 (ceph): ignore 'lock held when returning to user space' from btrfs sb_internal ...
- 11:28 PM Revision d5a048ce (ceph): doc: Removed legacy doc. Not needed.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:28 PM Revision 9fa57208 (ceph): doc: Removed legacy doc. Not needed.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:27 PM Revision 64198f8a (ceph): doc: Removed legacy doc. Not needed.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:26 PM Revision d50aab84 (ceph): doc: Added enabling Swift access to rgw config.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:25 PM Revision 49580635 (ceph): doc; Added troubleshooting to index.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:25 PM Revision a83f5038 (ceph): doc: Copied contents of rgw troubleshooting over to the new ops section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:24 PM Revision 280aeaf0 (ceph): doc: Reverted so that we don't force yes or non-interactive.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:23 PM Revision 9c8061fe (ceph): doc: Removed legacy usage.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:22 PM Revision b3651dac (ceph): doc: Cleanup, spell check, grammar check mostly.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:21 PM Revision b311a408 (ceph): doc: Updating the index to remove legacy and uneeded entries.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:34 PM Revision 5e9158eb (ceph): Fix description for --nodaemon
- Signed-off-by: Sam Lang <sam.lang@inktank.com>
- 09:34 PM Revision 0a83cb99 (ceph): mon: clean up recovered_leader() checks
- Assert they are called only once per machine per election epoch. Fix the
recovered_peon() caller to do that.
Signed... - 09:34 PM Revision 50e7251d (ceph): Abort on failure
- Signed-off-by: Sam Lang <sam.lang@inktank.com>
- 09:33 PM Revision 31f430a9 (ceph): Fixup usage to reflect options available
- Signed-off-by: Sam Lang <sam.lang@inktank.com>
- 08:57 PM Revision faddb80c (ceph): Swap current dir (.) with CEPH_BIN for OOT builds
- With out-of-tree builds, vstart.sh needs CEPH_BIN to be set, and
needs to look for init-ceph in CEPH_BIN rather than ... - 08:57 PM Revision 0f7c516f (ceph): Move keyring option to global section
- Using vstart.sh -n uses ceph-authtool to generate the keyring file
in ./keyring. The vstart.sh script then writes ou... - 06:53 PM Revision 259fffbe (ceph): mon: require MON_GV protocol feature
- Require the MON_GV feature when
- we see the ondisk feature is set on bootup
- we enable the ondisk feature
This ... - 06:53 PM Revision 414cd1b5 (ceph): mon: do not issue global versions if quorum does not support the feature
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:53 PM Revision 879ce01a (ceph): mon: move setting of ondisk GV feature into helper
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:53 PM Revision 3baae15d (ceph): mon: set new incompat GV feature when paxos stabilizes for the first time
- This is a marker that future versions will use to know whether they can
safely convert the monitor data to the new fo... - 06:49 PM Revision db04ce46 (ceph): mon: make MRoute encoding backwards-compatible
- If the target as the NULLROUTE feature, use a new encoding that explicitly
indicates whether a message follows. If t... - 06:38 PM CephFS Bug #1957: ceph-fuse: have "." and ".." entries consistently
- It looks like the trick is to use a fake inode number when filling in the parent (..) entry. I chose 2, as 1 is alre...
- 06:35 PM Revision 3e02b2fa (ceph): v0.48.2argonaut
- 04:23 PM Bug #3182 (Can't reproduce): No JSON object could be decoded - failure in the nightly run
- Logs: ubuntu@teuthology:/a/teuthology-2012-09-19_00:00:13-regression-next-testing-basic/25484...
- 11:30 AM Fix #3180 (Rejected): use of strerror() for possibly-negative return values
- System calls and similar routines throughout Ceph use negative constant return codes
to indicate error; strerror() i... - 06:36 AM Revision e068bd7c (ceph): rbd/copy.sh: fix typo
- Or maybe it was a spello, or a thinko, or something. In any case
I'm pretty sure Josh intended to call the function ... - 03:51 AM Revision ed43d4de (ceph): rbd/copy.sh: fix typo
- Or maybe it was a spello, or a thinko, or something. In any case
I'm pretty sure Josh intended to call the function ... - 01:51 AM Bug #2075: osd: recover_got assert
- Hi Sage,
as announced on the list, find the logfile attached. Don't know if it's exactly the same, at least almost...
09/18/2012
- 11:49 PM Revision 471acda4 (ceph): Merge remote-tracking branch 'gh/next'
- 11:31 PM Revision a101e491 (ceph): replace tab with spaces
- Signed-off-by: Sam Lang <sam.lang@inktank.com>
- 11:06 PM Revision f5306597 (ceph): Merge remote branch 'origin/wip-librbd-locking'
- Conflicts:
qa/workunits/rbd/copy.sh
Reviewed-by: Sage Weil <sage.weil@inktank.com> - 10:56 PM Revision 5ce4d70e (ceph): fix error on teardown failing to unmount /mnt
- 10:51 PM Revision 7a3d1e66 (ceph): librbd: bump version
- This marks the availability of the cloning and locking functions.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 10:45 PM Revision 855dff62 (ceph): cls_rbd: remove locking methods
- These are unnecessary now that librbd is using the generic cls_lock.
Fixes: #2951
Signed-off-by: Josh Durgin <josh.d... - 10:45 PM Revision 8f2a0d91 (ceph): rbd: add locking commands
- The locker (entity_name_t) will be different each time the rbd
command line tool is run, so 'lock remove' is always b... - 10:43 PM Revision b66ef430 (ceph): qa: update rbd tests and runner
- * no longer need to wait for watch timeout since #2948 was fixed
* use --format 2 instead of --new-format
* add test_... - 10:43 PM Revision 3a9e6650 (ceph): librbd: use generic cls_lock instead of cls_rbd's locking
- Update the librbd locking api to make more sense:
* Add an optional tag to shared locking
* only make shared vs exc... - 10:42 PM Revision 69ee9afa (ceph): cls_lock_client: add ObjectOperation-based get_lock_info
- This will be used by librbd to grab lock info along with
the rest of its header information in a single request.
Sig... - 10:40 PM Revision 6dcbbbb6 (ceph): cls_lock_types: add missing include
- msg_types defines entity-related types used here.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 10:40 PM Revision 67bbcf2c (ceph): cls_lock_client: return error when decoding fails
- Library code shouldn't be using cerr either.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 10:39 PM Revision d1252ea2 (ceph): cls_lock_client: fix indentation
- Add indentation settings to header, and reindent.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 10:39 PM Revision bf2e4892 (ceph): cls_lock_client: change modified reference parameters to pointers
- This makes it clear which parameters are modified,
as our style guide states.
Signed-off-by: Josh Durgin <josh.durgi... - 10:36 PM Revision 2dca3a86 (ceph): cls_lock_client: clean up reference parameters
- These should all be const. The remaining reference parameters
will be converted to pointers in another commit.
Signe... - 10:36 PM Revision e71fdc75 (ceph): cls_lock: fix some spacing
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 10:35 PM Revision b69a9599 (ceph): cls_lock: specify librados namespace explicitly
- librados namespace was not specified, hence required including
source files to add using namespace. This fixes it.
S... - 10:28 PM Revision 55673bab (ceph): radosgw-admin: fix cli test
- Signed-off-by: Sage Weil <sage@inktank.com>
- 10:21 PM Revision 3372f147 (ceph): rbd: only open the destination pool for import
- Otherwise importing into another pool when the default pool, rbd,
doesn't exist results in an error trying to open th... - 10:21 PM Revision ad2ba8e6 (ceph): qa: test args for rbd import
- Make sure that --pool/--dest-pool and --image/--dest all work
interchangeably.
Signed-off-by: Josh Durgin <josh.durg... - 10:21 PM Revision d14a31d3 (ceph): rbd: make --pool/--image args easier to understand for import
- There's no need to set the default pool in set_pool_image_name - this
is done later, in a way that doesn't ignore --p... - 10:21 PM Revision a583a605 (ceph): librbd, cls_rbd: close snapshot creation race with old format
- If two clients created a snapshot at the same time, the one with the
higher snapshot id might be created first, so th... - 10:19 PM Revision a4833bb2 (ceph): librbd: fix delete[]
- CID 716902: Non-array delete for scalars (DELETE_ARRAY)
At (15): Deleting array variable "buf" with non-array delete ... - 10:19 PM Revision 3401f004 (ceph): doc: clarify rbd man page (esp. layering)
- * a clone's size can't be overridden
* note which commands require format 2
* clarify details of copy
* add examples ... - 10:19 PM Revision 582001eb (ceph): rbd: add --format option
- This chooses whether to use the original (supported by krbd)
or the new (supports layering) format.
Signed-off-by: J... - 10:18 PM Revision a1124193 (ceph): librbd: prevent racing clone and snap unprotect
- If the following sequence of events occured,
a clone could be created of an unprotected snapshot:
1. A: begin clone ... - 10:18 PM Revision e85a2383 (ceph): rbd: add "children" command, update cli test files
- Fixes: #2720
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com> - 10:18 PM Revision bd940584 (ceph): librbd: add {rbd_}list_children() methods
- These iterate over all pools and check for children of a
particular snapshot.
Signed-off-by: Josh Durgin <josh.durgi... - 09:39 PM Revision 2ad62d52 (ceph): mon: decay laggy calculations over time
- Add a configurable halflife for the laggy probability and duration and
apply it at the time those values are used to ... - 09:39 PM Revision 5499778f (ceph): mon: apply grace period scaling to mon_osd_down_out_interval
- Scale the down/out interval the same way we do the heartbeat grace, so that
we give laggy osds a bit longer to recove... - 09:39 PM Revision be503915 (ceph): mon: add tunable to control laggy probability weighting. simplify deco...
- Default to .3. Setting to 0 effectively turns this off.
Also make OSDMap::osd_xinfo_t decode into a float to simplif... - 09:39 PM Revision f6b2f79c (ceph): mon: make heartbeat grace and down out interval scaling optional
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:39 PM Revision d328a28c (ceph): mon: send 'null' reply to requests we won't reply to
- This is a no-op if the client was talking to us, but in the forwarded
request case will clean up the request state (a... - 09:39 PM Revision 3eb7341a (ceph): mon: no_reply() to failure messages we don't reply to
- This makes use clean up request state when requests have been forwarded.
Signed-off-by: Sage Weil <sage@inktank.com> - 09:39 PM Revision 7952c359 (ceph): mon: locally apply osd heartbeat grace to failure checks
- Aggregate the failure reports into a single mon 'failed_since' value (the
max, currently), and wait until we have exc... - 09:39 PM Revision a3e8ed1e (ceph): mon: reply to all reporters when an osd is failed
- Track the latest report message for each reporter. When the osd is
eventually marked failed, send map updates to the... - 09:39 PM Revision 09b251cd (ceph): mon: clean up osd failure logging
- Debug log when we get a report, info log when we actual fail the osd.
Signed-off-by: Sage Weil <sage@inktank.com> - 09:39 PM Revision 3f51d316 (ceph): mon: check failures in tick
- Currently we only trigger a failure on receipt of a failure report. Move
the checks into a helper and check during t... - 09:39 PM Revision adf0fe6a (ceph): mon: scale heartbeat grace based on laggy probability, interval
- If, based on historical behavior, an observed osd failure is likely to be
due to unresponsiveness and not the daemon ... - 09:39 PM Revision abd2ae74 (ceph): mon: factor reporter lagginess into grace adjustment
- Use reporters as a proxy for laggy subclusters within the overall cluster.
See #3046.
Signed-off-by: Sage Weil <sage... - 09:38 PM Revision 4f1792d7 (ceph): osd: include failed_for in MOSDFailure reports
- The monitor will need this to dynamically adjust the heartbeat grace.
Closes: #3044
Signed-off-by: Sage Weil <sage@i... - 09:38 PM Revision b64641c3 (ceph): osd: include boot_epoch in MOSDBoot
- This will let the monitor infer whether we were wrongly marked down or
the daemon restarted.
Signed-off-by: Sage Wei... - 09:38 PM Revision e9f051ef (ceph): osdmap: include osd_xinfo_t to track laggy probabilities, timestamps
- Track information about laggy probabilities for each OSD. That is, the
probability that if it is marked down it is b... - 09:38 PM Revision 66f31c10 (ceph): mon: adjust or decay laggy probabilities on osd boot
- On each osd boot, determine whether the osd was laggy (wrongly marked down)
or newly booted. Either update the laggy... - 09:38 PM Revision e06818be (ceph): mon: refactor osd failure report tracking
- - use structs to track allegedly failed nodes, and reports against them.
- use methods to handle report, and failure ... - 09:30 PM Revision 6f5c4a91 (ceph): PG: need pg lock to read osdmap_ref
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 09:27 PM Revision 59f85874 (ceph): OSD: in enqueue_op, cannot read *pg since we don't hold the pg lock
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 09:26 PM Revision 78a322b2 (ceph): OSD: remove some commented out pg->unlock calls
- Signed-off-by: Samuel Just <sam.just@inktank.com>
- 09:17 PM Revision f418172e (ceph): Merge branch 'master' of https://github.com/ceph/ceph
- 09:16 PM Revision ed18eea4 (ceph): :doc: Changed rados.gateway to radosgw.gateway. Start with /etc/init.d.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:16 PM Revision 900e4ceb (ceph): workqueue: allow thread pool size to be adjusted dynamically
- Allow thread pool sizes to be adjusted on the fly by telling the
ThreadPool which config option to monitor. Add some... - 09:08 PM rbd Bug #3151: krbd: possible circular locking dependency (sysfs_lock and ctl_mutex) (testing branch)
- Strike that. I have now learned that the "copy.sh" script
was buggy (which I've now fixed) and have been unable to
... - 08:29 PM rbd Bug #3151: krbd: possible circular locking dependency (sysfs_lock and ctl_mutex) (testing branch)
- FINALLY. After going through a bunch of different tests
to try to narrow it down, I have found that simply running
... - 08:55 PM Revision 5e095ec8 (ceph): :doc: Trimmed up the stack diagram and fixed a few hyperlink refs.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:51 PM Revision 808e2b23 (ceph): update doc to be ceph-fuse task (instead of cfuse)
- 08:45 PM Revision 343e6079 (ceph): cls_rgw: if stats drop below zero, set them to zero
- This complements fix for #3127. This is only a band aid
solution for argonaut, the real solution fixes the original
i... - 08:44 PM Revision 53508306 (ceph): :doc: Modified hostname to hostname -s.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:43 PM Revision 7c178be1 (ceph): :doc: Modified hostname to hostname -s.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:43 PM Revision 5936ded8 (ceph): :doc: Modified hostname to hostname -s.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:30 PM Revision 92be908c (ceph): cls_rgw: change scoping of suggested changes vars
- Fixes: #3127
Bad variable scoping made it so that specific variables
weren't initialized between suggested changes it... - 08:24 PM Revision 9b628efe (ceph): Merge branch 'wip-3127'
- Conflicts:
src/rgw/rgw_rados.cc
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:20 PM Revision fa93ea11 (ceph): rgw: fix unused warning message
- Remove unused variable.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:13 PM Revision 496c5d45 (ceph): rgw: fix uninitialized var in error message
- Used the wrong variable in error message.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:08 PM Revision 6bc1ef0a (ceph): rgw, cls_rgw: a tool to rebuild bucket stats
- radosgw-admin bucket check [--fix] --bucket=<bucket>
The command will dump the existing bucket header stats,
and the... - 08:02 PM Revision 11b80ae7 (ceph): :doc: Changed title to RBD from RADOS. Clearer.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:01 PM Revision 9829039e (ceph): :doc: Made notice to stop i/o a warning. Initial format clean up.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:00 PM Revision 84c50dd8 (ceph): :doc: Added discussion and stack diagram. Initial format cleanup.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:00 PM Revision 171fcd81 (ceph): :doc: Remove legacy usage. Initial format cleanup.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:57 PM Revision b12c1477 (ceph): cls_rgw: change scoping of suggested changes vars
- Fixes: #3127
Bad variable scoping made it so that specific variables
weren't initialized between suggested changes it... - 07:57 PM Revision 6a054ea9 (ceph): rgw: set exists flag when suggesting existing bucket entry
- We weren't setting the 'exists' flag on the bucket entry,
so we ended up not updating the index correctly.
