Activity
From 08/02/2011 to 08/31/2011
08/31/2011
- 11:56 PM Revision 82776e15 (ceph): osd: one more flush before collection_list
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:49 PM Revision 3d69965c (ceph): workunits: remove unused variable
- 11:35 PM Revision 8c67d5ad (ceph): Don't specify kernel in any tasks.
- This can be added to every task at once when the suite is run.
- 11:24 PM Revision d5197893 (ceph): osd: whitespace
- 11:24 PM Revision f1cae577 (ceph): osd: flush previous operations to fs before collection list + destroy
- We need to flush any prior ops to the fs before we can rely on
collection_list to return all the objects we need to d... - 11:21 PM Revision f3325e65 (ceph): mds flock: make get_overlapping_locks and share_space handle 0 lengths
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 11:10 PM Revision 600bc8fd (ceph): MOSDSubOpReply/MSDSubOp: No need to update encoding version.
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 11:08 PM Revision 5bcea24d (ceph): architecture.rst: describe debug logs, cephcontext
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:27 PM Feature #488 (Resolved): osd: prehash pg content into subcollections
- 10:23 PM rgw Feature #1485 (New): rgw: support google storage api
- 10:23 PM rbd Feature #1484 (Rejected): libvirt: map rbd via kernel driver
- this should share a common image description schema as the qemu driver. we can treat qemu+librbd as a (preferred) op...
- 10:20 PM Feature #1483 (Resolved): glance: rbd backend
- store glance images in rbd
- 10:18 PM Feature #1335 (Rejected): osd: optionally tolerate per-object attr corruption
- this is really a meta-goal
- 10:17 PM rgw Feature #1482 (Resolved): qa: swift-tests
- incorporate some sort of swift api test suite. hopefully swift has something, or parts of s3-tests can be repurposed...
- 10:07 PM Feature #315 (Resolved): mds disaster recovery tools
- 10:06 PM Feature #314 (Closed): osd disaster recovery tools
- 09:59 PM Revision bcef985d (ceph): FlatIndex: lfn_get needs to set *exist for short filenames
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 09:54 PM Cleanup #67 (Closed): add 'autoscan' items to configure.ac
- 09:52 PM Bug #1470 (Closed): broken osd after filestore upgrade
- 02:09 PM Bug #1470: broken osd after filestore upgrade
- It's empty!...
- 09:50 AM Bug #1470: broken osd after filestore upgrade
- Can you post the contents of the meta collection?
- 03:30 AM Bug #1470 (Closed): broken osd after filestore upgrade
- upgraded to latest master (commit 961260d3a) and cosd began automatic upgrade of the filestore, which seemed to compl...
- 09:50 PM Bug #1454: cosd process dies and remains in defunct state
- Do you see anything in dmesg when this happens? The usually reason for this (that i've observed) is when we hit a ke...
- 09:48 PM CephFS Bug #1447 (Resolved): mds: does not validate pool IDs in handle_client_set[dir]layout
- 09:48 PM Linux kernel client Bug #1441 (Resolved): invalidation skips locked pages
- 09:48 PM Revision 00e0b77e (ceph): OSD: Fix encoding versions affected by hobject switch
- PG log did not previously store the object locator. To get the hash for
the hobject, scan the collection for the obj... - 09:45 PM CephFS Bug #1437: cfuse can't change permissions of a file
- Sam, is this something you can reproduce? All we should need is a client log.. something like '--log-file foo --log-...
- 09:43 PM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- I wonder if we can harass someone on IRC to get some partial 'sure, whatever' before we go write patches..
Or Wid... - 09:41 PM Tasks #1418 (Resolved): set up a no-atomic-ops gitbuilder
- 09:40 PM Feature #1401 (Closed): Support mutually untrusting clients using the same Ceph cluster
- Don't think there's anything in here not covered by #1237.
- 09:38 PM Feature #1385 (Closed): teuthology-nuke should support nuking "all hosts i have locked"
- I think this is better solved with
$ teuthology-lock --list-targets [--owner me] > list (added this recently)
... - 09:37 PM Revision 697f3419 (ceph): nuke: add option to reboot all nodes
- 09:36 PM Revision ec768ba3 (ceph): Fix pyflakes warnings.
- 09:34 PM Bug #1344 (Rejected): qa: move all fetched files to a central machine
- 09:32 PM Bug #1280 (Closed): gceph: missing dependencies
- 09:31 PM Bug #874 (Duplicate): Fix configure checks for g++
- 09:31 PM Bug #1255 (Duplicate): configure does not detect missing g++
- 09:27 PM Bug #1197: osd: make inconsistent state durable
- If we mark the pg inconsistent on disk, what does that actually mean? Is it just a 'tainted' flag that propagates to...
- 09:25 PM Tasks #1176 (Closed): qa: html test results
- 07:29 PM Revision 9a26100f (ceph): librbd: improve copy_with_progress
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:08 PM Feature #1476 (Resolved): teuthology: nuke rbd mounts
- 05:09 PM Feature #1476 (Resolved): teuthology: nuke rbd mounts
- can probably just reboot (as with kclient mounts) for now.
- 06:58 PM Revision 39fa9d8d (ceph): Merge remote branch 'origin/wip-scrub-oi-size'
- 06:51 PM CephFS Bug #1108 (Closed): Large number of files in a directory makes things grind to a halt
- Anything new here? Large directories aren't a part of our qa yet, but when they are this'll come up...
- 06:50 PM Bug #1072 (Won't Fix): cfuse mount fails when monitors have incorrect time synchronization
- 06:49 PM Bug #1029 (Rejected): ceph 0.27 run gceph, printf "Caught signal (Segmentation fault) " at a cli...
- 06:49 PM rgw Feature #1027 (In Progress): rgw log operations on non-existent bucket
- let's either add hte config option or close this bug.
- 06:48 PM Linux kernel client Bug #1014 (Closed): kclient can't complete fsstress on multi-mds cluster
- 06:46 PM Feature #1413 (Resolved): librbd asynchronous copy
- implemented
- 09:56 AM Feature #1413: librbd asynchronous copy
- Thanks, Yehuda. I suppose it does make sense to provide a C++-style API with a virtual base class, just to be consist...
- 06:41 AM Feature #1413: librbd asynchronous copy
- Works for me
- 06:42 PM Tasks #887 (Rejected): gitbuilder dash?
- 06:41 PM Linux kernel client Bug #853 (Resolved): bonnieloop.sh
- 06:41 PM Bug #836 (Rejected): atomic-ops error on Debian arm
- 06:40 PM Linux kernel client Feature #819 (Closed): robust nfs re-export
- 06:38 PM Linux kernel client Bug #804 (Resolved): read performance slow
- 06:38 PM Feature #800 (Rejected): pool namespace
- 06:38 PM rgw Feature #787 (Closed): rgw watchdog
- 06:37 PM Feature #748 (Rejected): Design/implement a RADOS copy
- 06:37 PM Feature #667 (Closed): Feature: ceph osd pool get data get pg_size
- 06:36 PM Bug #652 (Resolved): ReplicatedPG _scrub missing clone
- 06:35 PM Bug #644 (Closed): rsync can be sloooow
- 06:27 PM Bug #607 (Rejected): osd: ReplicatedPG: sub_op_modify: fix creation of ObjectState
- 06:26 PM Tasks #584 (Rejected): do throughput scaling tests on sepia
- 06:26 PM Linux kernel client Bug #576 (Can't reproduce): readdir returns too many results
- 06:25 PM Bug #538 (Closed): Write performance does not scale over multiple computers
- 06:24 PM Tasks #508 (Closed): test hadoop on sepia
- 06:24 PM Bug #492 (Rejected): osd: do not remove divergent objects
- 06:24 PM Tasks #454 (Resolved): c++ coding standards document
- 06:23 PM CephFS Cleanup #431 (Resolved): mds: clean up inode journaling internal interfaces
- 06:23 PM rbd Bug #391 (Closed): snap create/delete caused corruption
- 06:22 PM RADOS Bug #369 (Rejected): crush: calculate separate parent mappings for all possible hierarchies
- 06:21 PM Tasks #347 (Rejected): set up man2html or similar to get man pages on web site
- 06:21 PM Bug #323 (Duplicate): Get the wireshark dissector up to date
- 06:20 PM Feature #286 (Rejected): librados: set read chunk size
- 06:19 PM Feature #231 (Rejected): Slow OSDs shouldn't destroy cluster performance
- 06:19 PM Feature #242 (Rejected): fuse rbd driver
- 06:19 PM rbd Feature #1480 (Resolved): librbd: image locking
- Implement cooperative locking of RBD images (coordinated on the header, of course). Write public class functions to s...
- 06:18 PM Feature #191 (Rejected): rbd needs locking mechanism
- 06:17 PM Cleanup #146 (Rejected): Complete build options for Pthread API
- 06:16 PM Cleanup #129 (Rejected): msgr: separate message encoding into sections
- 06:16 PM Cleanup #103 (Closed): Introduction of namespaces
- 06:16 PM Cleanup #100 (Closed): #define ? enum?
- 06:14 PM Bug #35 (Duplicate): osd: pg split should queue transaction(s) under appropriate sequencer(s)
- 06:14 PM Linux kernel client Bug #28 (Won't Fix): gracefully fail on fill_trace errors
- 06:13 PM Bug #14 (Duplicate): osd: pg split breaks if not all osds are up
- 06:01 PM Feature #1479 (Resolved): librbd: remove_with_progress
- we can use the same progress callback infrastructure.
- 05:45 PM Cleanup #1478 (Resolved): remove last uses of old argument parsing macros, support -- everywhere
- The old-style argument parsing macros are a little confusing, since they're essentially flow control macros. They als...
- 05:42 PM Bug #1477 (Resolved): make g_conf constant
- 99% of the code is doing the right thing and treating g_conf as a constant, but there is some code out there that's w...
- 05:41 PM CephFS Bug #1318: directories disappear across multiple rsyncs
- added a workunit misc/multiple_rsyncs.sh to do a couple rsyncs and make sure no additional files are transfered. src...
- 05:08 PM Revision d08c7843 (ceph): Makefile: os/CollectionIndex.h
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:05 PM Revision 47a80635 (ceph): assert: work around libatomic_ops vs assert in a less lame way
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:00 PM Revision 37148621 (ceph): global_init: make startup line consistent, less ugly
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:00 PM Revision 5ae3e136 (ceph): crypto: reinclude our assert to clobber system one
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:00 PM Revision cf862c65 (ceph): assert: use our assert
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:00 PM Revision 5b7f9786 (ceph): filestore: reorder includes
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:58 PM CephFS Bug #1467 (Closed): cfuse crash during fsx workunit
- same, i think this was the MOSDOpReply bug
- 04:35 PM Bug #1468 (Rejected): mds crash after ffsb on cfuse
- This was the osd op reply bug (objecter->stat() decides data in the reply on success, but error code was munged)
- 04:35 PM Bug #1475 (Resolved): Our wait lock implementation is EINTR-happy
- So according to the fcntl man page,...
- 04:34 PM Bug #1469 (Rejected): kernel_untar_build workunit failed on cfuse
- I'm running this one in a loop and it's succeeding hundreds of times in a row.. gonna close this and see if it happen...
- 04:32 PM Bug #1471 (Resolved): osd: destroy_collection on non-empty dir
- I think the root bug is fixed by commit:f1cae577e3730a73dd5478785160745150095af5 and commit:e0776761a1d6866fda73fa5af...
- 03:34 PM Bug #1471 (In Progress): osd: destroy_collection on non-empty dir
- 12:19 PM Bug #1471: osd: destroy_collection on non-empty dir
- ...
- 10:03 AM Bug #1471: osd: destroy_collection on non-empty dir
- The assert.h #includes were screwed up so there's no line number in the log and I can't tell which assert this was hi...
- 07:13 AM Bug #1471 (Resolved): osd: destroy_collection on non-empty dir
- This was running 0.34 installed from packages.
Last 30 lines of log:... - 03:52 PM CephFS Bug #1472: cfuse hangs with v0.34
At about the time of that last client hang (_open), I do see these messages in the active mds log:
2011-08-31 17...- 03:28 PM CephFS Bug #1472: cfuse hangs with v0.34
- At most 20 processes running at any given time (different instances of the same application) from a single client, re...
- 03:15 PM CephFS Bug #1472: cfuse hangs with v0.34
- What does the workload look like?
- 03:12 PM CephFS Bug #1472 (In Progress): cfuse hangs with v0.34
- Well, so much for that then.
Are these actually new hangs compared to v0.33? Newly-noticed but possibly present be... - 03:05 PM CephFS Bug #1472: cfuse hangs with v0.34
- I have verified that this hang is not due to osds crashing. With all osds running, and all pgs active+clean, I still...
- 02:31 PM CephFS Bug #1472: cfuse hangs with v0.34
- Well with 3 OSDs down you probably lost access to some objects?