Signed-of... - 07:57 PM Revision 0f82a393 (ceph): rgw: fix suggested object size
- We can't just stat the head object, we need to go through
get_obj_state() to get obj attributes.
Signed-off-by: Yehu... - 07:57 PM Revision 31672284 (ceph): cls_rgw: update index stats based on correct category
- only refer to the on-disk category if the object existed.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:57 PM Revision 15e97d73 (ceph): cls_rgw: complete_op() only skip update if epoch is non zero
- An update shouldn't be skipped if epoch is zero. We'd see a zero
epoch if we tried to read an object and it didn't ex... - 07:57 PM Revision 08e9fc02 (ceph): rgw: obj_stat() returns object version (epoch), use it
- We now pass the object version returned by obj_stat. We use that
epoch for setting the object version through the ind... - 07:57 PM Revision c11e05f5 (ceph): test/rgw: add a test for index suggest
- test the bucket index suggest api
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:55 PM Revision 5b7e60d0 (ceph): Merge branch 'master' of https://github.com/ceph/ceph
- 07:55 PM Revision 12175386 (ceph): cls_rgw: configurable tag timeout
- we'll be using that for bucket index unitest
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:50 PM Revision 64b9a898 (ceph): test/rgw: add bucket index unitest
- Tests objects creation, removal, prepare, complete
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:43 PM Revision 519ed247 (ceph): cpeh.spec.in: Don't package boto_tool.
- 07:27 PM Revision 6ff23bb9 (ceph): cls_rgw: client api for bucket index init
- create a client api function for bucket index init.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:27 PM Revision b03adda4 (ceph): cls_rgw: add client api calls for index suggest
- This is needed for unitest creation.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:20 PM Revision 1bf345ec (ceph): doc/config-cluster/*: update recently added/changed config options
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:41 PM Revision 91f3893f (ceph): :doc: Title change for clarity. Initial formatting cleanup.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:23 PM Revision 93323134 (ceph): Merge branch 'master' of https://github.com/ceph/ceph
- 06:23 PM Revision 88dd90e1 (ceph): :doc: Removed old rec section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:17 PM Revision 2f73c075 (ceph): docs: add radosgw internals and usage testing section
- Document the current tests that are being done on the
radosgw usage feature.
Signed-off-by: Yehuda Sadeh <yehuda@ink... - 06:15 PM Revision 3369ea3a (ceph): :doc: Notes to add data directory. Notes to address key with escape char.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:14 PM Revision 0398b9f7 (ceph): :doc: Added note to clarify use of host=shortname.
- Fixes: #3155
Signed-off-by: John Wilkins <john.wilkins@inktank.com> - 06:08 PM Revision 3d86194f (ceph): :doc: Rewrote architecture paper. Still needs some work.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:05 PM Revision fafb1951 (ceph): :doc: Removing old control.rst. New one is in ceph/docs/cluster-ops.rst
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:02 PM Revision f60ea383 (ceph): :doc: Removing old init section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:44 PM Revision de026f31 (ceph): objecter: send queued requests when we get first osdmap
- If we get our first osdmap and already have requests queued, send them.
Backported from 8d1efd1b829ae50eab7f7f4c07da... - 05:44 PM Revision d3a8497b (ceph): objecter: fix osdmap wait
- When we get a pool_op_reply, we find out which osdmap we need to wait for.
The wait_for_new_map() code was feeding th... - 05:43 PM Revision 379aa80a (ceph): objecter: use ordered map<> for tracking tids to preserve order on resend
- We are using a hash_map<> to map tids to Op*'s. In handle_osd_map(),
we will recalc_op_target() on each Op in a rand... - 05:42 PM Revision 54788d0d (ceph): rbd: force all exiting paths through main()/return
- This properly destroys objects. In the process, remove usage_exit();
also kill error-handling in set_conf_param (nev... - 05:11 PM Revision a179ccee (ceph): rbd: only open the destination pool for import
- Otherwise importing into another pool when the default pool, rbd,
doesn't exist results in an error trying to open th... - 05:10 PM Revision 4f3642c5 (ceph): rbd: only open the destination pool for import
- Otherwise importing into another pool when the default pool, rbd,
doesn't exist results in an error trying to open th... - 04:43 PM rbd Feature #2951 (Resolved): cls_rbd: remove locking methods
- 04:42 PM rbd Feature #2686 (Resolved): rbd: let users specify a usage for shared locks
- 04:42 PM rbd Feature #2952 (Resolved): librbd: use generic rados locking class
- 04:42 PM rbd Feature #2556 (Resolved): rbd tool: break image locks
- 01:45 PM rgw Bug #3127 (Resolved): rgw: negative bucket stats
- Also wip-3127-stable contains the required fix and an additional band aid solution for argonaut.
- 01:42 PM rgw Bug #3127: rgw: negative bucket stats
- merged to master at commit:9b628efee1fc31fcb1f46818af4f6b262b460840
- 07:44 AM Documentation #3174 (Resolved): Cephx with RBD/Qemu/libvirt
- With cephx turned on by default. Cephx descriptions must be solid for:
- Client admin
- Ensuring Qemu can access... - 07:41 AM Documentation #3173 (Resolved): Cephx Overview
- We need a stronger overview of Cephx for bobtail. The overview should include:
- Creating Admin Users. This will ... - 05:21 AM Revision 46883a5a (ceph): ceph.spec.in: Add new rados libraries, libcls_kvs.so and libcls_refcou...
- 03:07 AM Bug #3172 (Resolved): ceph::buffer::bad_alloc downloading a large object using rados
- got a 1gig object, when I get it I get a bad_alloc error.
This is on a 64bit Ubuntu 12.04.1 LTS box, with packages... - 01:23 AM Revision de930440 (ceph): remove obsync
- This now lives at http://github.com/ceph/obsync
Signed-off-by: Sage Weil <sage@inktank.com> - 01:23 AM Revision 6b0c9ffc (ceph): filejournal: do not enforce that bdev size >= osd journal size
- If the configure osd journal size is > the block device size, warn, but
do not generate an error and abort startup. ... - 01:23 AM Revision cb498235 (ceph): osd: default journal size to 1 GB
- The default is used to create or extend the journal when it is a file.
Signed-off-by: Sage Weil <sage@inktank.com>
09/17/2012
- 11:25 PM Revision f1819f05 (ceph): :doc: Changed from bullet list to section titles.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:01 PM Revision 18cf1b6d (ceph): :doc: Usage consistent with Tyler's notes.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:00 PM Revision da1a9f91 (ceph): :doc: Corrected a path.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:59 PM Revision 00e58a50 (ceph): :doc: Addresses DOCUMENTATION #3155
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:01 PM Documentation #3098 (Resolved): doc: ceph.conf host= field still confuses people
- I updated the General Settings Reference and the Ceph Configuration section to point this out explicitly.
- 05:40 PM Revision 16aca749 (ceph): ceph-disk-activate, upstart: Use "initctl emit" to start OSDs.
- This avoids an error if the daemon was running already, and is
already being done with the other services.
Signed-of... - 05:26 PM CephFS Documentation #3077 (Resolved): doc: Apparently people still don't realize we don't recommend Cep...
- Added notes to http://ceph.com/docs/master/cephfs/ and http://ceph.com/docs/master/start/quick-cephfs/
- 04:28 PM Documentation #3155 (Resolved): doc: ceph --admin-daemon ... show config
- Added a section after http://ceph.com/docs/master/config-cluster/ceph-conf/#runtime-changes.
- 08:44 AM Documentation #3155 (Resolved): doc: ceph --admin-daemon ... show config
- Seems to be completely undocumented currently.
http://thread.gmane.org/gmane.comp.file-systems.ceph.devel/8944/foc... - 03:35 PM Documentation #3099 (In Progress): doc: write install ceph+openstack with crowbar
- 03:27 PM Revision f46a5f1c (ceph): Merge branch 'wip-rbd-import'
- 02:49 PM rbd Documentation #3171 (Resolved): doc: how to setup OpenStack (Glance and Cinder|Nova) to use rbd c...
- This can probably be an addendum to the current OpenStack docs page for releases Folsom and beyond.
- 02:41 PM rbd Documentation #3170 (Closed): doc: document image locking
- Some resources:
The original discussion:
http://comments.gmane.org/gmane.comp.file-systems.ceph.devel/7094
The... - 02:27 PM rbd Bug #2597 (Resolved): Import of image from file appears to succeed, but image not present in the ...
- This was unintentionally solved by commit:fd4b294280cc2238289dbd7b37236d8a8cb10160. The --pool parameter to rbd impor...
- 02:20 PM Feature #3167 (Resolved): auth: separate read/write capabilities for classes
- Currently class operations can only be allowed in an all-or-nothing manner, with the x capability.
It would be use... - 02:11 PM Feature #3097: create a teuthology task to run class unitests
- We don't need a separate teuthology task for this, it could just be another workunit. A localfs task that sets up the...
- 02:04 PM Bug #2563: leveldb corruption
- Experiencing the same issue on a production ceph cluster.
ceph version 0.48.1argonaut (commit:a7ad701b9bd479f204... - 12:34 PM Feature #3080: rados: support asynchronous replication of some sort
- Here's an write-up of what the proprietary XenServer product does: http://blogs.citrix.com/2012/08/24/storage_xenmotion/
- 10:35 AM RADOS Documentation #3166 (Closed): doc: Explain OSD up/down, in/out: what does it mean, where does it ...
- Something like: up/down is observed by peers, spread via gossip, reaches monitors that make a decision to declare an ...
- 10:33 AM Documentation #3165 (Rejected): doc: what to do after install to validate core / rbd / radosgw / ...
- 10:32 AM Documentation #3164 (Rejected): doc: "ceph auth add" syntax: * and ... are magic
- 10:22 AM Bug #3163 (Won't Fix): doc: explain meaning of pg dump output
- 10:16 AM devops Feature #3161 (Rejected): make gcov website public, via proxy on gitbuilder.sepia.ceph.com
- 09:38 AM devops Feature #3160 (Rejected): upstart: ceph-osd job leaves disks mounted
- 09:27 AM devops Feature #3158 (Closed): downburst: add --wait based on cd-rom eject detection kludge
- 09:26 AM devops Feature #3082 (In Progress): ceph-deploy: Support deployments without a framework like Chef, Pupp...
- 09:24 AM devops Feature #3157 (Resolved): upstart: move mds scripts to ceph-mds package.
- These are currently in ceph.deb, but the ceph-mds binary is in ceph-mds. The upstart scripts should move to the packa...
- 09:07 AM Messengers Bug #3156 (Resolved): port range is hard-coded
- Right now the port range for daemons is hard-coded as 6800-6900, which means if you get more daemons on a box than ca...
09/15/2012
- 10:21 PM Revision 4251d783 (ceph): rgw: update cache interface
- RGWRados::delete_obj() was updated in commit
93218aeab7615ced131f0f1af49f6cda41b4a661, but we
failed to update the co... - 10:17 PM Revision ff2ce1cb (ceph): Merge branch 'wip-2923'
- Conflicts:
src/rgw/rgw_admin.cc
src/rgw/rgw_log.cc
src/rgw/rgw_log.h
src/rgw/rgw_main.cc
src/rgw/rgw_op.cc
src/... - 08:33 PM Revision 6756f9af (ceph): rgw-admin: add cluster info command
- A new radosgw-admin command to dump cluster info
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:33 PM Revision 7fe1f0b1 (ceph): rgw: simplify create_pool
- create_pool should only create pool. A pool is not a bucket,
so we don't need to attach any attrs to it. Also, no rea... - 08:30 PM Revision ff0f88a6 (ceph): rgw: store cluster params in a special object
- We now have a cluster root pool that should hold the
cluster params. The cluster params are now read from
this object... - 08:02 PM Revision a16fa049 (ceph): Merge branch 'wip-rgw-refcount'
- 08:01 PM Revision a150cf3c (ceph): rgw: document how refcount objclass works
- Added some explanations in the client header.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:01 PM Revision 539a8c7e (ceph): cls_refcount: add a unitest
- test cls_refcount functionality.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:01 PM Revision 93218aea (ceph): rgw_rados: adjustments for refcount objclass
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 07:56 PM Revision cfe48305 (ceph): rgw: set atomic context for copy operation src and dest
- This is required so that we handle both src and dest atomically. We
also set the prefetch flag on the src object, so ... - 07:52 PM Revision 3cd17fb9 (ceph): rgw: copy of objects using refcount
- For objects with manifest that have a tail, we
copy only the head, and the manifest, and increasing
the reference cou... - 07:51 PM Revision 20c78527 (ceph): rgw: use refcount put instead of obj delete where needed
- Tie in new refcount objclass methods.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 03:22 PM rgw Feature #2923 (Resolved): rgw: non hard-coded pool names
- Merged, commit:ff2ce1cbd868a28b95f5a53e7004718630931fbf.
This still does not provide a tool to modify pool names, ... - 01:41 PM rgw Bug #2915 (Resolved): rgw: copy of large object times out
- 01:09 PM rgw Bug #2915 (Duplicate): rgw: copy of large object times out
- merged, commit:a16fa0492eaa1f03b8676f8793eaba6fdad21c89
- 01:14 PM rgw Cleanup #3154 (Rejected): rgw: configurable auid when creating pools
- We should revise the defaults we use when we create pools, e.g. auid. Maybe we should avoid creating pools altogether?
- 11:11 AM Revision 34d8eec9 (ceph): rbd: make --pool/--image args easier to understand for import
- There's no need to set the default pool in set_pool_image_name - this
is done later, in a way that doesn't ignore --p... - 12:17 AM Revision fd4b2942 (ceph): rbd: make --pool/--image args easier to understand for import
- There's no need to set the default pool in set_pool_image_name - this
is done later, in a way that doesn't ignore --p... - 12:17 AM Revision 78d6a60c (ceph): qa: test args for rbd import
- Make sure that --pool/--dest-pool and --image/--dest all work
interchangeably.
Signed-off-by: Josh Durgin <josh.durg...
09/14/2012
- 10:04 PM Revision 78b7b02c (ceph): imported subprocess module in nuke script
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 07:31 PM Revision 9348ea0d (ceph): ceph-create-keys: Refactor to share wait_for_quorum call.
- Signed-off-by: Tommi Virtanen <tv@inktank.com>
- 07:31 PM Revision f390ee68 (ceph): ceph-create-keys: Create a bootstrap-osd key too.
- Signed-off-by: Tommi Virtanen <tv@inktank.com>
- 04:35 PM rgw Feature #3153 (New): Add --preserve-basic-acl option to obsync
- A DHO Request:
The --no-preserve-acls flag is currently used to ensure that policies that are specific to external... - 03:29 PM devops Feature #2548: ceph-disk-activate: take mount options from ceph.conf
- Defaults for XFS, from an email from Sage:
When I talked to Christoph at Linuxcon a couple weeks back he suggested... - 11:34 AM rgw Tasks #3152 (Resolved): rgw: document usage testing
09/13/2012
- 11:58 PM Revision 0bf77237 (ceph): doc: fix indentation
- Before this the list of steps was split into two lists
after the conf example.
Signed-off-by: Josh Durgin <josh.durg... - 11:38 PM rbd Bug #3151 (Resolved): krbd: possible circular locking dependency (sysfs_lock and ctl_mutex) (test...
- Detected when unmapping an image:...
- 09:31 PM Revision d27806a2 (ceph): nuke: add missing import
- 09:29 PM Revision c8c7014f (ceph): rbd: fix typo and cast to int before comparing format
- 08:54 PM Revision cf6899d8 (ceph): Adding RBD CLI test automation script, RbdLib and cephscrub.sh
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 04:28 PM rgw Bug #3127 (Fix Under Review): rgw: negative bucket stats
- There were a few issues in the mechanism that recovers from failed operations. When we write/remove an object we go t...
- 03:42 PM rbd Bug #3133: perfcounters: sigfpe when rbd caching is enabled
- The floating point exception in this case was due to "inexact result". This kind of exception is turned off by defaul...
- 03:02 PM Revision 2223ac78 (ceph): osd: introduce PG::_qlock to protect queuing
- Taking the PG::_lock when queuing each op for the worker threads can intorduce
long delays that hold up subsequent op... - 03:02 PM Revision 221fc788 (ceph): osd: drop unnecessary can_discard_request() in PG::queue_op()
- We perform the same check in PG::do_request(), and it is no longer safe to
do this at enqueue_op() time because we ar... - 03:02 PM Revision 539786a8 (ceph): osd: move permissions check into pg op thread
- The permission check examines the PG::pool struct. Instead of adding
additional locking there, just push the check i... - 02:32 PM Bug #2691: osd/ReplicatedPG.cc: 5888: FAILED assert(latest->is_update())
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-13_04:00:05-regression-stable-master-basic/22002
- 08:34 AM Bug #3135 (New): mon_host without ports leads to failure
- 12:33 AM Revision 4d333adf (ceph): mon: track when peon paxos machines activate
- Make note of when peons get leases, indicating that all paxos machines have
recovered.
Signed-off-by: Sage Weil <sag... - 12:33 AM Revision fa86f7aa (ceph): mon: move start-up feature check into static helper
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:33 AM Revision e84f2e49 (ceph): mon: move feature read/write into helpers
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:33 AM Revision ebdfd4ba (ceph): mon: refactor monitor feature checks
- Three helpers:
- legacy features (if file isn't present)
- required features
- supported features
Write out the f... - 12:33 AM Revision 14252815 (ceph): mon: share quorum features on election victory
- Inform peons what the quorum features are. Previously only the leader
knew, which was of limited utility.
This is o... - 12:33 AM Revision 247ba35a (ceph): mon/MonitorStore: store global version (gv) alongside paxos values
- Keep a parallel set of files mapping paxos states (NNN of foo/NNN) to a
global version number, stored at foo_gv/NNN. ... - 12:33 AM Revision 6a1d34ac (ceph): mon: allow all paxos instances to recover before going writeable
- Wait for all paxos instances to settle (recover, and commit any proposed
but uncommitted value) before any of them go... - 12:33 AM Revision 8619d874 (ceph): mon: provide global paxos version helpers
- For each election epoch, we wait for paxos machines to settle, and then
start allocating global paxos version numbers... - 12:33 AM Revision f4210f83 (ceph): mon: include global versions in paxos, slurp messages
- Signed-off-by: Sage Weil <sage@inktank.com>
- 12:33 AM Revision 39e51387 (ceph): mon: specify/allocate a global version at begin time
- Each time the leader proposes a new state, it allocates a new global
version number to go with it.
Signed-off-by: Sa... - 12:33 AM Revision 85ed24af (ceph): mon: share global version with proposal, collect, commits, slurp, etc.
- Share the global version through all of the paxos messages that share
data, and when slurping. This ensures that com... - 12:33 AM Revision e91b16aa (ceph): mon: fix paxos completions
- If we are already recovered, trigger completions immediately. Otherwise,
wait, and when we do recover trigger them a...
09/12/2012
- 11:37 PM Revision d697b54c (ceph): test/rados-api/aio.cc: use read operation for omap cmp test
- The error code field is not filled in on writes.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage... - 11:01 PM Bug #3142: osd: crash induced by fsx workload
- Attempting a bisect from master to stable. Using
cd ~/src/ceph/ceph
git describe
make distclean && ./do_autoge... - 02:57 PM Bug #3142: osd: crash induced by fsx workload
- ubuntu@teuthology:/var/lib/teuthworker/archive/teuthology-2012-09-11_02:00:03-regression-testing-testing-basic/20743
- 02:56 PM Bug #3142 (Resolved): osd: crash induced by fsx workload
- ...
- 09:34 PM Revision c9d4c58b (ceph): objecter: fix skipped map handling
- If we skip a map, we want to translate NO_ACTION to NEED_RESEND, but leave
POOL_DNE alone.
Backported from 2a3b7961c... - 09:14 PM Revision c1372f91 (ceph): osd/OSD.cc: Fix typo in OSD::heartbeat_check()
- The check 'p->second.last_tx > cutoff' should always be false
since last_tx is periodically updated by OSD::heartbeat... - 09:13 PM Revision e89cab6b (ceph): osd/ReplicatedPG: set truncate_seq when handling CEPH_OSD_OP_APPEND
- We need set truncate_seq when redirect the newop to CEPH_OSD_OP_WRITE,
otherwise the code handles CEPH_OSD_OP_WRITE m... - 06:31 PM Revision 055bf73d (ceph): rbd: only specify --format if not using the default
- This lets older versions that do not support --format still work with
format 1 images. - 06:00 PM Revision c73c440e (ceph): Makefile: rename 'core' -> 'base', add a few things
- Signed-off-by: Sage Weil <sage@inktank.com>
- 04:45 PM Bug #3124 (Resolved): Rados api test failure in the nightly run
- d697b54c9bed80d9231f8f20fd22245d6f9d91f3
- 02:44 PM Bug #3124: Rados api test failure in the nightly run
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-11_19:00:29-regression-master-testing-gcov/21043
- 04:40 PM Revision d8dff93a (ceph): librbd, cls_rbd: close snapshot creation race with old format
- If two clients created a snapshot at the same time, the one with the
higher snapshot id might be created first, so th... - 04:28 PM Bug #3145 (Resolved): ceph-mon: mon_host insists on commas, does not understand spaces
- Everything else seems happy with either of...
- 04:15 PM Revision 24c3cae1 (ceph): librbd, cls_rbd: close snapshot creation race with old format
- If two clients created a snapshot at the same time, the one with the
higher snapshot id might be created first, so th... - 04:11 PM Bug #3135: mon_host without ports leads to failure
- I stumbled on this again. The trigger is this:
- no mon hosts or mon addr specified
- mon finds its own ip via pu... - 10:06 AM Bug #3135 (Can't reproduce): mon_host without ports leads to failure
- I swear this failed for me when I was at a customer site in NYC. That may have been when they were still using the ol...
- 03:51 PM rgw Feature #3144 (New): Make obsync able to tolerate N failures in migrating a bucket before abortin...
- DHO Request:
Currently if any one object fails to migrate, obsync aborts the entire bucket, making it very difficu... - 03:23 PM Bug #3143 (Won't Fix): Obsync object verification takes too long
- In Summary:
obsync verification for each objects takes about 3 seconds - with a large amount of objects per bucket... - 03:07 PM Revision dfcb48d4 (ceph): Merge pull request #23 from javacruft/master
- Refactor the way that dh_makeshlibs is called to be slightly more elegant
- 02:17 PM Bug #2691: osd/ReplicatedPG.cc: 5888: FAILED assert(latest->is_update())
- ubuntu@teuthology:/a/teuthology-2012-09-12_02:00:04-regression-testing-testing-basic/21369
- 01:16 PM rgw Tasks #3118: QA: radosgw upstart: test the radosgw upstart script respawn functionality
- There's no restarting ever with sysvinit, so I really don't expect Carl to have made that sort of a mistake. The "ser...
- 11:17 AM rgw Tasks #3118: QA: radosgw upstart: test the radosgw upstart script respawn functionality
- Using ceph version 0.51-516-gdfcb48d
My cluster is configured based off the examples in the web docs.
Following the... - 09:40 AM Revision 31560ca5 (ceph): Rejig the way the shared libraries are processed so that manual postins...
09/11/2012
- 11:58 PM Revision 9f967e3a (ceph): upstart: Start mds,mon,radosgw after a reboot.
- They had no "start on" stanzas, so they didn't get started earlier.
- 11:58 PM Revision 771ca71c (ceph): upstart: Give everything a stop on stanza.
- These are all tasks, and expected to exit somewhat quickly,
but e.g. ceph-create-keys has a loop where it waits for m... - 11:42 PM Revision 60e273ad (ceph): upstart: Start mds,mon,radosgw after a reboot.
- They had no "start on" stanzas, so they didn't get started earlier.
- 11:42 PM Revision 94747651 (ceph): upstart: Give everything a stop on stanza.
- These are all tasks, and expected to exit somewhat quickly,
but e.g. ceph-create-keys has a loop where it waits for m... - 11:40 PM Revision fb6c5c3e (ceph): upstart: Use "ceph osd crush create-or-move".
- Now the weight is only set when adding the OSD to the CRUSH map for
the first time. Once it's there, it's only moved,... - 11:04 PM Revision 331bbcfb (ceph): Merge remote-tracking branch 'gh/wip-crush'
- Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
- 10:33 PM Revision ce1e7d1b (ceph): upstart: Add ceph-create-keys.conf to package.
- Signed-off-by: Tommi Virtanen <tv@inktank.com>
- 10:31 PM Revision d8cb19dd (ceph): upstart: Add ceph-create-keys.conf to package.
- Signed-off-by: Tommi Virtanen <tv@inktank.com>
- 10:24 PM Revision ced6c2c3 (ceph): :doc: Fixed typo.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:51 PM Revision 0ff22ba0 (ceph): obsync: if OrdinaryCallingFormat fails, try SubdomainCallingFormat
- This blindly tries the Subdomain calling format if the ordinary method
fails. In particular, this works around bucke... - 09:50 PM Revision de811db9 (ceph): obsync: if OrdinaryCallingFormat fails, try SubdomainCallingFormat
- This blindly tries the Subdomain calling format if the ordinary method
fails. In particular, this works around bucke... - 09:06 PM Revision ef3eab74 (ceph): Merge remote-tracking branch 'upstream/next'
- Conflicts:
src/osd/ReplicatedPG.cc - 08:37 PM Revision 656ab158 (ceph): osd: fill in user log entry last after snapdir tran
- Reorder the snapdir logic and ctx->at_version adjustments prior to filling
in the object_info_t and user_versions and... - 08:37 PM Revision 4e5283d4 (ceph): ReplicatedPG: do not start_recovery_op if we are already pushing
- Should fix bug #2761.
If we are already pushing soid, recovery_ops will only be decremented once for
all current pus... - 08:05 PM Revision a4fb9c1a (ceph): :doc: Added tunables to cruch-map.rst.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:00 PM Revision 911433fd (ceph): :doc: Removed old pg tuning. New section was added.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:59 PM Revision 9256a295 (ceph): :doc: Trimmed the old ops tree. Will remove when all porting verified.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:52 PM Revision 203ba59e (ceph): :doc: Trimmed the tree for failures/troubleshooting. RGW remains.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:51 PM Revision 662fd032 (ceph): :doc: removed. RBD now has its own section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:05 PM Revision 0817b941 (ceph): mon: make redundant osd.NNN argument optional
- Instead of 'osd crush set NNN osd.NNN weight loc...', make the second
osd.NNN option optional, and allow either NNN o... - 07:05 PM Revision dd9819e3 (ceph): doc: make note of crush usage change
- Even tho it is compatible.
Signed-off-by: Sage Weil <sage@inktank.com> - 07:05 PM Revision babef41a (ceph): doc/control.rst: add 'osd crush create-or-move ...'
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:05 PM Revision 96369913 (ceph): crush: add const string& versions of accessors
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:05 PM Revision 50c957db (ceph): crush: constify loc map arguments
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:05 PM Revision 344fef77 (ceph): mon: move loc map parsing into a helper
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:05 PM Revision e6141005 (ceph): mon: adjust number of req args for loc
- At least one loc key/value pair is required to do anything useful with
these commands.
Signed-off-by: Sage Weil <sag... - 06:35 PM Revision 01a81469 (ceph): ceph tool: add 'osd crush create-or-move ...' to help
- Signed-off-by: Sage Weil <sage@inktank.com>
- 06:26 PM Revision 44fa233b (ceph): :doc: Deleting this. Wrote a new one, but will be revised a bit soon.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:25 PM Revision 79607eed (ceph): Don't lose tracebacks of exceptions raised in a greenlet.
- Exception objects don't contain the traceback of where they were
raised from (to avoid cyclic data structures wreckin... - 06:19 PM Revision 32f30f9a (ceph): :doc: Removed old ops pool section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:19 PM Revision 0313365d (ceph): :doc: Removed old authentication section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:18 PM Revision d1053d9d (ceph): :doc: Removed old resize OSD section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:15 PM Revision ad909f3f (ceph): :doc: Removed old mon resize section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:14 PM Revision 7d881dc8 (ceph): :doc: Removed from old ops doc. Still needs to be composed though.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:13 PM Revision bf342d14 (ceph): :doc: New cluster ops section addresses the todo.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:13 PM Revision e8449895 (ceph): :doc: Removed old OSD troubleshooting. New version to be updated shortly.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:12 PM Revision 72f802c5 (ceph): :doc: Removed old monitor troubleshooting. New version to be revised sh...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:12 PM Revision fe609b7a (ceph): :doc: Removed old mds troubleshooting. Still needs to be composed.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:11 PM Revision a4733b86 (ceph): :doc: Removed old cephfs disucssion.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:10 PM Revision d4e00bce (ceph): :doc: Trimmed toctree to last bits of legacy data.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:09 PM Revision 922c59ff (ceph): :doc: Updated FAQ with a friendlier message.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:48 PM Revision b2409a2c (ceph): mon: 'osd crush create-or-move <id> <initial-weight> <loc ...>'
- Create an item in the tree with the given weight, or move it (without
touching the weight) if it is already present.
... - 05:48 PM Revision 1da73e5d (ceph): mon: fail on trailing characters after parsing numbers
- parse '8' but not '8asdf'.
Signed-off-by: Sage Weil <sage@inktank.com> - 05:48 PM Revision f1b605c0 (ceph): mon: parse '<id>' or 'osd.<id>' for 'osd crush create-or-move ...'
- Signed-off-by: Sage Weil <sage@inktank.com>
- 05:48 PM Revision f8d9f86e (ceph): osdmap: 4 decimal places for osd tree
- This makes it print out more nicely for reasonable weights.
Signed-off-by: Sage Weil <sage@inktank.com> - 05:48 PM Revision 588b263d (ceph): crush: get_item_weight[f]()
- Get the weight for an item, or at least the first instance of an item in
the tree.