It probably shouldn't hang all other requests on t... - 02:26 PM CephFS Bug #1472: cfuse hangs with v0.34
- Only 3 osds crashed though. It seems like there should be other PGs on other osds that are still accessible, unless ...
- 02:14 PM CephFS Bug #1472 (Duplicate): cfuse hangs with v0.34
- Yeah, this is probably due to dead OSDs, so the client's unable to find anywhere to read the data from and is just wa...
- 01:34 PM CephFS Bug #1472: cfuse hangs with v0.34
- FYI: These hangs may have just been caused by osd failures (see #1473). I will update if this issue persists.
- 12:10 PM CephFS Bug #1472 (Can't reproduce): cfuse hangs with v0.34
- I see hangs with cfuse that appear to be at random (random requests to servers). Here are the backtraces of some cfu...
- 03:49 PM Bug #1465 (Rejected): error running kclient workunit on kclient
- It wasn't getting the kernel upgraded after all.
- 02:51 PM Bug #1465: error running kclient workunit on kclient
- It's in an overrides section that gets included when the suite is scheduled (the kernel task defaults to master branc...
- 02:33 PM Bug #1465: error running kclient workunit on kclient
- Where does it specify kernel master branch? It's not in the YAML fragment for that test.
- 10:02 AM Bug #1465 (In Progress): error running kclient workunit on kclient
- Hmm, I didn't check specifically, I just assumed that's what was going on. I'll take a look at it once I've dealt wit...
- 09:30 AM Bug #1465: error running kclient workunit on kclient
- The suite runs do specify kernel master branch. Did a node not get upgraded?
- 09:00 AM Bug #1465 (Closed): error running kclient workunit on kclient
- We should make these suite runs always use the kernel master branch.
- 08:57 AM Bug #1465: error running kclient workunit on kclient
- Ugh, this is because the kernel needs the fixes I applied for that part of the test to pass. We have a few options:
... - 01:59 PM Bug #1473: osd assert failure: FAILED assert(0 == "oi disagrees with stat, or error code on stat")
- Can you reproduce the second crash with the osd log enabled?
- 01:32 PM Bug #1473: osd assert failure: FAILED assert(0 == "oi disagrees with stat, or error code on stat")
If I comment out the if(1) check that is being done (and causing the assert failure), I'm able to restart the osds ...- 12:12 PM Bug #1473 (Resolved): osd assert failure: FAILED assert(0 == "oi disagrees with stat, or error c...
- 2011-08-31 10:04:29.898531 7fa020ddd700 -- 192.168.101.11:6801/7607 >> 192.168.101.14:6806/2173 pipe(0x78a7000 sd=34 ...
- 01:31 PM Bug #1453 (Resolved): osd: warn on object_info_t::size != st_size when building scrub_map
- This is detected by scrub now, but repairing it will take more work, since the scrub system isn't designed to repair ...
- 01:29 PM RADOS Feature #1474 (New): scrub: repair intra-node inconsistencies
- Currently scrub repairs inter-node inconsistencies in object existence, xattrs, and size. There are a bunch of intra-...
- 10:12 AM Bug #1463 (Resolved): rgw_admin crash
- Fixed by commit:6180c2cccf3e910cb78a139909b19b2333b79144.
- 10:10 AM Bug #1466 (Resolved): rados aio get_return_value tests fail
- Fixed by commit:6180c2cccf3e910cb78a139909b19b2333b79144.
- 10:01 AM Bug #1462 (Resolved): end of bufferlist reached
- Looks like this was fixed by commit:6180c2cccf3e910cb78a139909b19b2333b79144
- 09:19 AM CephFS Bug #1367 (Resolved): cfuse and mon crash after dbench
- 04:24 AM Revision 961260d3 (ceph): Makefile: add os/*Index.h to dist tarball
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 02:35 AM Revision 4dce9fe5 (ceph): doc/architecture: describe lib arch, config arch
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:21 AM Revision 5dd50b39 (ceph): coverage: remove debugging
- 12:13 AM Revision 5b42b085 (ceph): workunit: save coverage and coredumps
- Anything that runs a ceph utility should be using these commands.
08/30/2011
- 11:36 PM Revision ed2d1eac (ceph): basic workunits: switch to run on all clients simultaneously instead of...
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 11:30 PM Bug #1469 (Rejected): kernel_untar_build workunit failed on cfuse
- Logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/269
- 11:27 PM Bug #1468 (Rejected): mds crash after ffsb on cfuse
- Logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/274...
- 11:20 PM CephFS Bug #1467 (Resolved): cfuse crash during fsx workunit
- Logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/276...
- 11:11 PM Bug #1466 (Resolved): rados aio get_return_value tests fail
- Logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/294...
- 11:08 PM Bug #1465 (Rejected): error running kclient workunit on kclient
- ...
- 11:04 PM CephFS Bug #1464 (Can't reproduce): mds crash during shutdown (after trivial_sync workunit on kclient)
- Logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/293...
- 10:56 PM Revision 6180c2cc (ceph): osd: fix osd reply message
- 10:49 PM Revision 6d919152 (ceph): workunits: rework a little bit to allow "all" clients in a run
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:35 PM Revision be2187c8 (ceph): ObjectStore: Support older format on journal replay
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 10:35 PM Revision 5b709589 (ceph): cosd,OSD: Improve filestore upgrade path
- Previously, fsconverter was required to update an osd filestore to the
most recent version. cosd will now handle tha... - 09:17 PM Revision f9b85370 (ceph): librados: don't hide error in case of failed pool ctx creation
- 08:34 PM Revision ec97dd82 (ceph): cfuse: support running through valgrind
- Also switch up the config code so we can take per-client options.
- 08:15 PM Revision f0b626d4 (ceph): Merge branch 'stable'
- 08:14 PM Revision 6247df6b (ceph): client: fix iterator adjustment in readdir assimilation
- We need to move past the item we already had or else we'll remove it again
during the next iteration of the loop.
Si... - 08:14 PM Revision 7a8ab747 (ceph): client: plug leak of inode refs on seekdir(0)
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:14 PM Revision ed7ebf35 (ceph): client: drop mostly-useless relink()
- Just use unlink() and then link(). Carry an inode ref to avoid badness.
The relink() is left over from a simpler tim... - 07:47 PM Revision cc1cb8da (ceph): doc: Hide empty and broken "module index" page.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 07:38 PM Revision df0f32b7 (ceph): doc: Disable static file copying, we don't use it.
- Avoids a warning from sphinx on every run.
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com> - 07:37 PM Revision 3144fd73 (ceph): doc: Flesh out librados API example, to show doxygen & sphinx integration.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 06:50 PM Revision df38a701 (ceph): JournalingObjectStore: journal_replay committed_seq fix
- Previously, committed_seq was set to op_seq at the end of
replay_journal. committed_seq should not actually be set u... - 06:26 PM Revision c85c2f75 (ceph): doc: Adjust how table of contents is presented.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 05:12 PM Bug #1463 (Resolved): rgw_admin crash
- Full logs are in teuthology:~teuthworker/archive/nightly_coverage_2011-08-30/303/....
- 05:11 PM Bug #1462 (Resolved): end of bufferlist reached
- ...
- 04:48 PM Feature #1443 (Resolved): qa suite: make basic tasks run on mutiple clients
- Pushed to teuthology and ceph-qa-suite master!
(Hope I didn't just break all the workunits in the suite...) - 03:40 PM Revision ae552539 (ceph): client: additional sanity checks on link/unlink
- Signed-off-by: Sage Weil <sage@newdream.net>
- 03:38 PM Revision c44f134b (ceph): Merge remote branch 'vapre/stable'
- 03:38 PM Revision 8b8e47b1 (ceph): client: simplify Inode operator<<
- Signed-off-by: Sage Weil <sage@newdream.net>
- 02:32 PM CephFS Bug #1460 (Resolved): mds: file locks don't work right with 0-length locks
- Right now it just doesn't handle them properly. See, eg ...
- 02:15 PM Revision 3d58d007 (ceph): client: fix readder result merge
- When merging readdir results into the cache, we want to remove any names
_preceeding_ the current item before updatin... - 02:13 PM Revision 8d4c7e84 (ceph): client: relink_inode -> relink
- and some comments.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 02:10 PM Revision 70178046 (ceph): client: add some asserts
- Add sanity checks in close_dir(), and make sure we aren't replacing a dn
with the same name in link().
Signed-off-by... - 02:09 PM Revision afd5236c (ceph): client: fix readder result merge
- When merging readdir results into the cache, we want to remove any names
_preceeding_ the current item before updatin... - 01:43 PM Feature #1421 (Resolved): build release deb for oneiric
- 01:25 PM CephFS Bug #1456 (Resolved): cfuse: crash in snaptest2 during full snaps run
- 10:59 AM Feature #488 (Closed): osd: prehash pg content into subcollections
- Merged.
- 10:11 AM Feature #1413: librbd asynchronous copy
- For being more in line with the read_iterate() that'll call it, I suggest:
for C:
int callback(uint64_t total_... - 01:51 AM Feature #1413: librbd asynchronous copy
- I would say the callback route would be best.
Something along the lines of:... - 09:48 AM Feature #1459 (Resolved): rgw/qa: update teuth apache setup
- threaded, etc. match dho.
- 02:32 AM Revision 89fb93e4 (ceph): os/hashindex: fix #include
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:59 AM Revision b4e84cf7 (ceph): Merge branch 'wip_colindex'
- 12:43 AM Revision eb22b8df (ceph): os/FlatIndex: FlatIndex encapsulates previous FileStore behavior
- FlatIndex will be used to read and convert the previous on disk
format.
Signed-off-by: Samuel Just <rexludorum@gmail... - 12:43 AM Revision d59f34ab (ceph): FileStore: Integrates HashIndex/IndexManager into FileStore
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision e23e1924 (ceph): osd_types.h: Fix support for old object_info_t
- The object locator in the old encoding is necessary
to get the hash for soid.
Signed-off-by: Samuel Just <samuel.jus... - 12:43 AM Revision 341fb208 (ceph): FileStore: Add filestore version stamp
- A filestore will now be tagged with a version stamp during
mkfs. If on mount the version stamp detected lags the cur... - 12:43 AM Revision 42742ccd (ceph): FileStore: Added update_version_stamp
- update_version_stamp may be used to refresh the filestore version
stamp when all collections have been brought to the... - 12:43 AM Revision 279e1635 (ceph): FileStore: On mount, scan collections for unstable state
- CollectionIndex implementations may perform compound operations
leaving invalid state if interrupted. index->cleanup... - 12:43 AM Revision e6c91f6b (ceph): fsconverter.cc: Added fsconverter
- fsconverter updates an existing FileStore to the most recent FileStore
format
Signed-off-by: Samuel Just <rexludorum... - 12:43 AM Revision 05989673 (ceph): osd/,os/,osdc/: Convert collection_list_handle_t to a struct
- Previously, we stored the collection_list handle as an opaque
uint64_t. Now, collection_list_handle_t is a struct de... - 12:43 AM Revision 45d82964 (ceph): object.h: hobject_t definition
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision 1f591c29 (ceph): src/os/: s/sobject_t/hobject_t/g
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision f76baf1e (ceph): src/osd/: s/sobject_t/hobject_t/g
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision 8fed10e2 (ceph): src/messages/: s/sobject_t/hobject_t
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision a09c4913 (ceph): dupstore: s/sobject_t/hobject_t
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision 5e927ebe (ceph): osd/: fix hobject_t construction
- sobject_t requires only an object_t and a snapid_t. hobject_t also
requires the hash which should be used for the ob... - 12:43 AM Revision 5492cbdf (ceph): ReplicatedPG: find_object_context now takes an hobject_t
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:43 AM Revision d0744cdd (ceph): FileStore: CollectionIndex, HashIndex, IndexManager
- Adds ColletionIndex, an interface for collection indexing
systems, and HashIndex, a mechanism for organising a prehas... - 12:43 AM Revision de1f424a (ceph): ceph/src/test/store_test.cc: added store_test.cc
- Meant to test basic file store functionality
Signed-off-by: Samuel Just <samuel.just@dreamhost.com> - 12:42 AM Revision 30856dc1 (ceph): JournalingObjectStore: journal_replay committed_seq fix
- Previously, committed_seq was set to op_seq at the end of
replay_journal. committed_seq should not actually be set u... - 12:42 AM Revision 1fe54bfe (ceph): FileStore.cc: Fix error in collection_rename
- Previously, the rename operated on the collection names. Now, rename
operations on the full paths to the collections... - 12:42 AM Revision 71123b2d (ceph): ObjectStore.h: const cleanup in transaction method params
- Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
- 12:27 AM Revision f1d89644 (ceph): First draft of the documentation overhaul.
- To build the docs, run ./admin/build-doc. To browse them, either get
them on any static website, or just run ./admin/... - 12:27 AM Revision afede55c (ceph): Integrate Doxygen into Sphinx docs.
- Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
- 12:19 AM Revision 68f57f9a (ceph): test: make escaping unitest happy
- 12:12 AM Revision f2d19a03 (ceph): json escaping fix
- 12:02 AM Revision 19f09c0a (ceph): fix json escaping
- 12:02 AM Revision a2b0857c (ceph): radosgw_admin, rados tool: don't use camel casing for json output
08/29/2011
- 11:57 PM Revision cb89d005 (ceph): testlibrbdpp: test copy_with_progress
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:57 PM Revision 95127960 (ceph): librbd: add copy_with_progress API
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:50 PM Revision 8293dfab (ceph): pg: warn about inconsistent object_info/on-disk size
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 11:48 PM Revision e61fc663 (ceph): pg: remove useless line
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 11:48 PM Revision 281dae65 (ceph): osd_types: add standard header so spacing is correct
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 11:47 PM Revision 0c2bee15 (ceph): valgrind: don't run valgrind_post if there's no valgrind
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:53 PM Revision b140ff22 (ceph): client: debug link/unlink and dn_set parent relationship
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:04 PM Revision f8296ef5 (ceph): rados tool: df dumps formatted data
- 10:02 PM Revision 853658e9 (ceph): heartbeatmap: fix reset_timeout with mixed-used threads
- If you have a ThreadPool used by multiple WorkQueues, and some of them are
setting a suicide timeout, we need to clea... - 09:20 PM Revision c4219a11 (ceph): rgw: log show dumps some information about bucket owner
- 09:03 PM Revision 50a648bd (ceph): valgrind: use xml output for tools that support it
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:03 PM Revision 3a3c859f (ceph): valgrind: scan logs for bad results
- It's not sophisticated but it will warn you about a node
if at least one node has issues.
Signed-off-by: Greg Farnum... - 08:42 PM Revision 8cb25cab (ceph): bandaid for gitbuider
- 08:39 PM Revision 3ecdd112 (ceph): Merge remote branch 'vapre/wip-pgid'
- 08:39 PM Revision 698df7e1 (ceph): osd: remove dead lookup_lock_pg()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:42 PM Revision 7be9eaa0 (ceph): suite: add option to send an email if the entire suite passed
- 06:54 PM Revision 9d37c92c (ceph): osd: set suicide timeouts on some workqueues
- OpWQ: timeout * 10
RecoveryWQ: this does no io; it if stalls we're probably stuck in an
infinite loop. timeout * 1... - 06:46 PM Revision 502cf0b2 (ceph): filestore: add suicide grace to workqueue
- The main thing we are worried about here are kernel BUGs that kill off our
threads.
Signed-off-by: Sage Weil <sage@n... - 06:46 PM Revision 66b6289a (ceph): mon: health not ok when up < in osds
- We were warning if there were any not up or in osds. Instead, warn if
there are any osds that are in but not up. Th... - 06:46 PM Revision 5fde4011 (ceph): heartbeatmap: add suicide grace
- Add a second suicide grace period that will make us kill ourselves if
we are sufficiently catatonic.
Signed-off-by: ... - 06:33 PM Revision 8d64fee4 (ceph): testlibrbdpp: test librbd::Image::copy
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:58 PM Revision 68d1ba84 (ceph): rgw: don't check for time skew when Expires was set
- 05:23 PM Revision 0545e2fa (ceph): Add scripts to analyze coverage for a single teuthology run.
- 05:23 PM Revision b582764b (ceph): Add teuthology-coverage for analyzing test coverage for a suite run.
- 05:23 PM Revision 2b669382 (ceph): queue: delete every job when it finishes, so only running jobs are buried
- 05:23 PM Revision 4f4227a4 (ceph): Generate coverage at the end of a suite run,
- and optionally email failures and ongoing jobs.
- 04:31 PM Revision b9c824c0 (ceph): client: clean up Client::remove_session_caps() args
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:31 PM Revision 1bb0e3cc (ceph): client: simplify Client::remove_cap() args
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:31 PM Revision 16a5761f (ceph): client: don't falsely populate mds_sessions on bad MClientSession msg
- If from isn't in the map, we don't want to add it via operator[].
Fixes: #1367
Signed-off-by: Sage Weil <sage@newdre... - 04:26 PM Bug #1376: errant scrub stat mismatch logs after upgrade
- If you still have this cluster around, could you try applying commit:8293dfabb554883a30af549447995390fafa1f62 to see ...
- 04:10 PM Bug #1434 (Resolved): osds all failing each other
- 03:59 PM Revision 04153abb (ceph): testlibrbd: remove useless sizeofs
- Should be sizeof(*result), if anything.
Signed-off-by: Sage Weil <sage@newdream.net> - 03:56 PM Revision 3dcb04b9 (ceph): testlibrbd: fix assert side effect
- Coverity cid 1, 2
Signed-off-by: Sage Weil <sage@newdream.net> - 02:15 PM Bug #1458 (Resolved): Run ceph suite with valgrind enabled
- Once #1457 is done, let's see if we turn up any other problems via valgrind.
- 02:15 PM Bug #1457: Make daemons valgrind clean
- That is, I see they're not clean just when using vstart. Fix those issues first.
- 02:14 PM Bug #1457 (Resolved): Make daemons valgrind clean
- Noticed while doing #1411 that the nodes aren't actually valgrind clean. We probably don't need to worry about "Possi...
- 02:13 PM Feature #1411 (Resolved): teuthology: scan valgrind log for badness
- Pushed a simple one in commit:3a3c859f5bc8691891ff8a0f67a960a0d538083e
- 01:31 PM Feature #1411: teuthology: scan valgrind log for badness
- Hmm, I'm actually seeing a few errors in the MDS and OSD just by starting them up and shutting them down. (eg, vstart...
- 01:51 PM CephFS Bug #1456 (Resolved): cfuse: crash in snaptest2 during full snaps run
- cfuse seems to be failing on master with the following config:
roles:
- - mon.0
- mds.0
- osd.0
- - mon.1
... - 01:39 PM Feature #1455 (Resolved): librbd: convert testlibrbd, testlibrbdpp into gtest style unit test
- testlibrbd and testlibrbdpp should be converted into gtest style unit tests, similar to the rados-api tests.
- 12:55 PM Bug #1453 (In Progress): osd: warn on object_info_t::size != st_size when building scrub_map
- 12:54 PM Feature #1374 (Resolved): teuthology: email suite run results
- 12:54 PM Feature #1120 (Resolved): qa: gcov metrics
- 11:14 AM Bug #1454 (Can't reproduce): cosd process dies and remains in defunct state
- I'm seeing an occasional cosd failure where the process will die, but remains in a defunct state:
root 3165 ... - 09:17 AM CephFS Bug #1367 (In Progress): cfuse and mon crash after dbench
- ok, just hit the top one after 35 runs.
08/28/2011
- 09:46 PM Bug #1382: kclient: crash on resending osd ops
- Need to set up a teuthology job with rbd + thrasher and a suitable long-running workload.
- 09:43 PM Bug #1376: errant scrub stat mismatch logs after upgrade
- I think this is caused by an old bug. scrub needs to be fixed to properly detect (and ideally repair) it. See #1453.
- 09:43 PM Bug #1453 (Resolved): osd: warn on object_info_t::size != st_size when building scrub_map
- I belive that #1376 is caused by an st_size != object_info_t::size mismatch (from some bugs a few versions back). Sc...
- 09:28 PM Bug #1407 (Can't reproduce): _scrub crash on missing clones
- A whole bunch of objects were missing here. The immediate problem is that it crashed instead of printing an error me...
- 09:25 PM Feature #1411: teuthology: scan valgrind log for badness
- As soon as this is done (and/or as a test) let's do a full suite run with valgrind on (maybe one daemon at a time) an...
- 04:13 PM Bug #1415: cosd assertion: existing->state == STATE_CONNECTING || existing->state == STATE_OPEN
- From the log it doesn't look like it's a connect to self. The interesting thing is that existing->state is STANDBY. ...
- 03:53 AM Revision 66f1d41d (ceph): FileJournal: Also classify kernel major version 3 as safe with disk-cac...
- Signed-off-by: Wido den Hollander <wido@widodh.nl>
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 03:45 AM Revision 6c27367d (ceph): osd: decode pg_t in old PG::Info
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
08/27/2011
- 09:39 PM Feature #1451 (Resolved): librbd: instrument via perfcounter
- instrument librbd via perfcounter to keep track of things like
- # write ops
- # bytes written
- # read ops
- #... - 05:20 PM Revision 5d81c4fe (ceph): osd: requrie PGID64 of mons, other osds
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:20 PM Revision 7cedf7b8 (ceph): osd: add PGID64 feature bit
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:20 PM Revision fbeafdf9 (ceph): osd: make MOSDOp[Reply] encoding backwards compatible
- Move away from struct-based encoding of MOSDOpReply while we're at it.
Signed-off-by: Sage Weil <sage@newdream.net> - 05:20 PM Revision eae1922a (ceph): osd: decode old pg_stat_t
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:20 PM Revision d1ce4108 (ceph): osdmap: encode/decode old + new version
- In MOSDMap, reencode map payloads using old format if the target doesn't
have the PGID64 feature bit.
Signed-off-by:... - 05:20 PM Revision 6e58bc9d (ceph): osd: add 64-bit pools to ObjectLocator
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:20 PM Revision 92b524e1 (ceph): mdsmap: int64_t pool ids
- Encode old-format maps for clients without PGID64 feature bit.
Signed-off-by: Sage Weil <sage@newdream.net> - 05:20 PM Revision f093a2ee (ceph): pgmap: decode old pg_t
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:20 PM Revision efd43bbc (ceph): mon: require PGID64 of osds and other mons
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:20 PM Revision d996ce2a (ceph): mds: require PGID64 of mons
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 05:20 PM Revision 468e28ee (ceph): osd: int pool -> int64_t pool
- Internal interfaces.
Signed-off-by: Sage Weil <sage@newdream.net> - 05:18 PM Revision d8f1f1dd (ceph): osd: fix pg_pool_t::get_pg_num return types
- Unsigned!
Signed-off-by: Sage Weil <sage@newdream.net> - 05:18 PM Revision 877aa92b (ceph): osd: redo pg_t encoding with 64-bit pool id
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:27 PM Revision 458371e1 (ceph): Merge commit 'v0.34'
- 04:25 PM Revision 0b00350b (ceph): ceph_context: whitespace
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 02:27 PM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- Saw the e-mail, let's keep the discussion there to make sure everyone involved is up to date.
We could use this is... - 12:15 PM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- Hmm yeah, it's at least doable. I just sent an email to libvir-list asking about a more generic syntax for options.....
- 04:41 AM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- You can abuse the current libvirt implementation though....
- 04:48 AM Revision 2f039eee (ceph): v0.34
- 04:37 AM Feature #1423: qemu-kvm: allow user to specify client id
- Wasn't this one implemented with commit:20be92deef54e4ceb297e32fb2e98d4ef3868711 ?
I gave it a try and works fine ...
08/26/2011
- 11:54 PM Revision 44cb5ee2 (ceph): rgw: use formatter->dump_string() instead of dump_format()
- 09:11 PM Revision 0e8bd0dc (ceph): rgw: get bucket info only if bucket name is not empty
- 08:06 PM Revision fa757b7f (ceph): Makefile.am: install coverage files for libtool outputs as well
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 06:43 PM Revision 67cc408f (ceph): librados: watch request should wait for safe response
- 04:47 PM Revision 3a4d1936 (ceph): objectcacher: only want for commit
- There was some old, weird stuff going on here where we would wait for the
ACK and COMMIT separately. This is just wr... - 04:11 PM Revision 528703b9 (ceph): crush: comment fall-thru
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 04:11 PM Revision 09b08c5d (ceph): fix utime.h #include recursion
- Coverity cid 11, 13
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 04:11 PM Revision 85496ffa (ceph): mds: fix shadow
- Coverity cid 29
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 02:14 PM Bug #1376: errant scrub stat mismatch logs after upgrade
- it's a few weeks old yes, but there was no other evidence of of corruption (such as filesystem corruption).
I just... - 01:58 PM rgw Feature #1450 (Resolved): rgw admin remove user by access key
- this is needed in case the original user removal failed mid-way and there's some clutter left behind
- 01:53 PM Bug #1434: osds all failing each other
- Looks fixed with latest code.
- 12:53 PM Feature #1445 (Resolved): collectd: historical monitor stats
- 11:12 AM Bug #1449 (Resolved): osd: FAILED assert(0 == "we got a bad state machine event")
- I thought I submitted this already, but I can't find the ticket for it. Sorry if its a duplicate.
With my current...
08/25/2011
- 10:27 PM Revision fb33ef3c (ceph): thrasher: improve documentation a little
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:19 PM Revision 83e26342 (ceph): thrasher: add option to mark OSDs down instead of out.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:18 PM Revision 0f9b74e2 (ceph): thrasher: allow a config to set values
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:38 PM Revision 5d5de0e7 (ceph): thrasher: remove redundant wait_till_clean()
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:19 PM CephFS Bug #1444 (Resolved): client: crash on flush completion under blogbench
- 08:48 AM CephFS Bug #1444 (Resolved): client: crash on flush completion under blogbench
- ...