Signed-off-by: Sage Weil <sage@in... - 05:48 PM Revision adedd6b6 (ceph): crush: create_or_move_item()
- Create an item if it doesn't exist, with the specified weight. If it is
already in the tree, move it, but do not adj... - 05:41 PM rgw Bug #3127: rgw: negative bucket stats
- managed to reproduce with some code changes to simulate identified scenario. The code changes:
1. Randomly skip the b... - 05:15 PM Feature #3137 (Fix Under Review): osd: make default osd_journal_size 1GB, if existing (block or f...
- wip-osd-journal
- 04:55 PM Feature #3137 (Resolved): osd: make default osd_journal_size 1GB, if existing (block or file) use...
- This avoids having a "osd journal size = 1000" in every chef/ceph-deploy etc ceph.conf
- 05:00 PM Bug #3135: mon_host without ports leads to failure
- the code appears to set it to 6789. is it easy for you to reproduce? this must be some strange interaction with the...
- 04:53 PM Bug #3135 (Resolved): mon_host without ports leads to failure
- ceph-mon should default to :6789 if port is not mentioned, however using just...
- 04:56 PM Feature #3138 (Rejected): osd: Make filestore_xattr_use_omap=true the default? Or something?
- Sam said this is not desired on btrfs, but I think the typical ceph.conf should not need to set this. Make the logic ...
- 04:54 PM Feature #3136 (Resolved): for bobtail: Set auth_supported=cephx as default
- Well, set it via the non-deprecated options, but still.
- 04:34 PM Bug #3134 (Can't reproduce): osd: hang in journal?
- ...
- 04:12 PM Bug #3052: ceph-osd --mkfs fails on ext3
- IRC user mrjack_ confirms that this (from Greg) fixes the problem:
diff --git a/src/os/FileStore.cc b/src/os/Fil... - 04:05 PM Feature #3101 (Resolved): allow "ceph osd crush set" without overwriting weight
- 03:36 PM rbd Bug #3133 (Duplicate): perfcounters: sigfpe when rbd caching is enabled
- On irc, Damien Churchill reported this backtrace from qemu-kvm with rbd caching enabled:...
- 03:32 PM Bug #2787 (Resolved): RBD test copy fails when thrashing
- I think this was fixed shortly after it was discovered.
- 02:54 PM Bug #3128 (Resolved): Obsync doesn't appear to handle permanent redirect with specified endpoint ...
- 02:36 PM Bug #3128 (Fix Under Review): Obsync doesn't appear to handle permanent redirect with specified e...
- wip-obsync
- 02:10 PM Bug #2761 (Resolved): osd: failed to recover before timeout expired
- Now in master/next 4e5283d476fc3405f30c91f1447e6059a1b52c8c
- 11:23 AM Bug #2761: osd: failed to recover before timeout expired
- Testing fix in wip_push_fix c56b8a8bde6242e8b485a5d87798edf068d22642.
recovery_op leak. - 01:11 PM Bug #2691: osd/ReplicatedPG.cc: 5888: FAILED assert(latest->is_update())
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-11_02:00:03-regression-testing-testing-basic/20743
- 11:29 AM Bug #3124: Rados api test failure in the nightly run
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-10_19:00:08-regression-master-testing-gcov/20355
- 10:48 AM Bug #2784: osd hit suicide timeout
- Recent logs: /a/teuthology-2012-09-08_04:00:03-regression-stable-master-basic/19039
- 10:44 AM Bug #3130 (Resolved): osd: crash in SimpleMessenger::reaper
- Logs: ubuntu@teuthology:/a/teuthology-2012-09-08_02:00:03-regression-testing-testing-basic/18829...
- 10:38 AM Bug #3129 (Can't reproduce): valgrind reported error on uninitialized osd
- logs: ubuntu@teuthology:/a/teuthology-2012-09-09_02:00:03-regression-testing-testing-basic/19388...
- 04:31 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- we have created two files( 4GB and 5 GB) using dd and if=/dev/urandom in our local machine for 32 kernel client issue...
- 12:38 AM Revision f64cedf4 (ceph): rbd: allow xfstests task to specify rbd image formats
- This adds the ability to specify the rbd image format to use for the
scratch and test devices for the rbd.xfstests ta... - 12:37 AM Revision 73a29cdf (ceph): rbd: allow image format to be specified
- This adds the ability to specify an rbd image format (either 1 or 2)
for an rbd image.
Signed-off-by: Alex Elder <el... - 12:28 AM Revision d51d7b34 (ceph): rgw: fix rgw_dir_suggest_changes() to set header_changed
- Apparently we weren't setting header_changed to true in the
case where we handled the CEPH_RGW_UPDATE case and cur_di...
09/10/2012
- 10:45 PM Revision b6f5d12f (ceph): changed the debug value for mds from 10 to 20
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 10:02 PM Revision 39efbbcc (ceph): Suppress valgrind error "Invalid write 8"
- Signed-off-by: tamil <tamil.muthamizhan@inktank.com>
- 08:21 PM Revision 97d8a734 (ceph): librbd: ignore -ENOENT during discard
- This is a backport of a3ad98a3eef062e9ed51dd2d1e58c593e12c9703
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 08:21 PM Revision 0e58d951 (ceph): librbd: add test for discard of nonexistent objects
- This verifies librbd properly handles ENOENT during discard.
Signed-off-by: Sage Weil <sage@inktank.com> - 07:46 PM Revision 61d705e2 (ceph): objectcacher: fix bh leak on discard
- Fixes: #2950
Signed-off-by: Sage Weil <sage@inktank.com> - 07:42 PM Revision 63c09093 (ceph): Merge branch 'wip-swift-manifest'
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 07:41 PM Revision f8c365e6 (ceph): rgw: add missing ret code check
- Also, for some reason we used cerr instead of dout so fix
that too.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 06:51 PM Feature #3101 (Fix Under Review): allow "ceph osd crush set" without overwriting weight
- 10:55 AM Feature #3101: allow "ceph osd crush set" without overwriting weight
- Sage proposed this would be "ceph osd crush create-or-move".. I think the args could be "ID INITIAL_WEIGHT VAR=VAL.."...
- 06:28 PM Revision a4bc7953 (ceph): Merge branch 'wip-3060'
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 06:24 PM Revision d6c2ded0 (ceph): radosgw-admin: update task for new usage reporting
- Usage reporting output has been modified, also use the new
--categories input param.
Signed-off-by: Yehuda Sadeh <ye... - 06:09 PM Revision be426d1b (ceph): schedule_suite.sh: try to use same branch for s3-tests.git
- If it isn't there, warn and fall back to master.
- 06:08 PM Revision 3473c2ed (ceph): s3tests: run against arbitrary branch/sha1 of s3-tests.git
- 06:08 PM Revision 617534e7 (ceph): s3tests: run against arbitrary branch/sha1 of s3-tests.git
- 05:50 PM Revision 0cfac6da (ceph): librbd: bump version
- This marks the availability of the cloning and locking functions.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 05:41 PM rgw Bug #3127: rgw: negative bucket stats
- DHO reported the following details regarding this issue:
root@peon7935:~# radosgw-admin bucket stats --uid=andfar3... - 05:11 PM rgw Bug #3127 (Resolved): rgw: negative bucket stats
- We see a case where after all the objects were removed out of a bucket (40TB of 64k objects) the bucket index stats e...
- 05:18 PM Bug #3128 (Resolved): Obsync doesn't appear to handle permanent redirect with specified endpoint ...
- It seems that accessing the bucket from http://s3.amazonaws.com/mikesbackup3or results in a PermanentRedirect error. ...
- 04:24 PM Bug #3126 (Can't reproduce): mds crashed bool CDir::check_rstats()
- Logs: ubuntu@teuthology:/a/teuthology-2012-09-09_00:00:04-regression-next-testing-basic/19242...
- 04:19 PM Bug #3125 (Resolved): Assertion Error in peer.py - failure from the nightly run
- Logs: ubuntu@teuthology:/a/teuthology-2012-09-07_19:00:06-regression-master-testing-gcov/18598...
- 04:11 PM Bug #3124 (Resolved): Rados api test failure in the nightly run
- Logs: ubuntu@teuthology:/a/teuthology-2012-09-08_19:00:04-regression-master-testing-gcov/19111...
- 12:51 PM Documentation #3123 (Closed): doc: Where to see "ceph osd tell ID bench" results (ceph -w)
- http://www.spinics.net/lists/ceph-devel/msg08789.html
- 12:38 PM rgw Feature #1712 (Resolved): rgw: support swift manifest objects
- Done, merged commit:63c090939951a8c5a6c0abb41d811f62c6fc04b2.
- 11:31 AM devops Feature #3122: chef: Use "argonaut"/"bobtail" in apt sources.list, not "stable"
- debian is now a symlink to debian-argonaut on ceph.com. docs just need to be updated.
- 11:30 AM devops Feature #3122 (Resolved): chef: Use "argonaut"/"bobtail" in apt sources.list, not "stable"
- Depends: #3083
- 11:29 AM rgw Feature #3060 (Resolved): rgw: different usage info break down
- Done, merged, commit:a4bc7953a72fa2828e2aecc1d81f41d25222e0b1. Also update teuthology to reflect new radosgw-admin ou...
- 11:22 AM Documentation #2971 (Resolved): doc: expand/complete mon settings reference
- 11:18 AM devops Feature #3121 (Rejected): chef: allow specifying exact version for "package" LWRP
- This lets an admin have stricter control over what versions of packages are installed; decreases the chances of accid...
- 11:10 AM devops Feature #3120 (Resolved): provide command "ceph-disk-list" that lists osd data/journal disks and ...
- This should be easy after #3078 and #3079.
Something like... - 10:59 AM devops Feature #3119 (Duplicate): chef: support destroying an osd
- DH cookbooks do this by setting a node attribute that maps osd.id -> desired action, one of the actions is destroy.
... - 10:44 AM rbd Feature #2686 (Fix Under Review): rbd: let users specify a usage for shared locks
- 09:57 AM devops Feature #2554: chef: open question: How do we discover what disks we should use as Ceph data disks?
- What DH does: each Chef node has an attribute that's a list of block device pairs (data, journal).
- 09:52 AM rgw Tasks #3118 (Closed): QA: radosgw upstart: test the radosgw upstart script respawn functionality
- I have a vague bug report from Carl that radosgw, using the upstart script, is not restarting properly. Let's reprodu...
- 06:47 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- I tried your dd command with /dev/random
echo foo > /mnt/ceph/foo
dd if=/dev/random of=/mnt/ceph/foo bs=1M count=... - 03:45 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- I tried with the following command to create a file with random data
dd if=/dev/random of=cephtest bs=1M count=1 s... - 03:15 AM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- Hi Sage,
I tried with dd, it shows foo. But I have tried nearly 100 more video files, but i am getting same erro...
09/09/2012
09/08/2012
- 09:44 AM Bug #3115: Cannot compile ceph 0.48.1
- Apparently this is an issue with the tarballs from github. With the tarball from the blog post this error doesn't occur.
- 09:27 AM Bug #3115 (Won't Fix): Cannot compile ceph 0.48.1
- I'm trying to compile ceph on a Fedora 15 system but that fails when I try to run autogen.sh:...
- 04:03 AM Revision d77205d8 (ceph): objecter: remove the now-unused "registering" flag on LingerOps
- Signed-off-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com> - 01:45 AM Revision 33def83d (ceph): cls_rbd: remove locking methods
- These are unnecessary now that librbd is using the generic cls_lock.
Fixes: #2951
Signed-off-by: Josh Durgin <josh.d... - 01:38 AM Revision fd1c634b (ceph): qa: update rbd tests and runner
- * no longer need to wait for watch timeout since #2948 was fixed
* use --format 2 instead of --new-format
* add test_... - 01:38 AM Revision eeaa92c4 (ceph): rbd: add locking commands
- The locker (entity_name_t) will be different each time the rbd
command line tool is run, so 'lock remove' is always b... - 01:38 AM Revision 18427ecc (ceph): librbd: use generic cls_lock instead of cls_rbd's locking
- Update the librbd locking api to make more sense:
* Add an optional tag to shared locking
* only make shared vs exc...
09/07/2012
- 11:44 PM Revision c4c4b20f (ceph): rgw: encode total usage for backwards compatibility
- Adding the encoding of the total usage allows us to
make the format change backwards compatible.
Signed-off-by: Yehu... - 11:22 PM Revision 06290f6d (ceph): rgw: configurable swift auth entry point
- Fixes: #1776
Swift auth entry point is not hard coded anymore.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 10:34 PM Revision 03136d05 (ceph): osd: fill in user log entry last after snapdir tran
- Reorder the snapdir logic and ctx->at_version adjustments prior to filling
in the object_info_t and user_versions and... - 09:34 PM Revision 928865b1 (ceph): PG: create macro for simple events
- This should make defining no-information events a bit simpler.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 07:52 PM rgw Feature #3114 (Resolved): rgw: support cache-control header field
- 07:20 PM CephFS Bug #3088 (Need More Info): NULL pointer dereference at ceph_d_prune
- 07:20 PM Linux kernel client Bug #3087 (Need More Info): Hung ceph_msg_kfree
- 07:19 PM Linux kernel client Bug #3112: ceph 32 bit kernel client issue with file size more than 4GB.
- Hi,
A real simple test is to see if a write to a file offset > 4gb inappropriately wraps to a low file offset. Ca... - 01:19 AM Linux kernel client Bug #3112 (Resolved): ceph 32 bit kernel client issue with file size more than 4GB.
- We have been using 32 bit and 64 bit ceph cluster and kernel client. we have mounted ceph cluster using 32 bit and 64...
- 07:16 PM rbd Feature #2556 (Fix Under Review): rbd tool: break image locks
- wip-librbd-locking
- 07:16 PM rbd Feature #2951 (Fix Under Review): cls_rbd: remove locking methods
- wip-librbd-locking
- 07:15 PM rbd Feature #2952 (Fix Under Review): librbd: use generic rados locking class
- wip-librbd-locking
- 04:18 PM rgw Feature #1776 (Resolved): rgw: swift auth prefix should be configurable (and optional)
- Fixed, commit:06290f6dffec33f4a9f47e4c3733f6779173f595.
- 04:03 PM Bug #2834: osd/ReplicatedPG.cc: 3577: FAILED assert(waiting_for_ack.begin()->first == repop->v)
- Recent log: ubuntu@teuthology:/a/teuthology-2012-09-07_00:00:07-regression-next-testing-basic/17906...
- 03:57 PM Bug #2834: osd/ReplicatedPG.cc: 3577: FAILED assert(waiting_for_ack.begin()->first == repop->v)
- Recent log: ubuntu@teuthology:/a/teuthology-2012-09-07_00:00:07-regression-next-testing-basic/17906
- 03:35 PM Bug #3072 (Resolved): osd/ReplicatedPG.cc: 3548: FAILED assert(waiting_for_ondisk.begin()->first ...
- commit:03136d057f0048e9cd840a6e83efedfc20969247
- 09:29 AM CephFS Documentation #3113 (Rejected): Ceph FS Options Could Use Some Additional Information
- Documentation for mounting Ceph FS doesn't reflect some available options. Also, there are disparities -- e.g., cephf...
- 04:00 AM Revision e2129ab4 (ceph): :doc: Fixed broken hyperlinks.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 03:31 AM Revision 9bbe73cc (ceph): :doc: Addresses Documentation #3096. Also added new information.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
09/06/2012
- 11:46 PM Revision 793bebb8 (ceph): rgw: rgw-admin filter usage by categories
- rgw can now filter usage output by specific op
categories.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 11:38 PM CephFS Bug #3111 (Can't reproduce): Multiple MDS inevitable appear assert
- I use two sets of equipment installed suse11sp2, deployment of CEPH double MDS active?I do LOOKUP/GETATTR/CREATE/SYML...