- 09:19 PM CephFS Bug #1391 (Resolved): client: crash on std::string in insert_trace()
- 09:18 PM Bug #1359 (Can't reproduce): fsstress workunit hang on cfuse
- running in a loop, no problems.
- 08:04 PM Revision 75e53e92 (ceph): mds: remove unused Mutation in do_rmdir_rollback()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:51 PM Revision 65b30507 (ceph): mds: allow setting all options on a file layout.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 07:51 PM Revision d066cc8f (ceph): mds: server: should apply new layout settings on top of old layout
- This way, the MDS can handle updates of some values without needing
the user to specify the entire layout (ie, they c... - 07:51 PM Revision df8f3cbb (ceph): cephfs: fix the stripe_count documentation.
- Whoops!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 07:51 PM Revision 4dba8bc2 (ceph): cephfs: use strtol instead of atoi; handle 0 properly
- Besides being generally better, this means we can accept pool 0
as the pool to store stuff in.
Signed-off-by: Greg F... - 07:51 PM Revision b8267492 (ceph): workunits: make file_layout test setting only one parameter
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 07:37 PM Revision 05063867 (ceph): objectcacher: remove uncommitted xlist<>; fix broken purge() cleanup
- There was a problem where:
- we would dirty some buffers on an object
- bump dirty_tx count
- flush()
- this... - 07:11 PM Revision a439e150 (ceph): rgw: remove some compilation warnings
- 07:06 PM Revision 1e929cee (ceph): rgw: don't override ret value in do_read_permisions
- 06:25 PM Revision b7b5f2e2 (ceph): rgw: return correct bucket missing error when needed
- 06:11 PM Revision e478e92d (ceph): rgw: fix 500 error on bucket creation
- 03:58 PM Feature #1443: qa suite: make basic tasks run on mutiple clients
- I think I wrote the workunit task to begin with; I can modify it!
- 03:57 PM CephFS Bug #1318: directories disappear across multiple rsyncs
- Looking at these symptoms again, I wonder if this could have been a result of the path_traverse changes we were makin...
- 03:44 PM Bug #1376: errant scrub stat mismatch logs after upgrade
- Missing objects on disk sure make it look like data corruption. Your cluster's pretty old, right? Is it still in this...
- 03:42 PM Bug #1407: _scrub crash on missing clones
- Is this something we're likely to be able to track down without logs? (Especially now that it no longer asserts?)
... - 03:38 PM Feature #1410: teuthology: mark osds down in thrasher
- This makes most everything configurable: frequency of status changes, approximate frequency of wait_till_clean()s, ch...
- 03:37 PM Feature #1410 (Resolved): teuthology: mark osds down in thrasher
- Completed in commit:fb33ef3c69f6e5237c85c18b691f03545d1fce05
- 10:03 AM Feature #1410 (In Progress): teuthology: mark osds down in thrasher
- 01:08 PM CephFS Bug #1446 (Resolved): cephfs: pool option doesn't work
- Fixed in commit:65b30507590e9ef47623b7bfe1e672aba01ce823
- 09:14 AM CephFS Bug #1446 (Resolved): cephfs: pool option doesn't work
- While testing the pool layout option, it's accepted, but reading back the pool it's still located in pool 0.
This ... - 01:01 PM CephFS Bug #1405 (Resolved): cephfs: shouldn't have to specify all layout options
- Fixed in userspace commit:b8267492551f1adc5e0079a670b20f6180de18f0
and kernel client commit:7c296cadd05d28329e595b... - 08:14 AM CephFS Bug #1405: cephfs: shouldn't have to specify all layout options
- The in-kernel code rejects any layout that doesn't set the stripe unit (and if you set the object_size it makes sure ...
- 12:51 PM CephFS Feature #1448 (Resolved): test hadoop on sepia
- - set it up on some sepia nodes (8?)
- do some basic testing of ceph vs hdfs
from doug cutting:... - 12:46 PM Bug #1230 (Can't reproduce): cfuse: rm -r missed a directory
- i've run this a gazillion times now in a loop w/o problems.
- 12:45 PM CephFS Bug #1368 (Can't reproduce): mds crash after blogbench on cfuse
- 12:45 PM CephFS Bug #1367 (Can't reproduce): cfuse and mon crash after dbench
- 11:19 AM rgw Bug #1408 (Resolved): rgw: racing PUT to same object, can cause 404 on GET/PUT
- moving this to resolve for now. There were a few issues that were solved, and currently can't reproduce.
- 10:01 AM CephFS Bug #1447 (Resolved): mds: does not validate pool IDs in handle_client_set[dir]layout
- Yep, there's no checking that they're valid mds data pools or even that they exist!
- 09:02 AM Feature #1445 (Resolved): collectd: historical monitor stats
- track how long PGs have been in any interesting states ('degraded', 'peering', 'down') by bin (30 sec, 1, 2, 5, 10, 3...
- 12:32 AM Revision 9538e87e (ceph): rgw: fix clitest
- 12:20 AM Revision 2488e446 (ceph): mds/Server.c: fix memory leak in err path
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:20 AM Revision 5fb216e0 (ceph): MonClient: fix possible null dereference
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:12 AM Revision 83a248b5 (ceph): add multiclient collection with a locking task
- 12:07 AM Revision 44c68b1c (ceph): rgw: don't check for zero content when creating a bucket
- 12:01 AM Revision 94fefce5 (ceph): rgw: rgw_admin bucket stats dumps owner
08/24/2011
- 11:54 PM Revision 6dce3b1a (ceph): client: drop useless insert_trace timestamp argument
- This is reachable via request. It was also misnamed in the header :/.
Signed-off-by: Sage Weil <sage@newdream.net> - 11:51 PM Revision 3ef32bc6 (ceph): client: be careful about replacing dentries during readdir assimilation
- When we are assimilate readdir results into our cache, we need to be more
careful about replacing existing dentries. ... - 11:48 PM Revision 1130e5fe (ceph): coverage: create dir conditionally
- We don't need to create the dir if we aren't using coverage.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 11:41 PM Revision 0b1ad608 (ceph): rgw: add radosgw_admin bucket stats
- 11:39 PM Revision 0840d05a (ceph): lockfile: add a lockfile task
- This allows pretty highly configurable testing of
fcntl locking via a teuthology task.
Signed-off-by: Greg Farnum <g... - 11:25 PM Revision 2a4cf8f8 (ceph): client: factor update_dentry_lease() out of insert_dentry_inode()
- The first bit of insert_dentry_inode() handles the details of checking
whether an item still exists, un/relinking it,... - 10:51 PM Revision 53b3e146 (ceph): librados: fix memory leak in error path
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:46 PM Revision 94af6a4c (ceph): test/confutils.cc: fix resource leak
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:27 PM CephFS Bug #1405: cephfs: shouldn't have to specify all layout options
- This is going to be related to #1446, obviously. I'll take both.
- 10:26 PM Bug #1415: cosd assertion: existing->state == STATE_CONNECTING || existing->state == STATE_OPEN
- Do we know what version this was run on? This looks to me like the assert we saw when OSDs were connecting to themsel...
- 10:05 PM Revision 103a1983 (ceph): more kclient workunit/suites
- 10:04 PM Revision f2381f97 (ceph): client: carry inode reference across calls to _flush
- Also drop the unused _flush Context* argument.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 10:02 PM Revision 76123404 (ceph): add fsx workunit under cfuse, kclient, rbd
- 10:02 PM Revision 2ba01bbb (ceph): rgw: fix prepare_get_obj when ENOENT
- 09:50 PM Revision ebae5a72 (ceph): rbd: fix shared_ptr usage
- shared_ptr calls the disposal function even when the pointer is null
that is being disposed of.
Signed-off-by: Colin... - 09:43 PM Revision 57fd56cf (ceph): do_kernel_showmapped: fix resource leak
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:40 PM Revision d34743da (ceph): Merge remote branch 'origin/next'
- 09:38 PM Revision d69aa590 (ceph): client: fix readdir + cap recovation/reissue race
- The current race:
- we start readdir
- mds revokes dir cap
- some file gets removed
- mds reissues dir cap
- we ... - 09:15 PM Revision 2e1f2c1d (ceph): test_rados_tool.sh: test export with --workers
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:15 PM Revision ad1b2b0c (ceph): CephBroker.cc: fix allocation/deallocation mismatch
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:15 PM Revision a1afa6d7 (ceph): rados_sync: fix resource leak on err
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:15 PM Revision 6fd4f3a1 (ceph): rados_sync: make number of worker threads configurable
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:17 PM Revision 76718b08 (ceph): rgw: clear warning
- 07:29 PM Revision 1a002741 (ceph): Merge branch 'wip-decouple-bucket'
- 06:01 PM Revision e9b739f8 (ceph): journaler: fix waiting_for_zero clearing, again
- We can only clear this when we have >= a period between flush_pos and
write_pos.
Clear the flag in _do_flush() so th... - 05:44 PM Feature #1413: librbd asynchronous copy
- What would such an API look like? Copy is implemented using read_iterate... we could add a user-supplied callback th...
- 05:43 PM CephFS Bug #1391: client: crash on std::string in insert_trace()
- 03:11 PM CephFS Bug #1391: client: crash on std::string in insert_trace()
- Hmm, any other hints on what workloads might trigger this? I'm not getting anything from valgrind or my workloads.
... - 11:23 AM CephFS Bug #1391 (New): client: crash on std::string in insert_trace()
- Reopening this...
- 05:23 PM Bug #1314 (Resolved): teuthology: tasks for file lock behavior on process exit
- And done. Pushed a new collection using this task to ceph-qa-suite too.
- 05:00 PM Bug #1314: teuthology: tasks for file lock behavior on process exit
- Finally done and pushed. Just waiting to resolve it until I add some yaml to the qa suite; need to check on a few thi...
- 05:11 PM Feature #1443 (Resolved): qa suite: make basic tasks run on mutiple clients
- add a multiclient cluster config to the basic clusters facet.
change all of the basic tasks to run on N clients (r... - 05:03 PM Revision c2c5ac0c (ceph): lock: --list-targets: list locks and dump result in targets: yaml format.
- 04:57 PM rgw Feature #1431 (Resolved): rgw: get stats for specific pool by pool id
- commit:94fefce5ea6a04074739b16f3c251f1f7d35c82c
$ ./radosgw_admin bucket stats --bucket-id=63 --format=json - 04:38 PM Revision 23d94a4c (ceph): Revert "pipe.c: #define GNU_SOURCE"
- This reverts commit ae9ba2249e372e7435e56f54fc2d8e22f3e42ff8.
This is handled in master by the Makefile. - 04:38 PM Revision c2e4047a (ceph): Merge branch 'next'
- 04:31 PM Revision 6c6fa6df (ceph): client: only link directories into hierarchy once
- Directories can only link into the hierarchy once. We assert as much
in readdir_r_cb(). Fix link() so that it unlin... - 04:10 PM Revision ba44cdeb (ceph): journaler: fix waiting_for_zero clearing
- Clear this flag only if we know no subsequent flushes could be waiting on
a prezero operation.
Fixes MDS journaling ... - 04:10 PM Revision eaaefece (ceph): mon: fix store_state
- We need to bound the range we write to disk by m->last_committed; this got
lost in translation in commit dfab2c91f532... - 03:10 PM CephFS Bug #1442 (Resolved): client: non-empty ObjectSet on last inode->put()
- fixed by commit:f2381f97dea9f3563897857c0a0482281b449b61
- 12:56 PM CephFS Bug #1442 (Resolved): client: non-empty ObjectSet on last inode->put()
- commit:e9b739f8dd39f3373dd0869a0fd5436350e1e3f3...
- 02:30 PM Bug #1434: osds all failing each other
- 09:25 AM Bug #1434: osds all failing each other
- That assert is fixed 4 commits later by commit:cf3b7cf6a9d3f873ad27a313cc1635822bdd89a1. Can you still reproduce wit...
- 02:17 PM Feature #1419 (Resolved): rados export: parallelize to be faster
- 01:15 PM Bug #1369 (Can't reproduce): ffsb hang on cfuse (messenger?)
- I've run this a gazillion times now. There were many hangs, but they were a result of mon crashes and other bugs tha...
- 01:12 PM CephFS Bug #1429 (Resolved): cfuse assert failed assert(diri->dn_set.size() < 2)
- fixed by commit:6c6fa6dffddb6f388d03ca59e95844ddf845f491
- 12:54 PM Linux kernel client Bug #1441 (Resolved): invalidation skips locked pages
- ceph_invalidate_work() is skipping locked pages (e.g. those concurrently being written). it needs to be synchronous ...
- 11:21 AM Bug #1426 (Resolved): osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type =...