- 11:12 PM Revision 9d878bfb (ceph): gitignore: Remove "nbproject", no idea what that even is.
- Looks like this line was accidentally introduced in commit
af4d8db55f7268ab68ee5a7e17ac58c993528566.
Signed-off-by: ... - 09:06 PM Documentation #3096 (Resolved): doc: Hardware recommendations sort of look like ranges
- I also added a considerable amount of commentary, since people have asked for more guidance.
- 09:08 AM Documentation #3096 (Resolved): doc: Hardware recommendations sort of look like ranges
- http://ceph.com/docs/master/install/hardware-recommendations/ has entries like "2-1GB Ethernet NICs" and "1-500GB dri...
- 08:17 PM Revision ddaac56c (ceph): rgw: usage now account categories
- Instead of just keeping a flat usage info per bucket, we
now maintain a list of categories for which requests
usage i... - 06:08 PM rgw Bug #2652 (Resolved): Segmentation fault in rest-bench
- This should have been fixed.
- 06:06 PM rgw Feature #3102 (Need More Info): rgw: dedup
- 05:15 PM Revision a5b9ba77 (ceph): rgw: fix usage
- Fixes: #3085
usage was showing 'bucket info' command that never
existed.
Signed-off-by: Yehuda Sadeh <yehuda@inktank... - 04:45 PM rgw Feature #3060 (Fix Under Review): rgw: different usage info break down
- 04:22 PM Feature #3101 (Resolved): allow "ceph osd crush set" without overwriting weight
- Currently, a ceph-osd coming up gets its weight reset by src/upstart/ceph-osd.conf. This is not good. A node reboot s...
- 01:30 PM Feature #2894: cli: help command for ceph subsystems
- It also has "ceph --help", which gives a list of many commands, but not all. Remove the list of commands from --help ...
- 01:10 PM Feature #1783 (Resolved): osd: scrub incrementally across hash range using MOSDPGScan
- 01:10 PM RADOS Feature #3011 (Resolved): Remove "pool" terminology from CRUSH
- 01:09 PM rbd Feature #3100 (Resolved): rbd-fuse prototype
- 11:03 AM rgw Feature #1776 (Fix Under Review): rgw: swift auth prefix should be configurable (and optional)
- 11:01 AM Documentation #3099 (Resolved): doc: write install ceph+openstack with crowbar
- 10:59 AM Documentation #2981 (Resolved): doc: write add/remove a monitor
- 10:15 AM Documentation #3098 (Resolved): doc: ceph.conf host= field still confuses people
- I saw one group put in IP addresses in host=. I saw one person put in long hostnames (with domain).
The sysvinit i... - 10:12 AM rgw Cleanup #3085 (Resolved): rgw: remove 'bucket info' from radosgw-admin usage dump
- Fixed, commit:a5b9ba77737cbd7bf604492a40b291b2811f034f
- 09:59 AM Feature #3097 (Resolved): create a teuthology task to run class unitests
- We have some class unitests now (lock, refcount). We should be able to run these through a teuthology task.
- 08:46 AM Feature #3095 (Resolved): rbd tool resize improvements
- It might be handy if the rbd CLI tool could warn an admin when performing a resize operation that would ultimately en...
- 07:18 AM rbd Bug #3094 (Resolved): krbd: race between finding existing client and creating new one
- In rbd_get_client(), an attempt is made to reuse an existing
rbd client if there is already one with matching option... - 12:33 AM Revision 3fed44ac (ceph): :doc: Minor tweak to heading text.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:26 AM Revision 53d98292 (ceph): Merge branch 'master' of github.com:ceph/ceph
- 12:25 AM Revision 6957ae78 (ceph): :doc: Modified the index page to point to the new cluster-ops section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:24 AM Revision 4ff60c7a (ceph): :doc: Made minor changes to restructuredText headers.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:23 AM Revision 8fa81ab5 (ceph): :doc: Added comment redirecting editors to new page location.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:22 AM Revision c22ca723 (ceph): :doc: Added index/toctree page for cluster ops.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:22 AM Revision b8dfe822 (ceph): :doc: Added new auth settings to reference doc.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:21 AM Revision 199dc5d0 (ceph): :doc: Refactored and moved control.rst page.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:21 AM Revision d90fea6c (ceph): :doc: Consolidated file system recommendations.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:20 AM Revision ab020da9 (ceph): :doc: Minor syntax update.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:19 AM Revision 64b9e9f5 (ceph): :doc: restructuredText syntax corrections.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:17 AM Revision c94034a2 (ceph): :doc: Added index page. This will be refactored again soon.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:17 AM Revision 32398f34 (ceph): :doc: Refactored and moved to ceph/docs/cluster-ops/pools.rst
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:16 AM Revision c53a1104 (ceph): :doc: Removed. New section is in ceph/doc/cluster-ops/authentication.rst
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:15 AM Revision f1c869cb (ceph): :doc: Fixed heading syntax.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
09/05/2012
- 11:38 PM Revision 78c844e5 (ceph): PG: clear want_acting in choose_acting if want == acting
- Otherwise, a pg_temp from a previous peering sequence
(but not a different peering_interval) might leak through
into ... - 09:57 PM Revision 5868b00c (ceph): cls_refcount: a new class for objects refcounting
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 09:23 PM Revision d31140ce (ceph): Merge remote-tracking branch 'gh/wip_deep_scrub_final'
- 09:19 PM Revision a0b04bce (ceph): buffer: class for efficiently calculating CRC32 of >= 1 bufferlist
- Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
- 09:19 PM Revision 9013efd3 (ceph): osd: deep scrub, read file contents from disk and compare digest
- Deep scrub reads the contents of every file from the store and computes
a crc32 digest. The primary compares the dige... - 09:19 PM Revision 3fd5914c (ceph): doc: explain what scrub, deep-scrub, and repair actually do
- Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
- 09:19 PM Revision df89c963 (ceph): osd: initialize pg_log_entry_t::invalid_pool in default ctor
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:19 PM Revision a08a28d0 (ceph): pg: store scrubber state in its own object
- Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
- 09:19 PM Revision f5046798 (ceph): osd: chunky scrub, scrub PGs a chunk of objects at a time
- Chunky scrub is a more efficient scrub. It blocks writes on a subset of
objects and scrubs those, allowing writes thr... - 08:20 PM Revision 79b30543 (ceph): crush: change default type from 'pool' to 'root'
- The 'pool=default' in the default crush maps is confusing wrt rados pools.
'root' makes more sense given that we are ... - 06:28 PM Revision 24fe265c (ceph): ceph-disk-prepare: Take fsid from config file.
- Closes: #2546.
Signed-off-by: Tommi Virtanen <tv@inktank.com> - 06:28 PM Revision 471105a9 (ceph): ceph-disk-prepare: Partition and format OSD data disks automatically.
- Uses gdisk, as it seems to be the only tool that can automate GPT uuid
changes. Needs to run as root.
Adds Recommend... - 06:28 PM Revision 27372dc4 (ceph): upstart: Make ceph-osd always set the crush location.
- This used to be conditional on config having osd_crush_location set,
but with that, minimal configuration left the OS... - 06:28 PM Revision 6e34b8c1 (ceph): upstart: Make instance jobs export their cluster and id variables.
- This allows other jobs listening to Upstart "started ceph-mon" events
to see what instance started.
Signed-off-by: T... - 06:28 PM Revision 66aecb1e (ceph): config: Add a per-name default keyring to front of keyring search path.
- This lets us have e.g. /etc/ceph/ceph.client.admin.keyring that is
owned by root:admin and mode u=rw,g=r,o= without m... - 06:28 PM Revision 787a78e8 (ceph): upstart, ceph-create-keys: Make client.admin key generation automatic.
- This should help simplify Chef etc deployments. Now (when using the
Upstart jobs), when a ceph-mon is started, ceph-c... - 06:26 PM Revision 1f06b490 (ceph): upstart: fix regex
- Signed-off-by: Tommi Virtanen <tv@inktank.com>
Signed-off-by: Greg Farnum <greg@inktank.com> - 05:50 PM Revision a0c87b86 (ceph): Merge remote-tracking branch 'upstream/wip-kvstore'
- 05:13 PM Revision 8bad708e (ceph): pg: change _scrub() to take out parameters as pointers
- Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
- 05:13 PM Revision a83fb799 (ceph): FileStore: get objects whose names fall within a range
- Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
- 04:20 PM Bug #2761: osd: failed to recover before timeout expired
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-09-05_04:00:05-regression-stable-master-basic/16530
- 01:03 PM Feature #3047 (Fix Under Review): mon: apply heartbeat grace adjustment to down_out_interval
- 12:29 PM Linux kernel client Bug #3058 (Won't Fix): xfs: possible circular locking dependency detected
- 12:28 PM CephFS Bug #2444: null pointer deference in ceph_d_prune inside kvm
- Can you try...
- 12:28 PM CephFS Bug #3088: NULL pointer dereference at ceph_d_prune
- Can you try...
- 11:56 AM CephFS Bug #3088: NULL pointer dereference at ceph_d_prune
- Matt Garner wrote:
> Probably a duplicate of BUG #2444.
** Possibly - 11:55 AM CephFS Bug #3088: NULL pointer dereference at ceph_d_prune
- Probably a duplicate of BUG #2444.
- 11:52 AM CephFS Bug #3088 (Resolved): NULL pointer dereference at ceph_d_prune
- when I mount (via fstab) a specific folder
10.10.140.210:6789:/pool-hs-san-1 /mnt/ceph1-pool-hs-san-1 ceph ... - 11:03 AM Linux kernel client Bug #3087 (Can't reproduce): Hung ceph_msg_kfree
- Sep 4 00:37:00 rmi-orem-ceph1-mds1 kernel: [233040.316059] INFO: task smbd:31483 blocked for more than 120 seconds.
... - 08:52 AM Feature #3086 (Resolved): workqueue: dynamically adjust number of threads
- Add a config observer, and logic to start new threads and let extra threads die off.
- 08:32 AM Bug #2176: dependencies not checked by autoconf
- Gary, want to quickly check if configure.ac is checking these things and see if this is still a problem? IIRC Mark s...
- 08:30 AM CephFS Bug #1510 (Resolved): fsx failure on cfuse
- this has been back in the qa suite for weeks now, no problems.
- 03:31 AM Revision 195eda62 (ceph): Monitor.cc: Added include for limits.h.
- This include is needed on Centos. It seems to be included implicitly
on other platforms. - 12:17 AM Revision ca7fb838 (ceph): Merge branch 'master' of github.com:ceph/ceph
- 12:16 AM Revision 22668769 (ceph): :doc: Added a section for adding and removing monitors. Singificantly r...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 12:15 AM Revision c1ff83de (ceph): :doc: Incorporated Joao's feedback into the reference material.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
09/04/2012
- 11:48 PM Revision 0c78a03d (ceph): docs: Add CloudStack documentation
- The basic documentation about how you can use RBD with CloudStack
Signed-off-by: Wido den Hollander <wido@widodh.nl> - 11:34 PM Revision 2b8690be (ceph): :doc: Added recovering from OSD failures. Will be re-factored again soon.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:33 PM Revision 9db84be4 (ceph): doc: Added monitor failure recovery. Will be re-factored again soon.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:19 PM Revision 6dff5d56 (ceph): doc: Re-factored adding an OSD.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:18 PM Revision 32b44cb4 (ceph): doc: Minor updates.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:17 PM Revision 00dd6a63 (ceph): Merge branch 'master' of github.com:ceph/ceph
- 11:17 PM Revision d2332714 (ceph): doc: Added admonishments for Ceph FS per http://tracker.newdream.net/is...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:08 PM Revision 4b774b69 (ceph): doc: Updated to incorporate Sage's changes.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:03 PM Revision 327969ca (ceph): doc: Added anchor references.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:02 PM Revision a50f73ea (ceph): doc: Refactored the debug section to point back to reference.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 11:01 PM Revision 31d63287 (ceph): doc: Added QA reference to --valgrind option.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 10:21 PM Revision ccfd8413 (ceph): Revert "ReplicatedPG: fill in user log entry last after snapdir tran"
- This reverts commit 0aad5462eb79be0427004f2442903bb56c2057c1.
This gives us two events with the same version, and cr... - 09:37 PM Revision fd76083d (ceph): rgw: use prefetched buffer even when ofs != 0
- Beforehand we only used the prefetched buffer if read ofs
was 0.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 09:37 PM Revision 4d2a05f6 (ceph): rgw: implement user manifest (swift)
- Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
- 08:55 PM Revision 40fd6ba8 (ceph): OSD::handle_pg_stats_ack: grab pg refcount while processing pg
- If the queue refcount is the last one for the pg, the pg->put()
in the loop will destroy the pg while the lock is sti... - 08:52 PM Linux kernel client Bug #2754: cephfs show_location produces kernel "divide error: 0000 [#1]" when run against a dire...
- ceph_calc_file_object_mapping() does no divide-by-zero checking.
- 08:47 PM Revision 0aad5462 (ceph): ReplicatedPG: fill in user log entry last after snapdir tran
- The user log entry contains the request id, which will be used
by replay ops to put themselves in the correct place i... - 08:45 PM CephFS Bug #1509 (Can't reproduce): cfuse sometimes hangs after unmount
- 08:44 PM CephFS Bug #1603 (Can't reproduce): ceph-fuse crash during unmount
- 08:44 PM CephFS Bug #1806 (Can't reproduce): MDS won't start
- logs indicate an unresponsive osd.
- 08:41 PM Bug #2151 (Resolved): mon: pg creation retry is slow
- 08:40 PM CephFS Bug #1640 (Resolved): mds: failed assert(trim_to > trimming_pos)
- pretty sure this is out of order osd replies, due to #2947 and similar bugs.
- 08:09 PM rgw Feature #2923 (Fix Under Review): rgw: non hard-coded pool names
- 07:17 PM Revision 1db67c44 (ceph): PG: In Active, don't transition to WantActingChange
- want_acting is filled in during recovery completion in
order to move the newly backfilled osd into its correct
place.... - 07:17 PM Revision 46b86426 (ceph): Merge branch 'wip-msgr'
- Reviewed-by: Greg Farnum <greg@inktank.com>
- 06:54 PM Revision 035cf965 (ceph): msg/Pipe: kill useless onconnect arg
- This reduces debug output but nothing else, for know discernable reason.
Drop it.
Signed-off-by: Sage Weil <sage@ink... - 06:40 PM Revision b6ef17a1 (ceph): Merge branch 'master' of github.com:ceph/ceph
- 06:40 PM Revision 31b1c2f8 (ceph): doc: Added PG states.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:39 PM Revision ef6814d9 (ceph): doc: Promoting PG concepts into mainline docs. Redundant version still ...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:38 PM Revision 25b75939 (ceph): doc: New section on placement groups.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:37 PM Revision 3fcdbe42 (ceph): doc: Moved from configuration to operations. Updated with new info.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:37 PM Revision 9da49667 (ceph): doc: Created a more robust doc for monitoring a cluster.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:36 PM Revision 743e0b8d (ceph): doc: Consolidated start and stop.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:35 PM Revision 3e562001 (ceph): doc: Added a new "Data Placement" overview section for added continuity.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:34 PM Revision 2a1ac680 (ceph): doc: Added a new CRUSH map section. Will need to incorporate new tunabl...
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:33 PM Revision fc093f8b (ceph): doc: Moving new auth section from configuration to operations.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:29 PM Revision e09b2655 (ceph): objecter: fix osdmap wait
- When we get a pool_op_reply, we find out which osdmap we need to wait for.
The wait_for_new_map() code was feeding th... - 04:52 PM Revision f8e1f5c2 (ceph): task: die on ceph error or coredump
- This task allows ceph to signal to teuth that it should die immediately
by touching a file under /tmp/cephtest
Signe... - 04:28 PM CephFS Documentation #3077 (In Progress): doc: Apparently people still don't realize we don't recommend ...
- Added comments in Quick Start and Ceph FS sections. d233271434d1daa640273efe99ef3ee193da1de6
- 02:50 PM CephFS Documentation #3077 (Resolved): doc: Apparently people still don't realize we don't recommend Cep...
- Not sure what more we can or should do, but I run across this recently again.
- 04:18 PM rgw Cleanup #3085 (Resolved): rgw: remove 'bucket info' from radosgw-admin usage dump
- 04:13 PM Documentation #3084 (Resolved): doc: Why and when should I use latest stable kernel and not the 3...
- This will be a living document, more reasons added as time progresses.
Another similar one will probably be starte... - 04:11 PM Feature #3083 (Resolved): Provide separate APT repos for argonaut, bobtail, etc; stable would alw...
- I think we should by default make installations use the "argonaut" style name, so they don't accidentally upgrade jus...
- 03:51 PM devops Feature #3082 (Closed): ceph-deploy: Support deployments without a framework like Chef, Puppet, J...
- 03:34 PM rgw Feature #3081 (Duplicate): radosgw: asynchronous read-only replication mechanism
- See #3080 for a similar RADOS-level feature. Proposed separately for radosgw as it can offer different tradeoffs, and...
- 03:31 PM Feature #3080 (New): rados: support asynchronous replication of some sort
- Probably easiest to construct using pool-level snapshots, and computing a delta between two snapshots on the sending ...
- 03:20 PM Revision c1e000b9 (ceph): doc: Fix leftover "localhost" mention.
- Commit dd011aba90831bade3b67e99268429be10635dce changed
the conf file sample to say {hostname}, but changed the
prose... - 03:07 PM devops Feature #3079 (Resolved): support journal on any accessible block device
- 03:06 PM devops Feature #3078 (Resolved): support journal in 2nd partition on same disk as osd
- 02:22 PM Documentation #3076: doc: Explain how loopback mounts (using kclient, ceph-fuse should be immune)...
- this definitely qualifies as a faq
- 02:19 PM Documentation #3076: doc: Explain how loopback mounts (using kclient, ceph-fuse should be immune)...
- The discussion in this thread https://lkml.org/lkml/2004/7/26/68 is interesting for a more in-depth discussion of the...
- 02:00 PM Documentation #3076 (Resolved): doc: Explain how loopback mounts (using kclient, ceph-fuse should...
- Currently, consuming CephFS/RBD services provided by the same machine can lead to deadlock. People in general don't r...
- 02:08 PM rbd Cleanup #1761: krbd: make block/segment naming consistent
- I just committed the following to the ceph-client/testing
branch. It renames field "block_name" to be "object_prefi... - 10:49 AM rbd Cleanup #1761 (Resolved): krbd: make block/segment naming consistent
- 02:02 PM Bug #3071 (Resolved): osd: crash in ~ReplicatedPG
- 40fd6ba8ed9ba70c8d20a79936f53f10f2dfe839 should take care of this one.
- 01:49 PM Bug #3072: osd/ReplicatedPG.cc: 3548: FAILED assert(waiting_for_ondisk.begin()->first == repop->v)
- 0aad5462eb79be0427004f2442903bb56c2057c1 should take care of this one.
- 12:06 PM Bug #3072: osd/ReplicatedPG.cc: 3548: FAILED assert(waiting_for_ondisk.begin()->first == repop->v)
- ubuntu@teuthology:/a/sage-h/15556 and 15540
- 08:01 AM Bug #3072 (Resolved): osd/ReplicatedPG.cc: 3548: FAILED assert(waiting_for_ondisk.begin()->first ...
- ...
- 01:48 PM Bug #3025 (Resolved): WaitActingChange
- 1db67c443d84dc5d1ff53cc820fdfd4a2128b680
- 01:38 PM Feature #3045 (Fix Under Review): mon: factor osd laggy interval into heartbeat grace
- 01:38 PM Feature #3046 (Fix Under Review): mon: factor osd failure reporters into heartbeat grace
- 12:04 PM Bug #3070 (Resolved): client/mds: ms failure injection causes problems
- 11:29 AM Bug #3075 (Resolved): rados python tests occasionally hang with ms failures
- commit:e09b26555c6132ffce08b565780a39e4177cbc1c
- 10:04 AM Bug #3075: rados python tests occasionally hang with ms failures
- ...
- 09:36 AM Bug #3075 (Resolved): rados python tests occasionally hang with ms failures
- ...
- 11:20 AM Bug #2954 (Resolved): osd: scrub stat mismatch, got 18/19 objects, 14/15 clones, 22478527/2538528...
- Most likely fixed in b273c376ca6455f1e36be82cbc91606debd5fb1e.
- 09:13 AM rgw Feature #3074 (Resolved): radosgw needs --help support
- User reports he only discovered radosgw supports -d via radosgw-admin (which shouldn't even have that option).
Needs... - 09:12 AM Feature #3044 (Resolved): osd: include fail stamp in failure messages
- 09:12 AM rgw Bug #3073 (Resolved): radosgw-admin: is not a daemon, should not have -d/-f options
- ...
09/03/2012
- 09:06 PM Revision 3302a2d4 (ceph): doc: Added debug ref to toctree. Trimmed title names a bit.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:05 PM Revision 4609639b (ceph): doc: Added "how to" for debug/logging config. Trimmed titles too.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:04 PM Revision 7465dbeb (ceph): doc: Added logging reference under configuration section.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 09:00 PM Revision 4ab9f02f (ceph): msg/Pipe: do not special-case failure during connect
- Do not special case failure during connect. In particular, we may be
reconnecting and experience a second fault, and... - 08:35 PM Revision 7b6c7738 (ceph): doc: Added runtime configuration example.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:24 PM Documentation #2983: doc: write performance monitoring
- From Sage:
-- osds --
overall throughput
osd.op
osd.op_r
osd.op_w
osd.op_in... - 12:12 PM Bug #3071: osd: crash in ~ReplicatedPG
- ubuntu@teuthology:/a/sage-fuse2/14470
- 12:12 PM Bug #3071 (Resolved): osd: crash in ~ReplicatedPG
- i suspect this was from the coverity fixes?...
09/02/2012
- 08:46 PM Revision 73d016fd (ceph): test, key_value_store: added distributed flat btree key-value store
- Uses one index object and many sub objects to store key-value pairs. The pairs
are stored in the omaps of librados ob... - 11:53 AM Bug #3070 (Resolved): client/mds: ms failure injection causes problems
- most recently, ubuntu@teuthology:/a/teuthology-2012-09-01_19:00:04-regression-master-testing-gcov/13271
09/01/2012
- 09:51 PM Revision bfa9a59f (ceph): vstart.sh: -r to start radosgw
- Uses a fixed access/secret key for easier testing. Starts a standalone
apache2 process with basic config (based on t... - 09:50 PM Revision cbb8b96a (ceph): Merge remote-tracking branch 'gh/wip-coverity'
- 09:54 AM Subtask #2614 (Closed): Single Paxos instance shared across the existing services
- 09:53 AM Subtask #2745 (Closed): mon: Single-Paxos: Sync: Add new message support to the Monitor class
- 09:52 AM Subtask #2746 (Rejected): mon: Single-Paxos: Sync: Test message passing
- We tested the message passing using a ill-built bash script based on grepping logs on both sides.
Further testing ... - 09:51 AM Subtask #3069 (Rejected): mon: Single-Paxos: messaging: log MMonSync messages for offline matching
- Keep a log of sent/received MMonSync messages, obtainable through the admin socket, for offline matching.
Completi... - 09:44 AM Subtask #2737 (Resolved): mon: Single-Paxos: Sync: Force trimming to be proposed through Paxos
- 09:43 AM Subtask #2757 (Resolved): mon: Single-Paxos: Sync: pack chunks of the MonitorDBStore into transac...
- We now synchronize the stores of two different monitors, being able to checksum the transferred contents on both side...
- 12:06 AM Revision a386d8a9 (ceph): Merge branch 'wip-osd-flags'
- Reviewed-by: Samuel Just <sam.just@inktank.com>
08/31/2012
- 11:51 PM Revision 8578b216 (ceph): cls_lock_client: add ObjectOperation-based get_lock_info
- This will be used by librbd to grab lock info along with
the rest of its header information in a single request.
Sig... - 11:50 PM Revision e06031dd (ceph): cls_lock_types: add missing include
- msg_types defines entity-related types used here.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 11:31 PM Revision 8ac81779 (ceph): osd: defer backfill with NOBACKFILL osdmap flag is set
- If we encounter nobackfill, let ourselves to fall out of the recovery
queue. If we encounter a map that has does not... - 10:47 PM Revision e920a615 (ceph): Clarify CodingStyle with respect to tab compression of space runs
- Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Greg Farnum <gregory.farnum@inktank.com>
Reviewed-by: Jos... - 10:45 PM Revision fb379174 (ceph): Fix rados put from '-' (stdin)
- Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Mike Ryan <mike.ryan@inktank.com>
Reviewed-by: Greg Farnu... - 10:12 PM Revision 0a28d795 (ceph): osdmap: add NOBACKFILL flag
- This will tell the OSDs to please not initiate any backfill operations.
Signed-off-by: Sage Weil <sage@inktank.com> - 10:12 PM Revision 07ad9251 (ceph): osdmap: add NORECOVER flag
- This will stop recovery via log catch-up and via backfill both.
Signed-off-by: Sage Weil <sage@inktank.com> - 10:12 PM Revision 49375a0e (ceph): osd: pause/unpause recovery based on NORECOVER osdmap flag
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:43 PM Revision b273c376 (ceph): PG: do not update stats in ReplicaActive from info
- Bug #2954
Consider the following case:
1) Primary calls share_pg_info()
2) Primary processes client op and sends of... - 08:20 PM Revision 5ac52de0 (ceph): crushtool: Miscellaneous cleanup.
- Clean up the output messages; add some function documentation and some
unit tests.
Signed-off-by: caleb miles <caleb... - 08:11 PM Revision db9e1209 (ceph): osd/OSD.cc: OSD::shutdown() leaks lock
- CID 716952: Missing unlock (LOCK)At (48): Returning without unlocking
"this->osd_lock._m".
Signed-off-by: Samuel Jus... - 08:11 PM Revision d2356261 (ceph): osd/OSD.h: big parameter passed by value, publish_superblock
- CID 717051: Big parameter passed by value (PASS_BY_VALUE)At (1): Passing
parameter block of type OSDSuperblock (size ... - 08:11 PM Revision d41c253b (ceph): osd/OSDMap.cc: uninit vars in ctor struct qi
- CID 717337: Uninitialized scalar field (UNINIT_CTOR)At (6): Non-static class
member "weight" is not initialized in th... - 08:11 PM Revision 9a93eb32 (ceph): osd/OpRequest.h: uninit vars in ctor OpRequest
- At (2): Non-static class member "hit_flag_points" is not initialized in this
constructor nor in any functions that it... - 08:11 PM Revision 10636735 (ceph): osd/PG.cc: PG constructor pass PGPool by reference
- CID 717053: Big parameter passed by value (PASS_BY_VALUE)At (1): Passing
parameter _pool of type PGPool (size 336 byt... - 08:11 PM Revision f84e7764 (ceph): osd/PG.h: uninit var in ctor IndexedLog
- CID 717339: Uninitialized scalar field (UNINIT_CTOR)At (2): Non-static class
member "last_requested" is not initializ... - 08:11 PM Revision 471787f1 (ceph): osd/PG.h: uninit var in ctor OndiskLog
- CID 717342: Uninitialized scalar field (UNINIT_CTOR)At (2): Non-static class
member "has_checksums" is not initialize... - 08:11 PM Revision 6fbf9e6f (ceph): osd/PG.h: uninit var in ctor NamedState
- CID 717340: Uninitialized pointer field (UNINIT_CTOR)At (2): Non-static class
member "state_name" is not initialized ... - 08:11 PM Revision f30facab (ceph): osd/ReplicatedPG: pass PGPool to ReplicatedPG ctor by ref
- CID 717057: Big parameter passed by value (PASS_BY_VALUE)At (1): Passing
parameter _pool of type PGPool (size 336 byt... - 08:11 PM Revision 0e2efbc8 (ceph): osd/ReplicatedPG.h: uninit var in ctor OpContext
- CID 717343: Uninitialized pointer field (UNINIT_CTOR)At (3): Non-static class
member "snapset" is not initialized in ... - 08:11 PM Revision 75acb7bd (ceph): osd/ReplicatedPG.h: uninit var in ctor RepModify
- CID 717344: Uninitialized scalar field (UNINIT_CTOR)At (2): Non-static class
member "epoch_started" is not initialize... - 08:11 PM Revision 84216d59 (ceph): osd/Watch.h: uninit var in ctor Watch
- CID 717345: Uninitialized pointer field (UNINIT_CTOR)At (8): Non-static class
member "obc" is not initialized in this... - 08:11 PM Revision fd4d567b (ceph): osd/osd_types.h: fix pg_history_t::merge copy paste error
- CID 716882: Copy-paste error (COPY_PASTE_ERROR)At (2): "last_epoch_started" in
"other.last_epoch_started" looks like ... - 08:11 PM Revision 08a29f93 (ceph): ceph-dencoder: don't leak object
- Signed-off-by: Sage Weil <sage@inktank.com>
- 08:11 PM Revision b960c623 (ceph): librbd: fix delete[]
- CID 716902: Non-array delete for scalars (DELETE_ARRAY)
At (15): Deleting array variable "buf" with non-array delete ... - 08:11 PM Revision ec5133b0 (ceph): msg/Pipe: make banner buffer slightly bigger
- CID 717023: Out-of-bounds access (OVERRUN_DYNAMIC)
At (4): Allocating insufficient memory for the terminating null of... - 08:11 PM Revision 929892fe (ceph): LogEntry: pass by reference
- CID 717030: Big parameter passed by value (PASS_BY_VALUE)
At (1): Passing parameter w of type entity_inst_t (size 152... - 08:11 PM Revision f51f5685 (ceph): common/LogEntry: pass by value for contains()
- CID 717031: Big parameter passed by value (PASS_BY_VALUE)
At (1): Passing parameter k of type LogEntryKey (size 168 b... - 08:11 PM Revision 46c08d6c (ceph): log: protect m_stop with lock
- CID 716965: Data race condition (MISSING_LOCK)
At (2): Accessing "this->m_stop" ("_ZN4ceph3log3LogE.m_stop") requires... - 08:11 PM Revision 4eb31463 (ceph): msg/Accepter: move nonce init to ctor
- This was suggested by Greg too but I was too lazy.