- 09:17 AM Bug #1440 (Resolved): mon: irtual void PGMonitor::encode_pending(ceph::bufferlist&)
- Tracked this down last night.. should be fixed by commit:eaaefece778b042b91d7c7e7fd8e00d418d7e93a
- 07:18 AM Bug #1440 (Resolved): mon: irtual void PGMonitor::encode_pending(ceph::bufferlist&)
- After upgrading to v0.33 I'm encountering:...
- 04:00 AM Revision 42318c57 (ceph): check ceph cluster log for badness (ERR, WRN, SEC)
- 01:53 AM Revision d630693b (ceph): Merge branch 'master' into wip-decouple-bucket
- 01:52 AM Revision 2144c7a4 (ceph): add missing include, required for certain environments
- 01:52 AM Revision 190b8936 (ceph): Merge remote-tracking branch 'origin/master' into wip-decouple-bucket
- Conflicts:
src/rgw/rgw_rados.cc - 01:33 AM Revision adf1efe9 (ceph): rgw: fix multipart upload
- don't try to reset meta object with category
- 01:12 AM Revision d81d4580 (ceph): rados_sync:don't need to declare CHANGED_XATTRS 2x
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 01:12 AM Revision 32694765 (ceph): librados: fix IoCtx copy constructor
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 01:12 AM Revision b20f3622 (ceph): test_rados_tool.sh: fix xattr test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 01:12 AM Revision b76e054b (ceph): WorkQueue: don't be chatty about finishing workers
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 01:12 AM Revision 1a7eac4d (ceph): parallelize rados export and import
- * use workqueues to parallelize rados export and import
* Put export and import into separate files
* Fix RADOS_SYN... - 01:12 AM Revision 31bf064b (ceph): Some fixes to double-dash support, fix warning
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 01:06 AM Revision 301ae6c5 (ceph): rgw: if pool is empty recreate
08/23/2011
- 11:30 PM Revision eecf4d61 (ceph): ceph_argparse: handle double dashes consistently
- Handle double dashes in the ceph_argparse functions, so that any piece
of code doing argument parsing will correctly ... - 10:46 PM Revision 084fe47f (ceph): injectargs test: test foo=true
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:46 PM Revision 42a65b8d (ceph): config: print warning on injectargs parse fail
- injectargs should print a warning message when its parser fails.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost... - 10:46 PM Revision 18ec5655 (ceph): injectargs: binary flags only consume 1 token
- Never consume more than one token when parsing a binary flag.
Basically, your choices are --foo=false, --foo=true, a... - 10:24 PM Revision ddc56742 (ceph): mon: allow wildcards for mon cap command matching
- We need wildcards, actually. * is any token, ... is 1 or more tokens.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:26 PM Revision 34327293 (ceph): common/config: fix get_val for booleans
- md_config_t::get_val should return true/false for booleans, not 0/1.
This is for consistency with the setter.
Signed... - 09:12 PM Feature #1263 (Resolved): teuthology: scan cluster log for badness
- 09:12 PM Revision ae9ba224 (ceph): pipe.c: #define GNU_SOURCE
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:03 PM Revision 9c498e0f (ceph): qa: add unit tests for mon cap 'allow command ...' stanzas
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:56 PM Revision 6c3d8103 (ceph): mon: whitelist individual commands in mon cap string
- This lets you whitelist a command prefix. For example,
'allow command osd foo'
will let you do any command that b... - 07:59 PM Revision ea0f2f30 (ceph): mon: adjust max_devices when adding to crush map
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:58 PM Revision 58f98e7a (ceph): crush: refuse to add item >= max_devices
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:47 PM Revision dc2be1d8 (ceph): mon: 'osd gstcrushmap' or 'osd crush set' both work
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:47 PM Revision a4f32ba7 (ceph): mon: fix arg count checks
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:40 PM Revision d896d401 (ceph): mon: 'osd crush reweight <name> <weight>'
- Reweight a node in the crush map.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:33 PM Revision def8382e (ceph): mon: 'osd crush rm <name>'
- This is similar to crushtool --remove-item.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:31 PM Revision db8ad462 (ceph): mon: 'osd crush add <id> <name> <weight> [<loc1=foo> [<loc2=bar> ...]]'
- This is identical to the crushtool --add-item functionality, but does not
require a racy and tedious
- get crush ma... - 07:26 PM Revision 051c4a36 (ceph): mon: return error message string when no other output string specified
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:56 PM Revision 29cf7333 (ceph): common/config: honor --
- Argument parsing should stop after --
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 06:23 PM Revision 6d292397 (ceph): mon: remove dead 'setmap' code
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:21 PM Revision deb25e00 (ceph): mon: make auto-marking of booting osds 'in' optional
- By default we mark a booting OSD 'in' so that it immediately gets data.
Make this optional.
Eventually we may want t... - 06:20 PM Revision 09f4c464 (ceph): client: kill dentry_of() macro
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:04 PM rgw Bug #1439 (Resolved): multipart meta object is not atomic
- we should use tmap for it.
- 04:25 PM Revision 5e97a1cb (ceph): Merge branch 'next'
- 04:24 PM Revision 7ef44b73 (ceph): osd_types: fix category encoding in object_info_t
- This was mis-merged in f6ff0f14e9a5b70d5ce0277975d321f10e3b39c7.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:38 PM Bug #1438 (Resolved): gitbuilder kernel versioning is considered broken
- I upgraded a sepia node to ubuntu 11.04 to test kvm migration. It keeps repeating:...
- 11:33 AM CephFS Bug #1437 (Can't reproduce): cfuse can't change permissions of a file
- I've hit a case where I cannot change the permissions of a script to 755.
> chmod 777 ./extract_full.sh
> ls -... - 10:33 AM CephFS Bug #1391: client: crash on std::string in insert_trace()
I've been seeing a segfault in a similar spot regularly, but its been hard to reproduce. The segfault is always in...- 10:14 AM rgw Feature #1436 (Resolved): rgw_admin user info output should be formatted
- All rgw_admin output for user info should be formatted, or at least can dump formatted data when --format is specified.
- 05:04 AM Revision 21d04419 (ceph): ceph: copy cluster log file to archive/ceph.log
- 04:18 AM Revision d5a9301f (ceph): qa: define _GNU_SOURCE for qa helpers; add btrfs dir makefile
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:21 AM Revision ba4551d1 (ceph): librados: whitespace
- Don't do
some::class::
method(int foo)
{
}
...because then I can't '^S::method'. :)
Signed-off-by: Sage Weil <sag...
08/22/2011
- 11:53 PM Revision 39b6a9c4 (ceph): qa: add rbd copy workunit
- Test copying an image and snapshot.
Signed-off-by: Sage Weil <sage@newdream.net> - 11:53 PM Revision 76e18e3d (ceph): librbd: fix copy
- This was croaking on sparse images.
Use an image ctx for the dest.
We should probably use read_iterate instead.
Si... - 11:53 PM Revision f19e6131 (ceph): rbd: fix check_io for snap
- Need to look at image size for snapshot, not the head version.
Signed-off-by: Sage Weil <sage@newdream.net> - 11:53 PM Revision 3205441d (ceph): librbd: reimplement copy using read_iterate
- Ah, so much simpler!
Signed-off-by: Sage Weil <sage@newdream.net> - 11:02 PM Revision a585c4d4 (ceph): librbd: specify copy src image as image handle
- By specifying via an image handle we can set the snapshot on the src
image. This also makes the API cleaner.
Fixes:... - 11:00 PM Revision 4a17d71c (ceph): mdcache: handle unreadable null dentries
- Previously, we were asserting (farther down) if the dentry was
null but blocked by a lock. Instead, we should here be... - 11:00 PM Revision a1433a1d (ceph): mds: remove the unused MDS_TRAVERSE_FAIL flag.
- This isn't used, and judging by the comments was a horrible lie.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhos... - 11:00 PM Revision e898fe7d (ceph): mdcache: Reduce the gratuitous references to onfail
- We've been setting flags at the beginning of path_traverse
for a while; use those instead of ongoing comparisons.
Si... - 11:00 PM Revision b03a1841 (ceph): mdcache: documentation for path_traverse
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:35 PM Revision 3ce41446 (ceph): Define _GNU_SOURCE in src/Makefile.am
- We end up needing _GNU_SOURCE in a bunch of places-- to get direct i/o,
pipe2, and some other Linux-specific interfac... - 10:23 PM Feature #1409 (Resolved): teuthology: gather mon log
- 10:06 PM Revision 4e9a0b6d (ceph): pipe.c: #define GNU_SOURCE
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:59 PM Revision 4b7b5ec7 (ceph): Merge branch 'next'
- 08:58 PM Revision 3d85872a (ceph): mds: open+pin stray dirfrags on startup
- This ensures that the stray dirfrags are always open, which in turn ensures
that whenever we add straydn items the rs... - 08:19 PM Revision 3f134257 (ceph): Revert "mds: do not complain/assert about stray inode rstat/fragstat co...
- This reverts commit d3d767a2a36caefdeb8a94abba1673cf8304c793.
A better solution is to keep the stray frags loaded+pi... - 08:11 PM Revision b00b6f68 (ceph): common/pipe: add function to do pipe2(O_CLOEXEC)
- Support old systems that don't have O_CLOEXEC.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 08:11 PM Revision 93c0565e (ceph): PG: Move reset_last_warm_restart to Initial::exit
- Previously, reset_last_warm_restart was only invoked when handle_create
was used. This misses cases where the pg is ... - 06:13 PM Revision abe5cd28 (ceph): common/pipe: add function to do pipe2(O_CLOEXEC)
- Support old systems that don't have O_CLOEXEC.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 05:20 PM Revision cf3b7cf6 (ceph): PG: Move reset_last_warm_restart to Initial::exit
- Previously, reset_last_warm_restart was only invoked when handle_create
was used. This misses cases where the pg is ... - 04:59 PM Bug #1416 (Resolved): rbd: "cp --snap=mysnap myimage newimage" crashes
- the --snap specifies the src snap to copy from. the target is always the head of a fresh image.
fixed the librbd ... - 04:24 PM Revision 3e915254 (ceph): osd: flush writes prior to PGLS
- Make sure previous writes (which will include both object creation and
PG creation) are flushed before listing object... - 04:12 PM CephFS Bug #1435 (Resolved): mds: loss of layout policies upon mds restart
- Cluster running ceph 0.33 + patch to add support for “ceph mds add_data_pool”.
I set up layout policies for variou... - 04:09 PM CephFS Bug #1433 (Resolved): mds: assert in path_traverse
- Fixed by commit:b03a1841b4b08c82fa37a45dc31a0c0255949235
- 03:10 PM CephFS Bug #1433: mds: assert in path_traverse
- Yep! Need it to wait in that case. Pushing as soon as I write some documentation for path_traverse.
- 02:46 PM CephFS Bug #1433: mds: assert in path_traverse
- Looks like it's bailing out because another client is holding a lock, so the (existing) null dentry isn't readable. R...
- 01:44 PM CephFS Bug #1433 (Resolved): mds: assert in path_traverse
- While testing my teuthology lock test: ...
- 04:01 PM Revision c412da8a (ceph): osd: remove dead code, whitespace
- Signed-off-by: Sage Weil <sage@newdream.net>
- 03:58 PM Bug #1434: osds all failing each other
- reproduced, but with "debug ms = 1" this time.
now I notice that all but 1 osd crashes:... - 03:46 PM Bug #1434 (Resolved): osds all failing each other
- build of git tree 3a623bb327
cluster boots, then within a minute all the osds start failing each other. logs show... - 03:16 PM CephFS Bug #1366 (Can't reproduce): mds segfault
- 02:05 PM Bug #1426: osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type == Query::MI...
- 10:19 AM Bug #1426: osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type == Query::MI...
- This is probably caused by the same last_warm_restart as before. My previous patch handled the case where the handle...
- 08:12 AM Bug #1426: osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type == Query::MI...
- type=0 is INFO.
- 07:27 AM Bug #1426: osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type == Query::MI...
- (gdb) p query.query.type
$3 = 0
(gdb) p query.query
$4 = (const PG::Query &) @0x78db028: {type = 0, since = {versi... - 07:15 AM Bug #1426 (Resolved): osd assert fail ../../src/osd/PG.cc: 4271: FAILED assert(query.query.type =...
- I'm seeing assertion failures on a few nodes on my cluster. 10/33 hit the following assertion, the other 17 remain r...
- 02:05 PM CephFS Bug #1428 (Resolved): MDS: Load and pin stray dirs in memory
- 10:28 AM CephFS Bug #1428 (Resolved): MDS: Load and pin stray dirs in memory
- MDCache::populate_mydir() already does some of this; we also need to it load each frag and set the STICKY flag on the...
- 02:01 PM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- Idea from gregaf was to "abuse" the name attribute to do so, have to look into this.