CID 717331: Uninitialized scalar field (UNINIT_CTOR)
At (2): Non-... - 08:11 PM Revision a3bf558a (ceph): fix accepter bind fix
- 08:11 PM Revision 650ec35b (ceph): msg: pass by reference
- CID 717048: Big parameter passed by value (PASS_BY_VALUE)
At (1): Passing parameter a of type entity_addr_t (size 136... - 08:11 PM Revision b7af8d8b (ceph): msg/Message: fix possible null deref
- cct may be null
CID 716930: Dereference after null check (FORWARD_NULL)
At (11): Dereferencing null pointer "cct".
... - 08:11 PM Revision ea078e18 (ceph): msg/DispatchQueue: simplify shutdown()
- lcok isn't held during dispatch, so we can take it unconditionally. THis
also makes coverity happier:
CID 716966: D... - 08:11 PM Revision 89add0f7 (ceph): paxos: init vars in ctr
- At (2): Non-static class member "latest_stashed" is not initialized in this constructor nor in any functions that it ...
- 08:11 PM Revision 0b4ddc27 (ceph): mon/MonitorStore: uninit var in ctor
- CID 717329: Uninitialized scalar field (UNINIT_CTOR)
At (2): Non-static class member "lock_fd" is not initialized in ... - 08:11 PM Revision 50ed1fca (ceph): monclient: uninit var in ctor
- CID 717328: Uninitialized scalar field (UNINIT_CTOR) [select defect]
At (2): Non-static class member "newest" is not ... - 08:11 PM Revision e0b8cb7c (ceph): osd/ClassHandler: uninit var in ctor
- CID 717335: Uninitialized scalar field (UNINIT_CTOR)At (2): Non-static class
member "flags" is not initialized in thi... - 08:11 PM Revision b0412607 (ceph): osd/ClassHandler.cc: uninit var in ctor
- CID 717336: Uninitialized pointer field (UNINIT_CTOR)At (2): Non-static class
member "handler" is not initialized in ... - 08:11 PM Revision 532d1b10 (ceph): rgw: init RGWRados::cct in ctor
- Coverity cid 717382
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 139ba9f3 (ceph): client: unconditional teardown
- CID 717125: Dereference before null check (REVERSE_INULL)
At (2): Null-checking "this->objectcacher" suggests that it... - 08:11 PM Revision 484fd0c0 (ceph): mds: fix broken EMetablob inode_t* return value
- We pass this back to the caller, but it points to a member that is never
looked at, because the inode has already bee... - 08:11 PM Revision 1b1a0bf6 (ceph): msg/Accepter: check setsockopt return value
- CCID 716856: Other violation (CHECKED_RETURN)
At (7): Calling function "setsockopt(this->listen_sd, 1, 2, &on, 4U)" w... - 08:11 PM Revision 3fd13f57 (ceph): crypto: fix fd leak from get_random_bytes()
- Signed-off-by: Sage Weil <sage@inktank.com>
- 08:11 PM Revision 7cb3d2e4 (ceph): auth: fix uninit vars
- Signed-off-by: Sage Weil <sage@inktank.com>
- 08:11 PM Revision 757457ad (ceph): rgw: init ptr
- Coverity cid 717384
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision c67eda26 (ceph): rgw: init RGWGetObj::sent_header in ctor
- Coverity cid 717383
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 1247d4d5 (ceph): test/rados-api/aio: fix uninit ptr
- Coverity cid #717392
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 8d9d4f61 (ceph): test/osd/RadosModel: fix uninit ptr
- Coverity cid 717391
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 7c10bdc3 (ceph): test/osd/RadosModel: fix uninit values
- Coverity cid 717390
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 26a822bc (ceph): test/osd/RadosModel: fix uninit value
- Coverity cid 717389
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 20372126 (ceph): test/osd/RadosModel: uninit value
- Coverity cid 717388 717387
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 2214e838 (ceph): omap_bench: fix uninit var
- Coverity cid 717386
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision ca4e0607 (ceph): rgw: add missing ctor, init parent
- Coverity cid 717385
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision ec0a2662 (ceph): test_librbd: add missing va_end()
- Coverity cid 717407
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision a6dfa3e5 (ceph): rgw: fix use-after-free
- Coverity cid 717405
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 1355a676 (ceph): filer: add missing return
- Coverity cid 717404
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision b9adc27b (ceph): mds: drop unused value
- Coverity cid #717400
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 97bd0848 (ceph): mds: fix unused value
- This value was never used before it was clobbered.
Coverity cid #717398
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 998e2079 (ceph): test/Object: comment out dead code
- Coverity cid #717397
Signed-off-by: Sage Weil <sage@inktank.com> - 08:11 PM Revision 220980c9 (ceph): cephtool: fix uninitialized scalar in constructor
- Signed-off-by: Sage Weil <sage@inktank.com>
- 08:06 PM Revision dd4b9caa (ceph): cls_lock_client: return error when decoding fails
- Library code shouldn't be using cerr either.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 06:17 PM Revision 1779d5ab (ceph): cls_lock_client: fix indentation
- Add indentation settings to header, and reindent.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com> - 06:14 PM Revision b4298fc7 (ceph): cls_lock_client: change modified reference parameters to pointers
- This makes it clear which parameters are modified,
as our style guide states.
Signed-off-by: Josh Durgin <josh.durgi... - 04:53 PM Feature #2943 (Resolved): mon: norecovery and/or nobackfill
- 04:05 PM Bug #3068 (Resolved): rados: Can't pipe to the rados command
- commit:fb379174bfa32ea5123a451fc5ddc30ecf15d2f1
- 02:11 PM Bug #3068: rados: Can't pipe to the rados command
- It appears this was broken by the reorganization in some past delta; it shouldn't
be difficult to fix.
- 12:42 PM Bug #3068 (Resolved): rados: Can't pipe to the rados command
- while "rados -p poolname put objectname filename" works, "cat filename | rados -p poolname put objectname -" doesn't....
- 03:28 PM Revision 9c6752e8 (ceph): cls_lock_client: clean up reference parameters
- These should all be const. The remaining reference parameters
will be converted to pointers in another commit.
Signe... - 01:36 PM Feature #3044 (Fix Under Review): osd: include fail stamp in failure messages
- 01:36 PM Feature #3043 (Fix Under Review): mon: track osd laggy rate/interval
- 01:19 PM Linux kernel client Cleanup #2432 (Resolved): ceph-client: messenger: refactor to simplify state model
- 01:17 PM Linux kernel client Bug #3063 (Resolved): rbd: snapshots must be added to end of list
- 12:46 PM Bug #3059 (Resolved): osd:FAILED assert(waiting_for_ondisk.empty())
- by commit:5f36b8d78416b7a1d1bbefecddfcee00b7bfcfa3
- 11:11 AM rbd Feature #2952 (In Progress): librbd: use generic rados locking class
- 09:43 AM rgw Feature #1712 (Fix Under Review): rgw: support swift manifest objects
- 08:23 AM Bug #3067: mon: runaway memory
- If you can't ssh you probably need to power cycle the machine and restart the daemons. It sounds like there is some ...
- 08:20 AM Bug #3067 (Resolved): mon: runaway memory
- ...
- 08:21 AM Bug #2026: osd: ceph::HeartbeatMap::check_touch_file
- This is unrelated to the heartbeta problem; opened new bug #3067!
08/30/2012
- 11:39 PM Revision 3de0d4a3 (ceph): cls_lock: fix some spacing
- Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
- 11:06 PM Revision 82999c0e (ceph): Merge branch 'wip-rbd-children'
- 10:23 PM rgw Bug #2915 (Fix Under Review): rgw: copy of large object times out
- 10:20 PM rgw Feature #1712: rgw: support swift manifest objects
- 09:34 PM Revision dcbb87cc (ceph): rbd: add --format option
- This chooses whether to use the original (supported by krbd)
or the new (supports layering) format.
Signed-off-by: J... - 09:34 PM Revision 649bdc15 (ceph): doc: clarify rbd man page (esp. layering)
- * a clone's size can't be overridden
* note which commands require format 2
* clarify details of copy
* add examples ... - 09:32 PM Bug #2026: osd: ceph::HeartbeatMap::check_touch_file
- Ceph cluster is totally unreachable at this point, the only interesting in the log (on one of the machines, not the o...
- 08:58 PM Bug #2026: osd: ceph::HeartbeatMap::check_touch_file
- Some more information:
OS: Debian Wheezy 3.2.0-3-amd64
Ceph: ceph version 0.48argonaut (commit:c2b20ca74249892c8e... - 09:03 PM Revision 1b27a7cd (ceph): client: renew/submit subs on startup
- This fixes a delay in getting the initial MDSMap during startup (we wait
for the MonClient tick()).
Reported-by: Noa... - 08:37 PM Revision f85863e9 (ceph): objclass: add remove operation
- Added a missing remove object operation
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:36 PM Linux kernel client Bug #3040 (Resolved): btrfs: recursive locking of sb_internal#2
- 08:36 PM Linux kernel client Bug #3031 (Resolved): btrfs: lock returned to userspace
- 08:36 PM Bug #2842 (Won't Fix): mon: health detail lists pgs multiple times
- 06:48 PM Revision a23351ef (ceph): :doc: Added constraints and revised a description per joshd.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:41 PM Revision f2c45108 (ceph): upstart, ceph-create-keys: Make client.admin key generation automatic.
- This should help simplify Chef etc deployments. Now (when using the
Upstart jobs), when a ceph-mon is started, ceph-c... - 05:00 PM rbd Feature #2720 (Resolved): rbd: add children command
- Added in commit:465fcca1febb62d8871d482b5bc079aa21704454
- 05:00 PM rbd Feature #2719 (Resolved): librbd: provide functions for listing parents and their children
- Added in commit:78f69752a44f0f754ebdba2c06399c493f2be99f
- 03:45 PM Bug #3059: osd:FAILED assert(waiting_for_ondisk.empty())
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-08-30_00:00:06-regression-next-testing-basic/11576
- 03:44 PM rgw Bug #2527: RGW may return 409 Conflict when deleting a bucket
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-08-30_00:00:06-regression-next-testing-basic/11588
- 03:40 PM rgw Bug #2527: RGW may return 409 Conflict when deleting a bucket
- Logs:ubuntu@teuthology:/a/teuthology-2012-08-29_19:00:06-regression-master-testing-gcov/11434
- 03:38 PM Linux kernel client Bug #3058: xfs: possible circular locking dependency detected
- Logs: ubuntu@teuthology:/a/teuthology-2012-08-29_19:00:06-regression-master-testing-gcov/11363
- 02:25 PM Revision 3b0e2a2d (ceph): config: Add a per-name default keyring to front of keyring search path.
- This lets us have e.g. /etc/ceph/ceph.client.admin.keyring that is
owned by root:admin and mode u=rw,g=r,o= without m... - 02:25 PM Revision 0de71956 (ceph): upstart: Make instance jobs export their cluster and id variables.
- This allows other jobs listening to Upstart "started ceph-mon" events
to see what instance started.
Signed-off-by: T... - 01:59 PM Bug #3066 (Resolved): impermissible OSD ops get dropped on the floor
- Right now, an OSD operation that has insufficient permissions gets dropped without an EPERM error being sent back. I ...
- 07:38 AM Revision 8b4a9ade (ceph): doc: Modified index page with table. Pools and Auth will go to Ops.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:37 AM Revision b961a660 (ceph): doc: Some improvements. This still needs review.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:35 AM Revision c41fa394 (ceph): doc: Added additional note to mount data disk to data path.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:33 AM Revision b80a1b73 (ceph): doc: Added a document for general config settings.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:32 AM Bug #3065 (Won't Fix): don't always default to "client.admin"; perhaps client.$USER or something?
- It would be desirable to have a read-only, status-view-only key that is able to run trivial operations like "ceph -s"...
- 07:29 AM Revision 2407b0cd (ceph): doc: Added a doc for filestore settings.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:28 AM Revision 17dee291 (ceph): doc: Added a document for RBD cache settings.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:26 AM Revision 69e10167 (ceph): doc: Added substantial detail to OSD setting descriptions.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:25 AM Revision c0c9fe54 (ceph): doc: minor edit to use # for comments.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:24 AM Revision 98eaaccd (ceph): doc: Restructured ceph config. Added network and OSD config details.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:19 AM rbd Feature #3064 (Resolved): librbd: A way to read from nearby replicas
- In the style of CEPH_OSD_FLAG_LOCALIZE_READS, there are cases when an rbd image is stored in a pool where some of the...
- 12:44 AM Documentation #2970 (Resolved): doc: expand/complete osd settings reference
- This has been largely resolved. However, there are still a few fields that could use some attention. Updates will occ...
- 12:42 AM rgw Documentation #2972 (Resolved): doc: expand/complete rgw settings reference
- This was reviewed with Yehuda's help, and has been staged.
- 12:41 AM Documentation #3062 (In Progress): doc: osd tuning config options
- Added filestore section to configuration. Should be reviewed.
- 12:35 AM Revision 3c9bc47a (ceph): librbd: prevent racing clone and snap unprotect
- If the following sequence of events occured,
a clone could be created of an unprotected snapshot:
1. A: begin clone ... - 12:35 AM Revision 78f69752 (ceph): librbd: add {rbd_}list_children() methods
- These iterate over all pools and check for children of a
particular snapshot.
Signed-off-by: Josh Durgin <josh.durgi... - 12:35 AM Revision 465fcca1 (ceph): rbd: add "children" command, update cli test files
- Fixes: #2720
Signed-off-by: Dan Mick <dan.mick@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
08/29/2012
- 10:05 PM Linux kernel client Bug #3063 (Resolved): rbd: snapshots must be added to end of list
- I just found and fixed this bug and I'm going to try to make a
practice of at least documenting the fixes to actual ... - 09:43 PM Revision ac481e8e (ceph): rgw: cleanup create_bucket
- Pool creation is now being done through an excplicit
method, get rid of the unused user param when creating
a pool, a... - 09:27 PM Revision ef92157c (ceph): rgw: remove static store object
- We used to instantiate a single RGWRados object.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 09:22 PM Revision f71a4163 (ceph): rgw: rgwstore is not global
- Instead of using a global rgwstore param, just pass it around.
We now do it almost all around, except for in rgw_admi... - 09:22 PM Revision ed8922df (ceph): rgw: RGWRados holds domain root info
- Continuing cleanup work.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 09:22 PM Revision be00f9b6 (ceph): rgw: pool names are not global
- Move all hard coded pool names outside of the global
namespace.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 09:20 PM Revision 4d9802c4 (ceph): rgw: remove unused block of code
- We were reading bucket info, but that wasn't necessary.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:45 PM Revision e3834f02 (ceph): rgw_admin.cc: Prevent clobbering the index when linking a bucket.
- Prevent the 'bucket link' command from overwriting the index of an
existing bucket. Corrects bug 2935:
http://tracke... - 06:36 PM Documentation #3062 (Resolved): doc: osd tuning config options
- Copy-paste from recent ceph-devel email from Sam:
filestore [min|max] sync interval:
Periodically, the filestor... - 01:15 PM Bug #3061 (Resolved): osd crash during shutdown
- Logs: ubuntu@teuthology:/a/teuthology-2012-08-28_19:00:05-regression-master-testing-gcov/10931...
- 01:05 PM rgw Bug #2366: rgw: bucket index update rely on pg state
- I think a simple solution for that would be attaching the pool id to the version number. Then, assuming that differen...
- 12:46 PM rgw Bug #2935 (Resolved): rgw: radosgw-admin bucket link clobbers index
- Fix merged at commit:e3834f023be6d897995a2a7f6829f442be0795ff.
- 12:34 PM Bug #2761: osd: failed to recover before timeout expired
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-08-28_19:00:05-regression-master-testing-gcov/10872
- 12:32 PM Bug #3059: osd:FAILED assert(waiting_for_ondisk.empty())
- Logs: ubuntu@teuthology:/a/teuthology-2012-08-28_19:00:05-regression-master-testing-gcov/10868
- 11:25 AM rgw Feature #3060 (Resolved): rgw: different usage info break down
- We should be able to provide usage break down by the requests types, e.g. put_obj, get_obj, etc.
08/28/2012
- 11:47 PM Revision 82c62bd9 (ceph): rgw: clear usage map before reading usage
- Fixes: #3057
Since we read usage in chunks we need to clear the
usage map before reading the next chunk, otherwise
we... - 11:40 PM Revision 657ca118 (ceph): rgw: clear usage map before reading usage
- Fixes: #3057
Since we read usage in chunks we need to clear the
usage map before reading the next chunk, otherwise
we... - 10:15 PM Revision 0052a787 (ceph): Merge branch 'next'
- 10:14 PM Revision 5f36b8d7 (ceph): osd: fix waiting_for_disk assertion
- If requeue is false, we won't have cleared out waiting_for_ondisk; adjust
assert placement as appropriate. Also, mak... - 09:21 PM Revision 745a3c9b (ceph): rados_bench: wait for completion callbacks before returning
- If we don't wait for the callback, the finisher may cleanup the callback
context before the callback is actually invo... - 09:13 PM Revision dfb7f804 (ceph): Makefile.am: add missing .h
- Was missing rgw_html_errors.h
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:36 PM Revision 0ab62325 (ceph): Merge remote-tracking branch 'origin/wip-multi-delete'
- Conflicts:
src/Makefile.am
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 08:28 PM Revision 2d79a62a (ceph): rados_bench: wait for completion callbacks before returning
- If we don't wait for the callback, the finisher may cleanup the callback
context before the callback is actually invo... - 08:25 PM Revision a7867c27 (ceph): doc: Completed and reviewed RGW config reference.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 08:24 PM Revision 8d7912cb (ceph): doc: added admonishment. Updated header syntax, copy semantics and x-ref.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 07:51 PM Revision 5507665a (ceph): rgw: fix usage log read
- The usage log read got broken in a recent cleanup work.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 06:41 PM Revision 3eea9104 (ceph): doc: Added internal references. Clarified language in disk prepare.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 06:02 PM Revision a466dfff (ceph): doc: added sudo for hdparm command.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:55 PM Revision 080e5722 (ceph): doc: added internal hyperlink targets.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 05:49 PM Bug #3042: monitor hangs when osds are shut down
- ubuntu@teuthology:/a/teuthology-2012-08-28_00:00:10-regression-next-testing-basic/10259
- 05:41 PM Bug #3059: osd:FAILED assert(waiting_for_ondisk.empty())
- Logs: ubuntu@teuthology:/a/teuthology-2012-08-28_00:00:10-regression-next-testing-basic/10315...
- 05:29 PM Bug #3059 (Resolved): osd:FAILED assert(waiting_for_ondisk.empty())
- Logs: ubuntu@teuthology: /a/teuthology-2012-08-28_00:00:10-regression-next-testing-basic/10302...
- 05:17 PM Linux kernel client Bug #3058 (Won't Fix): xfs: possible circular locking dependency detected
- Logs: ubuntu@teuthology:/a/teuthology-2012-08-28_02:00:05-regression-testing-testing-basic/10406...
- 05:01 PM Revision dd011aba (ceph): doc: Clean up quick start to ensure nobody uses "localhost".
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 04:45 PM rgw Bug #3057 (Resolved): rgw: usage show show duplicate info when there's a large number of buckets
- Fixed, commit:82c62bd977c74c22385b18791943cb2054920f47
- 03:46 PM rgw Bug #3057 (Resolved): rgw: usage show show duplicate info when there's a large number of buckets
- Apparently we don't reset the aggregated usage before each iteration.
- 04:30 PM rbd Bug #2937: btrfs filesystem on rbd device kernel BUG writing large file
- This smells a bit like it's related to this discussion:
https://patchwork.kernel.org/patch/1271871/
I terminated ... - 04:24 PM Revision 97554666 (ceph): doc: Cleaned up syntax errors, and converted table to list.
- Signed-off-by: John Wilkins <john.wilkins@inktank.com>
- 03:17 PM rgw Bug #2916 (Resolved): radosgw does not check command line options for correctness
- The rados tool alone is affected by this issue and due to the nature of valid input to the command this will not be f...
- 10:01 AM rgw Bug #2916: radosgw does not check command line options for correctness
- 01:41 PM rgw Feature #2797 (Resolved): rgw: support multi-objects delete
- Fixed, merged at commit:0ab623250344705f55d0c249ad977f43bed8fd54.
- 01:30 PM Bug #3048 (Resolved): rados bench: use after free?
- fixed in 2d79a62a11ee4c18cbc8619da7921a0707ec401c
- 12:51 PM rgw Bug #3056 (Resolved): rgw: usage show doesn't show anything
- Fixed, commit:5507665a58d909c7c659b233d472c3d3a868130a.
- 12:48 PM rgw Bug #3056 (Resolved): rgw: usage show doesn't show anything
- Problem following cls rgw cleanup work. Affects current master only.
- 08:33 AM Bug #3055: ceph command stuck until all predefined monitors are up
- Holger Macht wrote:
> One thing that still bothers me is how the correct way of doing this looks like. '/sbin/mkceph... - 08:29 AM Bug #3055: ceph command stuck until all predefined monitors are up
- One thing that still bothers me is how the correct way of doing this looks like. '/sbin/mkcephfs -d /tmp/mon-init -c ...
- 07:45 AM Bug #3055: ceph command stuck until all predefined monitors are up
- Thanks for the clarification, that's all I wanted to know.
- 07:29 AM Bug #3055 (Won't Fix): ceph command stuck until all predefined monitors are up
- By putting all three monitors in the monmap, you've declared them to all be part of the monitor group, and you'll nee...
- 07:05 AM Bug #3055 (Won't Fix): ceph command stuck until all predefined monitors are up
- I'm using ceph-0.48.1 and encountered a problem from which I'm not sure if it's a bug or a wrong setup sequence. So p...
- 07:45 AM Bug #2026: osd: ceph::HeartbeatMap::check_touch_file
- It would be great if we could keep this issue going, and if there is anything I could provide, I'd be happy to do so....
- 06:39 AM RADOS Documentation #3054 (Closed): doc: omap, tmap, xattrs
- Just about everything in this thread belongs in docs, though some of it should go under dev/
http://thread.gmane.o... - 12:44 AM Revision 5630e812 (ceph): Merge branch 'next'
- 12:26 AM Revision 15995ea1 (ceph): Merge branch 'wip-objecter' into next
- Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
- 12:25 AM Revision e59b9dae (ceph): objecter: fix is_latest_map() retry on mon session restart
- If the mon session drops, we get an EAGAIN callback, which we already
correctly ignored. (Clean this up and comment ... - 12:25 AM Revision 8d1efd1b (ceph): objecter: send queued requests when we get first osdmap
- If we get our first osdmap and already have requests queued, send them.
Fixes: #3050
Signed-off-by: Sage Weil <sage@... - 12:25 AM Revision 2a3b7961 (ceph): objecter: fix skipped map handling
- If we skip a map, we want to translate NO_ACTION to NEED_RESEND, but leave
POOL_DNE alone.
Signed-off-by: Sage Weil ... - 12:25 AM Revision 7d40cba2 (ceph): monclient: pass EAGAIN to is_latest_map() callers
- If our map get_version check needs to be retried, tell the
is_latest_map() callers instead of giving returning 0 ("no... - 12:25 AM Revision 0adc2289 (ceph): monclient: document get_version(), and fix return value
- Return -EAGAIN instead of -1, since that's more meaningful, and
document it.
Signed-off-by: Sage Weil <sage@inktank.... - 12:08 AM Revision 0a1f4a97 (ceph): Implement multi-object delete.
- An implimentation of multi-object delete described in
the latest Amazon S3 API provied at
http://docs.amazonwebservi...
08/27/2012
- 11:47 PM Revision 17ceec0d (ceph): osd: requeue dup ops inline with in-progress ops
- We should requeue the dups along with the originals. This avoids
situations where, after requeue, the dups are reord... - 10:14 PM Bug #3048: rados bench: use after free?
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-08-27_00:00:04-regression-next-testing-basic/9564
- 10:10 PM Revision c7054933 (ceph): Merge remote-tracking branch 'gh/wip-mon-intparsing'
- Reviewed-by: Joao Eduardo Luis <joao.luis@inktank.com>
- 09:57 PM Revision d5cacaca (ceph): osd: include notif pointer in notify debug output
- Signed-off-by: Sage Weil <sage@inktank.com>
- 09:57 PM Revision 0a2ec988 (ceph): config: add 'fatal signal handlers' option
- This will let us disable the sighandlers for SEGV, etc.
Signed-off-by: Sage Weil <sage@inktank.com> - 09:55 PM Revision aaeb5519 (ceph): librados,ReplicatedPG: add omap_cmp
- Allows atomic checking of omap values.
Signed-off-by: Samuel Just <sam.just@inktank.com> - 09:55 PM Revision bc90c9aa (ceph): test/: renamed omap_bench.hpp to .h, fixed histogram formatting
- Signed-off-by: Eleanor Cawthon <eleanor.cawthon@inktank.com>
- 09:55 PM Linux kernel client Bug #2590: possible irq lock inversion dependency with con->mutex and osdc->request_mutex
- Recent logs: ubuntu@teuthology:/a/teuthology-2012-08-27_02:00:06-regression-testing-testing-basic/9679
- 09:41 PM Revision 7a631f94 (ceph): cls_rgw_client: fix #include path
- Signed-off-by: Sage Weil <sage@inktank.com>
- 07:43 PM Revision fa74e047 (ceph): Merge remote-tracking branch 'origin/master' into wip-gc2
- 07:42 PM Revision 6f68ff5c (ceph): cls_rgw: add cls_rgw unitest, test gc api
- Test various cls_rgw gc related functionality.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> - 07:41 PM Revision a30f7140 (ceph): rgw-admin: get rid of lazy remove option, other fixes
- was mishandling parsing of binary flag arguments.
also, fix argument parsing and update radosgw-admin
cli test refere... - 07:41 PM Revision 721a6bef (ceph): rgw: implement garbage collector
- Add a garbage collector thread that is responsible for clean
up of clutter. When removing an object, store info about... - 06:41 PM Bug #3049 (Resolved): mds: startup+suicide failure, MDLog::handle_journaler_write_error
- 03:09 PM Bug #3049: mds: startup+suicide failure, MDLog::handle_journaler_write_error
- 06:41 PM Bug #3050 (Resolved): objecter: need to resend requests when we get first map
- 03:09 PM Bug #3050: objecter: need to resend requests when we get first map
- 05:45 PM Bug #2947 (Resolved): osd: out of order reply
- commit:17ceec0d10cc6540c958e8f2d5ec1961750ced48
- 03:09 PM Bug #2947: osd: out of order reply
- 04:56 PM Linux kernel client Bug #3040: btrfs: recursive locking of sb_internal#2
- 03:36 PM Revision bd534bf3 (ceph): mon: make parse_pos_long() error message more helpful
- Signed-off-by: Sage Weil <sage@inktank.com>
- 03:13 PM Feature #3010 (Resolved): Make it easy to find a list of data locations from a cephfs file
- 03:12 PM Feature #3053 (Duplicate): osd: (de)prioritize recovery
- 03:11 PM Bug #2858 (Resolved): mon: osd id parsing returns 0 when passed 'osd.1234'
- 03:11 PM Bug #3014 (Resolved): ceph mds set_data_pool pool doesn't fail
- 12:46 PM rgw Feature #2839 (Resolved): rgw: garbage collection
- Done, commit:fa74e0476c0ca852a70f08805634661d89aa17de.
- 11:34 AM Bug #3052 (Resolved): ceph-osd --mkfs fails on ext3
- ...
- 11:27 AM Documentation #2981 (Fix Under Review): doc: write add/remove a monitor
- 11:26 AM Documentation #2973 (Resolved): doc: expand/complete ceph general settings
- 11:25 AM rbd Feature #3051 (Resolved): ObjectCacher: negative object caching
- 11:17 AM rbd Feature #2720 (Fix Under Review): rbd: add children command
- 11:17 AM rbd Feature #2560 (Resolved): rbd: safe parent deletion
- 11:16 AM rbd Feature #2724 (Resolved): rbd: add protect/unprotect command
- 11:16 AM rbd Feature #2718 (Resolved): librbd: map parent -> child in a per-pool rbd_children object when cloning
- 11:15 AM rbd Feature #2717 (Resolved): cls_rbd: add methods for maintaining mapping from parent to child images
- 10:52 AM Feature #2943 (Fix Under Review): mon: norecovery and/or nobackfill
- 10:42 AM rgw Feature #771: rgw: POST
- Requires #1717 for parsing policies
08/26/2012
- 10:16 PM Bug #3050 (Resolved): objecter: need to resend requests when we get first map
- ...
- 10:10 PM Bug #3049 (Fix Under Review): mds: startup+suicide failure, MDLog::handle_journaler_write_error
- was able to reproduce after a few attempts with...
- 09:30 AM Bug #3049 (Resolved): mds: startup+suicide failure, MDLog::handle_journaler_write_error
- ...
- 03:42 PM Revision c7d11cd7 (ceph): osd: turn off lockdep during shutdown signal handler
- We don't shut down all threads, and the surviving ones fight with
exit()'s teardown. Kludge until we have a clean sh... - 03:18 PM Revision 0e091d81 (ceph): Merge tag 'v0.51'
- v0.51
- 09:31 AM Bug #2947 (In Progress): osd: out of order reply
- nooo!...
- 09:27 AM Bug #3048 (Resolved): rados bench: use after free?
- ...
08/25/2012
- 10:58 PM Revision c03ca95d (ceph): v0.51
- 10:29 PM Revision aa91cf81 (ceph): mon: require --id
- Fixes: #2997
Signed-off-by: Sage Weil <sage@inktank.com> - 09:28 PM RADOS Bug #2874 (Resolved): apparent CRUSH mapping failure
- Glad to hear the tunables resolved this for you, Alex!
- 05:42 PM Feature #2944 (Duplicate): mon: dynamically adjust heartbeat grace
- #3044 #3043 #3047 #3045 #3046
- 05:40 PM Feature #3047 (Resolved): mon: apply heartbeat grace adjustment to down_out_interval
- 05:39 PM Feature #3046 (Resolved): mon: factor osd failure reporters into heartbeat grace
- When a sufficient number of failure reports come in to mark an OSD
down, additionally compute the laggy probability ... - 05:39 PM Feature #3045 (Resolved): mon: factor osd laggy interval into heartbeat grace
- Adjust the "heartbeat grace" locally on the monitor according to
the following formula:
adjusted_heartbeat_grace = ... - 05:38 PM Feature #3044 (Resolved): osd: include fail stamp in failure messages
- 05:36 PM Feature #3043 (Resolved): mon: track osd laggy rate/interval
- 1) Keep track of when an OSD boots if it reports itself as fresh or as
wrongly-marked-down. Maintain the probability... - 04:15 PM Feature #2742 (Resolved): qa: ms socket inject failures in regression suite
- 02:23 AM Revision dc1c247a (ceph): disable lockdep recursive warnings until #3040 is fixed
Also available in: Atom