We should check this with the... - 01:52 PM Bug #1432: libvirt: fix definition for rbd params/sources/etc
- The virtual disk was added to libvirt in 036ad5052b43fe9f0d197e89fd16715950408e1d.
It only lets you specify server... - 12:07 PM Bug #1432 (Resolved): libvirt: fix definition for rbd params/sources/etc
- 01:26 PM CephFS Bug #1429: cfuse assert failed assert(diri->dn_set.size() < 2)
- Hopefully -- we'll have to reproduce with logging and check it out in more detail. My concern is that it may be revea...
- 01:23 PM CephFS Bug #1429: cfuse assert failed assert(diri->dn_set.size() < 2)
- Sounds like an easy fix. For dirs it should just unlink the old link in insert_trace (or whatever it is).
- 01:19 PM CephFS Bug #1429: cfuse assert failed assert(diri->dn_set.size() < 2)
- There's probably something wonky going on with the way the client is handling moved directories -- that assert is bec...
- 11:28 AM CephFS Bug #1429 (Resolved): cfuse assert failed assert(diri->dn_set.size() < 2)
- This assertion happens when a directory is moved on one client, and then the other client changes to that directory. ...
- 12:03 PM rgw Feature #1431 (Resolved): rgw: get stats for specific pool by pool id
- radosgw_admin bucket info --bucketid=123
may include new message to monitor to get the correct stats
dumps ... - 09:32 AM Bug #1427 (Resolved): osd: pgls needs to be ordered wrt writes
- commit:3e9152546e6b3294b16db0afa1017df3701a4b47
- 08:36 AM Bug #1427 (Resolved): osd: pgls needs to be ordered wrt writes
- If you do an object write then read, the ObjectContext ensures things are ordered (waits for the write to his the fs ...
- 12:53 AM Revision 6decfa74 (ceph): Makefile: allow gtest to be linked in statically
- Use UNITTEST_STATIC_LDADD instead of UNITTEST_LDADD.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 12:53 AM Revision 2fb3af45 (ceph): librados: statically link gtest into librados unit tests
- This way they can run via teuthology.
We should do this for anything that we expect to run outside of the
immediate ... - 12:53 AM Revision 3a623bb3 (ceph): librados: prefix api test pools
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:27 AM Revision fc4da528 (ceph): rados_api_tests: run on top of cfuse, not rbd
- Slightly less inclined to break (e.g. due to old kernel)
- 12:26 AM Revision e79dda9a (ceph): workunits: set CEPH_CONF environment
- This allows any ceph util we run (including the rados-api tests) find
the config and keyrings they need.
08/21/2011
- 11:46 PM Revision 239f61e7 (ceph): .gitignore: scratchtool
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 11:45 PM Revision b02a3471 (ceph): Makefile: rule to upload + submit coverity job
- You need to
make project.tgz
make submit-coverity
and have .coverity.build.pass with the right password in CWD
(c... - 11:42 PM Revision aa3a8dd6 (ceph): addr_parsing: fix warning
- warning: include/addr_parsing.c:45: ‘saveptr’ may be used uninitialized in this function
Signed-off-by: Sage Weil <s... - 11:26 PM Revision 7d2f4a51 (ceph): escape: fix snprintf args
- Include space for null termination.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 11:17 PM Revision 952009c6 (ceph): .gitignore: scratchtool
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 10:14 PM Revision aa575c13 (ceph): rbd: make default image 10G instead of 1G
- 09:34 PM CephFS Bug #1425 (Resolved): mds: stuck in prexlock
- See mds.a.log on sepia78.
- setattr request starts locking
- auth_pins auth stuff
- rdlocks parent dirs, does no... - 09:24 PM Revision febe2327 (ceph): librados: fix shadow
- Coverity cid 23
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:23 PM Revision 244b6ee3 (ceph): mds: fix shadow
- Coverity cid 24
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:22 PM Revision 56b1484d (ceph): mds: fix shadow
- Coverity cid 25
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:22 PM Revision 6de1e538 (ceph): mds: fix shadow
- Coverity cid 26
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:21 PM Revision 6b8ccb21 (ceph): mds: fix shadow
- Coverity cid 27
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:20 PM Revision e6467ef4 (ceph): mds: fix shadow
- Coverity cid 28
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:18 PM Revision b3821cc5 (ceph): moncaps: fix shadow
- Coverity cid 30
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:17 PM Revision 14096a8f (ceph): monclient: fix shadow
- Coverity cid 31
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:16 PM Revision 0d73ad8e (ceph): monclient: fix shadow
- Coverity cid 32
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:15 PM Revision 1cbc242c (ceph): msg: fix shadow
- Coverity cid 33
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:14 PM Revision 87a50261 (ceph): osd: fix shadow
- Coverity cid 34
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:12 PM Revision 3283e8ee (ceph): osdcaps: fix shadow
- Coverity cid 35
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:10 PM Revision b5a2ad0e (ceph): escape: use snprintf
- Coverity cid 41, 40
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:07 PM Revision 41745a91 (ceph): secret: use strncpy, snprintf
- Coverity cid 42
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:03 PM Revision 5fb99fa7 (ceph): addr_parsing: use strncpy
- Coverity cid 43
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:03 PM Revision 05c38faa (ceph): addr_parsing: avoid strcpy
- Make coverity shut up.
Coverity cid 44
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:00 PM Revision ef80b0ac (ceph): mount: fix path leak
- canonicalize_path() does not free/touch it's argument; no need to strdup()
here.
Coverity cid 39
Signed-off-by: Sag... - 08:56 PM Revision 2ebf1414 (ceph): mount: avoid unguarded strcpy
- Use strdup here, mostly to make coverity shut up.
Coverity cid 45
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:52 PM Revision 40fcd2e1 (ceph): test_ioctls: reduce stack use
- Coverity cid 51
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:51 PM Revision 8b0d0651 (ceph): formatter: less big buffer for dump format string
- Coverity cid 53
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:50 PM Revision 3ff9068e (ceph): mount: avoid big stack item
- Coverity cid 54
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 08:49 PM Revision b920fac4 (ceph): mount: avoid big stack item
- Coverity cid 55
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 06:23 PM Feature #1353 (Resolved): qa: full librados c++ api gtest coverage
- fixed (statically linked now)
- 02:56 PM Feature #1353 (In Progress): qa: full librados c++ api gtest coverage
- Bleh, when run via teuthology the binaries can't find libgtest.so. Probably needs to be included in the tarball we'r...
- 04:05 PM Revision c6156851 (ceph): testlibrbd: add missing va_end()
- Coverity cid 57, 58
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 06:32 AM Revision 29c1ce55 (ceph): rgw: remove redundant line
- 03:42 AM Revision d3d767a2 (ceph): mds: do not complain/assert about stray inode rstat/fragstat consistency
- We instantiate the stray dirfrags without reading the fragstat off of disk
because it's faster, we know the dentry is...
08/20/2011
- 10:04 PM Revision f6c07efd (ceph): paxos: fix store_state
- Stop if we empty out the map completely.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 09:21 PM Bug #1369: ffsb hang on cfuse (messenger?)
- unlocked nodes
- 04:56 PM Revision 09356734 (ceph): monclient: be less chatty about stray messages
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 02:32 PM Revision 64ce3030 (ceph): debian: librbd headers require librados headers
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:33 PM Bug #1376: errant scrub stat mismatch logs after upgrade
- Just tried writing some data to the ceph filesystem on this cluster and got this message:...
- 11:14 AM rgw Feature #1424 (New): rgw: short-term object version retention policy
- Version all object puts, like when full versioning is enabled on the bucket, but
- only keep each version for a l... - 11:04 AM Feature #1377: teuthology: use btrfs to back osds
- I think the key here is to have some per-node metadata indicating which disks (e.g. /dev/sdb) are available as scratc...
- 10:56 AM Feature #685: libcephmon: interact with ceph monitors via a library
- The monitors now dump most of the cluster state in json, so this gives us more reliable visibility. You still need t...
- 10:50 AM Feature #1423 (Resolved): qemu-kvm: allow user to specify client id
- specify id=foo (for client.foo) in conf string so you aren't stuck with client.admin.
already pushed a patch to do... - 09:07 AM Feature #1422 (Resolved): libvirt: rbd storage pool
- The libvirt storage pools let you set up a location/source and then browse images. This would use librbd to browse/m...
- 08:57 AM rbd Feature #341 (Resolved): libvirt bindings
- 06:40 AM Revision c75f0fca (ceph): scratchtool: not a test suite
- 06:36 AM Revision ab40c3c0 (ceph): Revert "Remove old rados tests"
- This reverts commit 665fd8214d4f8fa54a675c2998b22613424a6cb3.
Conflicts:
src/testradospp.cc - 06:32 AM Revision 331d0aa7 (ceph): osd: fix compilation
- 05:54 AM Revision c095f74f (ceph): rgw: fix bucket removal
- 03:37 AM Revision 6fca6249 (ceph): osd: call incorrect osdmap addrs errors
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:35 AM Revision 3794bd9c (ceph): rgw: fail operation when trying to clone to the same shadow
- not necessary to recreate the same shadow, and also provides mutual
exclusion
08/19/2011
- 11:23 PM Revision 410d963e (ceph): add multimds subtree thrashing test
- 11:21 PM Revision 1dfdc730 (ceph): rgw: fix prepare_atomic_for_write
- 11:18 PM Revision a42d9a80 (ceph): add rados_api_tests.yaml
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:16 PM Revision a35ac29b (ceph): mds: only client hold xlock in xlockdone can change lock state
- If we are in xlockdone, only the client holding that xlock can adjus the
lock state (e.g., relock). Other clients ha... - 11:16 PM Revision e5cf2a4f (ceph): mds: xlocker_caps are supplemental to caps
- Like loner_caps, xlocker_caps are additional caps to the any caps.
In practical terms, this only affects (currently)... - 09:14 PM CephFS Bug #1367: cfuse and mon crash after dbench
- nuked and unlocked nodes, nothing useful there.
- 09:10 PM Feature #1353 (Resolved): qa: full librados c++ api gtest coverage
- 09:09 PM Bug #1392 (Resolved): osd: PGLog received in Initial state
- being optimistic and closing this unless/until we see it again.
- 08:53 PM Revision ac248f1a (ceph): mds: update truncate_pending on O_TRUNC; pass old_size on O_TRUNC open
- We were missing truncate_pending++ in O_TRUNC path. Use a common helper.
Pass old_size on O_TRUNC open. (This was ... - 08:07 PM Revision 2d0bd2df (ceph): rados-api: fix #include order
- Fixes conflict with linux types on fatty.
Signed-off-by: Sage Weil <sage@newdream.net> - 08:05 PM Revision 513f6128 (ceph): cconf: fix clitest
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:56 PM Revision f3396bdc (ceph): rgw: add missing cache locking
- this was overlooked when switch to the multithreaded configuration
- 07:54 PM Revision f5b012c8 (ceph): mds: fix LOCK_LOCK_XLOCK lock state name/string
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:54 PM Revision 72044499 (ceph): cconf: --filter-key and --filter-key-value
- Filter out section list to sections that define a given key or that
set a key to a given value.
This is useful to, f... - 07:33 PM Revision 0b216bdc (ceph): rgw: was missing return value
- 07:27 PM Revision 32db9dc5 (ceph): rgw: some more race fixes
- 04:53 PM Revision ed580c55 (ceph): mon: whitespace
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:53 PM Revision bec73904 (ceph): monitorstore: implement put_bl_sn_map
- This will commit a big batch of values safely. We push it down into
MonitorStore so we can avoid an fsync(2) per fil... - 04:53 PM Revision dfab2c91 (ceph): paxos: use MonitorStore::put_bl_sn_map() to commit batches of values
- This allows us to (safely) do fsync vs sync optimizations. The old code
would write values to the final names and th... - 04:53 PM Revision f0ab8b06 (ceph): monitorstore: drop unused sync value
- This is inherently unsafe, because we write to a final filename without
ensuring the data is on disk.
Signed-off-by:... - 04:51 PM Revision e759021b (ceph): test/rados-api/misc.cc: update tests after API change
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:38 PM CephFS Bug #1417 (Resolved): mds: failed assert on xlock
- Well, I hit a path_traverse bug instead. I'm going to mark this particular one as resolved unless it pops up again.
- 04:26 PM CephFS Bug #1417: mds: failed assert on xlock
- Testing that fix I worked out with Sage.
- 02:33 PM CephFS Bug #1417: mds: failed assert on xlock
- Okay:
1) dispatch client1 request, gets xlock on filelock (lock_xlock)
2) early_reply to client1 request, which cal... - 09:17 AM CephFS Bug #1417 (Resolved): mds: failed assert on xlock
- ...
- 04:38 PM Feature #1421 (Resolved): build release deb for oneiric
- 04:37 PM Feature #1420 (Resolved): build release rpms
- centos
scientific linux
rhel 6
are the ones people seem to want. - 04:15 PM Revision 635bbe83 (ceph): atomic_t: update spinlock implementation.
- Somebody added a 'set' function without adding the non-atomic_ops
version!
Signed-off-by: Greg Farnum <gregory.farnu... - 03:00 PM Feature #1411: teuthology: scan valgrind log for badness
- Valgrind can be run in a mode where it outputs XML.
generally, you would do it like this:... - 12:01 PM Feature #1419 (Resolved): rados export: parallelize to be faster
- Currently, buckets with lots of small objects take a long time to backup with rados export.
To saturate the networ... - 09:59 AM Bug #1414 (Resolved): cmon crash Paxos::handle_collect (this=0x22a9000, collect=0x22e56c0) at ../...
- fixed by commit:dfab2c91f53289b1d527fb87126a639f5d5275f1
- 09:25 AM Bug #1414: cmon crash Paxos::handle_collect (this=0x22a9000, collect=0x22e56c0) at ../../src/mon/...
- I think i see the problem. Almost everywhere we write to a .new file, fsync, and then rename over. Except on place,...
- 09:11 AM Bug #1414: cmon crash Paxos::handle_collect (this=0x22a9000, collect=0x22e56c0) at ../../src/mon/...
- With debug mon = 20
2011-08-19 10:15:24.033788 7f4b5fb05760 ceph version .commit: . process: cmon. pid: 14962
201... - 08:04 AM Bug #1414 (Resolved): cmon crash Paxos::handle_collect (this=0x22a9000, collect=0x22e56c0) at ../...
- I see this repeatedly in my current setup:
Core was generated by `/usr/ceph/bin/cmon -i bravo -c /etc/ceph/ceph.co... - 09:38 AM Tasks #1418 (New): set up a no-atomic-ops gitbuilder
- When the atomic_t "set" operation got added, it was only put into the atomic_ops version, not the spinlock version. S...
- 08:56 AM Bug #1416 (Resolved): rbd: "cp --snap=mysnap myimage newimage" crashes
- ...
- 08:07 AM Bug #1415 (Duplicate): cosd assertion: existing->state == STATE_CONNECTING || existing->state ==...
- end of the log:
2011-08-18 19:45:52.794270 7f90a6161700 -- 192.168.101.11:6803/12333 >> 192.168.101.112:6819/10241... - 07:04 AM Feature #1413 (Resolved): librbd asynchronous copy
- It would be handy if there was an asynchronous version of the librbd copy method. Or at least some means of getting s...
- 06:36 AM Revision fb7831c3 (ceph): rgw: reread obj meta on racing write operation
- 05:59 AM Revision 4a1a64e9 (ceph): rgw: bucket removal through intent log, fixes
- 05:22 AM Revision f336fef2 (ceph): rgw: fix some issue, basic stuff works
- 04:57 AM Revision 40752e9b (ceph): rgw: create preallocated pools
- 01:15 AM Revision 8db7daf6 (ceph): ReplicatedPG: Don't crash when a clone is missing
- Print an error rather than crashing when _scrub finds a missing
clone.
Signed-off-by: Samuel Just <samuel.just@dream... - 12:36 AM Revision 9f5d1638 (ceph): file_layout.sh: add test for setting a new data pool
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 12:00 AM Revision 8c822a80 (ceph): Merge branch 'wip-librados-tests'
- Conflicts:
src/include/rados/librados.hpp
src/librados.cc
src/rgw/rgw_rados.cc
src/testradospp.cc
08/18/2011
- 11:34 PM Revision 665fd821 (ceph): Remove old rados tests
- These tests have been supplanted by the rados-api gtests.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 11:33 PM Revision eb8925a7 (ceph): rgw: fix locator for shadow objects
- 11:30 PM Revision cfd5456f (ceph): suite: support a suite consisting of multiple collections
- suite = many collections, and maybe some shared files
collection = a collection of facets
facet = a config fragment - 10:15 PM Revision 1546beaf (ceph): mds: share max_size on any client_range adjustment
- We were missing this on open + O_CREAT (now triggered by cfuse), on open +
O_TRUNC, and on mknod.
Signed-off-by: Sag... - 10:09 PM Revision 87e20c90 (ceph): rgw: add missing files
- 10:06 PM Revision b8b8f9dc (ceph): add WatchNotifyTimeoutTestPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:06 PM Revision 0c0071a4 (ceph): rgw: fix a double free in GetObj
- 09:51 PM Feature #1412 (Resolved): qa: spec out messenger testing
- use Colin's multiprocess framework to do a messenger test for reconnects, with socket failure injects.
types:
- ... - 09:44 PM Feature #1411 (Resolved): teuthology: scan valgrind log for badness
- the memory leak stuff and stupid library issues aside, i don't remember seeing any false warnings from valgrind. is ...
- 09:32 PM Feature #1410 (Resolved): teuthology: mark osds down in thrasher
- this exercises some additional code paths. shouldn't be _too_ frequent by default.
the 'wrongly marked down' thin... - 09:26 PM Feature #1409 (Resolved): teuthology: gather mon log
- include the cluster log ($mon_data/log) in archive
- 09:20 PM Revision 93b484b1 (ceph): rgw: prepare for bulk pool creation
- 08:21 PM Revision 3ebe6b77 (ceph): rgw: store bucket info
- 07:55 PM Revision d454c584 (ceph): test/rados-api/misc.cc: test CloneRange
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:52 PM Revision 17ee7f8f (ceph): test/rados-api/misc.cc: test CloneRangePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:32 PM Revision 9c8722a6 (ceph): test/rados-api/misc.cc: test AioOperatePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:32 PM rgw Bug #1408 (Resolved): rgw: racing PUT to same object, can cause 404 on GET/PUT
- I think what happens is that:
The following operations happen concurrently around the same time:
(1) PUT obj
(2)... - 06:16 PM Revision 8ce65447 (ceph): osd: fix _activate_committed() crash
- Do not dereference acting[0] unless we know it is still valid.
Take a reference when scheduling the transaction, and... - 06:15 PM Revision 0ae230b1 (ceph): test/rados-api/misc.cc: test Operate2PP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:12 PM Bug #1407: _scrub crash on missing clones
- 8db7daf should at least prevent it from asserting in this case. The question is how the pg came to be clean with tha...
- 06:11 PM Bug #1407 (Can't reproduce): _scrub crash on missing clones
- (05:07:56 PM) slang1: osd crash: http://pastebin.com/raw.php?i=A08FRV15
(05:10:20 PM) slang1: stack trace: http://... - 06:11 PM Revision 2903de0a (ceph): Operate1PP: test CEPH_OSD_CMPXATTR_OP_EQ
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:07 PM Revision 6d90796f (ceph): compatset: dump json
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:03 PM Revision f5f2cd1c (ceph): librados: remove unused operate param
- The last parameter of operate, aio_operate was unused and unchecked.
Remove it.
Signed-off-by: Colin McCabe <colin.m... - 04:39 PM Bug #1379 (Resolved): osd segfault during end of recovery
- 10:15 AM Bug #1379: osd segfault during end of recovery
- slang hit this too,
(10:08:17 AM) slang: http://pastebin.com/raw.php?i=fYFGnVPJ
- 03:44 PM Revision d21c918b (ceph): uninline a bunch of buffer.h stuff
- 03:44 PM Revision f6ff0f14 (ceph): uninline osd types
- 11:33 AM Bug #1404 (Resolved): cephfs: test pool option
- Well, there's a test. Unfortunately, it doesn't pass. #1406.
- 11:02 AM Bug #1404: cephfs: test pool option
- Also created #1405 for later.
- 11:30 AM CephFS Bug #1405: cephfs: shouldn't have to specify all layout options
- And you also need to specify these even if you only want to set the pool. :(
- 11:02 AM CephFS Bug #1405 (Resolved): cephfs: shouldn't have to specify all layout options
- Right now, you need to specify all layout options in cephfs (of the stripe unit, stripe count, and block size, anyway...
- 01:38 AM Revision df99cd9b (ceph): mds: add debug_mds_locker
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 12:11 AM Revision ff9537e5 (ceph): rgw: prepare for bucket/pool decoupling
- still not working
08/17/2011
- 11:54 PM Revision a724bc46 (ceph): librbd: rbd_list: return buffer size, not element count
- Fixes: #1397
Signed-off-by: Sage Weil <sage@newdream.net> - 11:53 PM Revision 92746916 (ceph): mon: add/remove data_pg_pools from mdsmap
- Signed-off-by: Sage Weil <sage@newdream.net>
- 11:35 PM Revision a88c1790 (ceph): PG: last_warm_restart must be set to the current epoch
- Previously, last_warm_restart did not get set until the
first time the pg was restarted.
Signed-off-by: Samuel Just ... - 11:12 PM Revision 47d75af5 (ceph): mds: explain weird encoding
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:53 PM Revision a21a5ef8 (ceph): mds: whitespace
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:47 PM Revision b4fb82bb (ceph): mds: drop unused projected dirfrag_t arg for EMetaBlob::add_primary_dentry
- Not used. If it is later, we need to project it properly.
Signed-off-by: Sage Weil <sage@newdream.net> - 10:43 PM Revision 93b71672 (ceph): mds: drop unused snapbl arg
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:41 PM Revision a9998c8d (ceph): mds: avoid passing snapbl around explicitly
- We now properly track projected snaprealm/sr_t's. Use that for journaling
instead of manually passing around a snapb... - 10:41 PM Revision d011c623 (ceph): mds: avoid explicit passing of projected_xattrs
- No need to pass this in explicitly; we can look in the projected inode for
it. This actually fixes a race where a jo... - 10:40 PM Revision 213ce6ab (ceph): test/rados-api/misc.cc: test Operate1PP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:33 PM Revision 08fb8a61 (ceph): mds: use snaprealm projection for MDCache::snaprealm_create()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:33 PM Revision 6ee52e7d (ceph): mds: simplify sr_t journaling
- Drop encode/decode in SnapRealm and encode the sr_t's directly. This way
the caller says what they mean (current or ... - 10:32 PM Revision d60d5319 (ceph): mds: journal inode on mksnap
- We were cowing but not journaling a simply-dirty inode. Der.
Fixes: #1389
Signed-off-by: Sage Weil <sage@newdream.net> - 09:11 PM Revision 79db9fc5 (ceph): mds: handle O_TRUNC when size is already 0
- We always want to go through the truncation path. Two reasons:
- even if the size is already 0, we still need to u... - 09:00 PM Revision 45740cf8 (ceph): client: send O_TRUNC open requests to MDS
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:59 PM Revision 266110d6 (ceph): client: move Inode operator<<
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:52 PM Revision f3260643 (ceph): Test rados_exec, C and C++ version
- * Test rados_exec, C and C++ version
* add test_exec action
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 08:18 PM Revision 3c3089ff (ceph): test/rados-api/misc.cc: test tmap_update
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:56 PM Revision d9df58e1 (ceph): rgw: fix virtual bucket name handling
- 07:56 PM Revision 0ed2fafc (ceph): rbd: remove dup 'rm' command from man page
- Fixes: #1395
Signed-off-by: Sage Weil <sage@newdream.net> - 07:53 PM Revision d3145064 (ceph): cfuse: take out accidental bad channel add
- This was a brainstorm that snuck into last commit, grr.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:41 PM Revision 184c2f5d (ceph): qa/workunits/rados/test.sh:explicitly invoke tests
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:35 PM Revision 930bd329 (ceph): Add qa/workunits/rados/test.sh
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:25 PM Revision 2df3f2c9 (ceph): cfuse: make use of fuse invalidate callback optional (and off by default)
- This is hitting a deadlock in fuse itself (or so it appears). Disable it
until that's sorted out.
Signed-off-by: Sa... - 07:16 PM Revision 1179a31b (ceph): formatter: remove bad asserts
- 06:42 PM Revision afdb1f38 (ceph): rgw: list objects shows time in GMT, thread safety fix
- 06:21 PM Revision 14de6d26 (ceph): client: compile with older libfuse2
- 06:08 PM Revision a3117152 (ceph): src/perfglue/heap_profiler.cc: fix snprintf
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:00 PM Revision 82baaeb1 (ceph): Merge branch 'wip-librados-tests'
- 05:47 PM Revision cb1f0500 (ceph): osd_types: fix merge
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:47 PM Revision 46645936 (ceph): Merge commit 'v0.33'
- 05:35 PM Revision e20bae2a (ceph): valgrind: Document!
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 05:32 PM Revision 73de620c (ceph): Merge branch 'wip-valgrind'
- 05:30 PM Revision 4efc95fa (ceph): include log in valgrind log file names
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 05:30 PM Revision d5eb2c2b (ceph): ceph task: split up arguments a little more
- This allows selective daemon kill signal changes. With valgrind
daemons we want term instead of kill, for instance.
... - 05:29 PM Revision 5323e179 (ceph): valgrind: move valgrind logs to log dir
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 05:00 PM Bug #1404: cephfs: test pool option
- fixed by commit:92746916dc19ef155b943c326ab0f8062811edac, but still need to add tests for all of the cephfs options!
- 04:41 PM Bug #1404 (Resolved): cephfs: test pool option
- It's been broken for a long time since the pools needed to be added to the available list in the MDSMap. Add some tes...
- 04:59 PM Bug #1397 (Resolved): rbd_list and rados_pool_list differ
- yup! fixed by commit:a724bc4667a0cc1a058343459845a38ca631ff1c
- 04:48 PM Bug #1392: osd: PGLog received in Initial state
- Actually, it looks like last_warm_restart isn't set in the Initial state. Thus, between when a PG is created and whe...
- 04:33 PM Bug #1392: osd: PGLog received in Initial state
- The crash happened on just that osd (out of 30), and it hasn't happened since, so I don't have a way to easily reprod...
- 03:56 PM Bug #1392: osd: PGLog received in Initial state
- Sorry for the delay. It looks like the osd received a log message during GetInfo. Would it be possible to reproduce...
- 04:16 PM Revision 89eeb34e (ceph): osd: pg_stat_t: throw exception on unknown encoding
- This is better then decoding nonsense, eating ram, etc.
Signed-off-by: Sage Weil <sage@newdream.net> - 04:16 PM Revision 340b4348 (ceph): v0.33
- 04:07 PM Revision ce005640 (ceph): qa: test rename into snapped dir
- Tests fix for #1399.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 04:07 PM Revision e98669ea (ceph): mds: mark newly created snaprealms as open
- Fixes: #1399
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 03:58 PM CephFS Bug #1389 (Resolved): re-created snapshot gets removed by mds journal replay
- fixed by commit:d60d5319ad5d6674488cab96b4a452ff553e779b
- 03:26 PM CephFS Bug #1399: mds crash
- replay crash looks like the one fixed in commit:8c5e7dcf8cf7f3daa65eb9905, yay!
- 02:17 PM CephFS Bug #1399 (Resolved): mds crash
- I'm not sure I can reproduce the second (replay) crash. Sam, next time you see one of these, please capture a replay...
- 09:12 AM CephFS Bug #1399 (In Progress): mds crash
- original crash is fixed by commit:e98669ea69059e26e0c4aa72c46e0be5bfc96386
- 08:07 AM CephFS Bug #1399: mds crash
- Hmmm. If ...
- 07:53 AM CephFS Bug #1399: mds crash
- As for the original error, it does seem reproducible by creating a snapshot of a directory using the mkdir system cal...
- 07:28 AM CephFS Bug #1399: mds crash
- I removed the assertion: assert(in->is_head());
That allowed the mds servers to restart and complete recovery, and... - 02:44 PM Bug #1403: osd: FAILED assert(0 == "we got a bad state machine event")
- Oops. Here's the rest of the report:
../../src/osd/PG.cc: 3891: FAILED assert(0 == "we got a bad state machine ev... - 02:40 PM Bug #1403 (Resolved): osd: FAILED assert(0 == "we got a bad state machine event")
- 02:16 PM CephFS Bug #1393 (Resolved): cfuse failed 3 pjd tests
- 01:00 PM Bug #1395 (Resolved): rbd man page lists the rm command twice
- 10:45 AM Feature #1207 (Resolved): qa: run daemon(s) through valgrind
- Merged in commit:e20bae2a7f15cd4fe899f6bc711bec1ef1f2d3e8
- 09:53 AM Feature #1401: Support mutually untrusting clients using the same Ceph cluster
- Tommi Virtanen wrote:
> Depends on #1402.
Make that #1237. - 09:52 AM Feature #1401: Support mutually untrusting clients using the same Ceph cluster
- Depends on #1402.
- 09:40 AM Feature #1401: Support mutually untrusting clients using the same Ceph cluster
- Copy-pasting IRC conversation for the record:...
- 09:24 AM Feature #1401: Support mutually untrusting clients using the same Ceph cluster
- The last piece is to restrict a client to only be allowed to mount a subtree on the mds. This means extending the md...
- 09:22 AM Feature #1401 (Closed): Support mutually untrusting clients using the same Ceph cluster
- QA & document this feature -- all the necessary code might already be there.
From the mailing list:... - 09:53 AM Feature #1402 (Closed): MDS: use caps to restrict clients to a subtree of the filesystem
- Oops this already existed as #1237.
- 09:51 AM Feature #1402 (Closed): MDS: use caps to restrict clients to a subtree of the filesystem
- This allows mutually untrusting users to share a ceph filesystem.
Without this, root@customerA can rm -rf /custome... - 12:55 AM Revision 2dd5510c (ceph): test/rados-api/misc.cc: test Rados::version
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:53 AM Revision d45b0f63 (ceph): test/rados-api/list.cc: C++ version
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:48 AM Revision b8b14c9c (ceph): test/rados-api/snapshots.cc: C++ version
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:37 AM Revision 65626ede (ceph): test/rados-api/watch_notify.cc: implement C++ ver
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:30 AM Revision 9ca1527b (ceph): test/rados-api/stat.cc: implement C++ tests
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:16 AM Revision d8da2ce6 (ceph): rados/librados.h: use rados_snap_t consistently
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/16/2011
- 09:58 PM CephFS Bug #1399: mds crash
- Sam, do you still have this cluster? Can you restart the mds with debug mds = 20 and attach the resulting log? There...
- 03:11 PM CephFS Bug #1399 (Resolved): mds crash
- After running successfully with one active mds and two standbys, the active mds has crashed, and on restart, it crash...
- 08:06 PM Revision 80a094d2 (ceph): test/rados-api/aio.cc: test RoundTripWriteFull
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 07:59 PM Revision 9df46059 (ceph): test/rados-api/aio.cc: implement FlushPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:25 PM Revision 4d9dab04 (ceph): test/rados-api/aio.cc: test ReturnValuePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:10 PM Revision 8cc794b0 (ceph): test/rados-api/aio.cc: test IsSafePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:49 PM Revision 3b81656a (ceph): test/rados-api/aio.cc: add IsCompletePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:40 PM Revision f4d73829 (ceph): test/rados-api/aio.cc: Add RoundTripAppendPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:31 PM Revision cc252669 (ceph): test/rados-api/aio.cc: add RoundTripPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:44 PM Feature #1400 (Resolved): throw exceptions on unknown encoding
- Any struct that encodes with a...
- 03:58 AM Revision 8072bdca (ceph): Fix bug in turning off logging. Remove log_dir.
- Fix a bug that made it impossible to turn off logging to a file.
Remove the old "log_dir" setting. It has been depre...
08/15/2011
- 11:34 PM Revision e7de7cba (ceph): Merge branch 'next'
- 11:33 PM Revision 77c780df (ceph): osd: fix heartbeats after bad markdown
- The heartbeat start message comes from hbin messenger, which has no port
and a nonce of the pid (at startup). When w... - 11:33 PM Revision 8df56b50 (ceph): mds: break #include recursion
- CInode.h -> SessionMap.h -> CInode.h
Coverity cid 14
Signed-off-by: Sage Weil <sage@newdream.net> - 11:33 PM Revision 915aa412 (ceph): crushtool: don't inc bool
- Coverity cid 21
Signed-off-by: Sage Weil <sage@newdream.net> - 11:33 PM Revision 44d3193a (ceph): monmaptool: don't inc bool
- Coverity cid 22
Signed-off-by: Sage Weil <sage@newdream.net> - 10:35 PM Revision aa744817 (ceph): ceph: split up daemon-running arguments and insert valgrind ones
- This setup should let us insert other kinds of things too, if we
need them.
Signed-off-by: Greg Farnum <gregory.farn... - 10:32 PM Revision 9ec19f13 (ceph): ceph: Set up valgrind as a flavor, and create a dir for logging.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:31 PM Revision 98ac89a5 (ceph): ceph task: pass the full config to the daemon startup subs
- So far as I can tell there is no reason to reduce them to
the coverage config, and I want the full config for my
soon... - 09:15 PM Revision 19ec8d0f (ceph): rgw: adjust returned error when got invalid auth request
- 09:15 PM Revision b10b3a39 (ceph): rgw: create CommonPrefixes section for each prefix
- 09:15 PM Revision 3b9fbcfe (ceph): rgw: handle zero and invalid max_keys
- 09:15 PM Revision 3e66b807 (ceph): rgw: some encoding fixes
- 05:08 PM Feature #1207 (In Progress): qa: run daemon(s) through valgrind
- Have it working for the monitors; something's broken with the other daemon types. Also, it's grabbing archives before...
- 04:36 PM Revision 747deeca (ceph): Add assert to catch simple typos in roles list.
- Input of "roles:\n- [mds,1]" used to make teuthology crash
in a non-obviou way. - 01:31 PM CephFS Cleanup #1307 (Closed): client cleanup
- 01:31 PM Bug #1194: kclient: NFS reexport does not survive ceph fs remount
- pushed this to for-linus branch.
- 09:58 AM CephFS Feature #1398 (New): qa: multiclient file io test
- test read/write consistency across clients.
i thinking:
- teuthology task gets list of client names (or uses all... - 09:49 AM CephFS Bug #1391 (Can't reproduce): client: crash on std::string in insert_trace()
- It's not clear from code inspection where this might be coming from, unless there is general heap corruption. If you...
- 06:29 AM Bug #1397 (Resolved): rbd_list and rados_pool_list differ
- Using the C API the @rbd_list@ and @rados_pool_list@ functions differ in how you call them, and what they return, whi...
08/13/2011
- 12:44 AM Revision 80dfc981 (ceph): obsync: add SRC_SECURE, DST_SECURE
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/12/2011
- 09:55 PM Revision e2d4462a (ceph): LibRadosAio: WaitForSafePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:46 PM Revision ffab58f7 (ceph): test/rados-api/aio.cc: SimpleWritePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:13 PM Revision 1d6224ef (ceph): rados_setxattr: return 0 on success
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:07 PM Revision 431e98a9 (ceph): rados_write_full: remove offset parameter
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:56 PM Feature #1396 (Resolved): Environment variables to specify secure connections for source and dest...
- implemented by 80dfc981a0f70127b475d30b36f97e39eae49994
- 11:58 AM Feature #1396 (Resolved): Environment variables to specify secure connections for source and dest...
- It would be a handy feature for obsync if there could be SRC_SECURE and DST_SECURE environment variables that would s...
- 05:48 PM Revision 361997af (ceph): librbd: output copy progress
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 05:33 PM Revision 2f7e3715 (ceph): Merge branch 'next'
- 05:33 PM Revision 8c5e7dcf (ceph): mds: avoid issue_caps on snapped inodes
- Only head inodes have caps. Don't set need_issue if it's not a head inode.
This is cleaner than bailing out of issu... - 12:21 PM Bug #1394 (Resolved): rbd tool hangs when doing rename
- That's odd, all the transactions are complete. Now that I try on 0.32, I can reproduce the problem. This is fixed in ...
- 11:27 AM Bug #1394: rbd tool hangs when doing rename
- My rbd and rados libs are the same commit, 0.32 packages.
Here's the command with the debugging added:... - 10:42 AM Bug #1394: rbd tool hangs when doing rename
- The rbd rename command works on my machine. Is it only hanging with this particular image? If so, running it with the...
- 10:36 AM Bug #1394: rbd tool hangs when doing rename
- the fix should include a script in qa/workunits/rbd that tests this.
- 07:55 AM Bug #1394 (Resolved): rbd tool hangs when doing rename
- Running the following results in a successful rename, however it just hangs....
- 10:34 AM CephFS Bug #1390 (Resolved): MDS crash in function 'bool Locker::issue_caps(CInode*, Capability*)', in t...
- pushed to next/master, will be in v0.33.
- 04:01 AM CephFS Bug #1390: MDS crash in function 'bool Locker::issue_caps(CInode*, Capability*)', in thread '0x7f...
- It's looking good, cluster has started up okay, no metadata crashes :-)
- 08:42 AM Bug #1395 (Resolved): rbd man page lists the rm command twice
- As the title says:...
- 12:54 AM Revision dc305839 (ceph): test/rados-api/snapshots.cc:RollbackPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:50 AM Revision 8f2933f8 (ceph): test/rados-api/snapshots.cc:SnapGetNamePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:45 AM Revision 2a766bce (ceph): test/rados-api/snapshots.cc:SelfManagedRollbackTest
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 12:30 AM Revision 19ed6b65 (ceph): SelfManagedSnapTest: test more
- * test more snapshot functions in SelfManagedSnapTest
* rados_ioctx_selfmanaged_snap_set_write_ctx: fix name in C fi... - 12:09 AM Revision 54810b6d (ceph): test/rados-api/snapshots.cc: SelfManagedSnapTestPP
- Also test SelfManagedSnapRollbackPP
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/11/2011
- 09:43 PM Revision db368161 (ceph): test/rados-api/io.cc: test XattrListPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:06 PM Revision 020f063c (ceph): test/rados-api/io.cc: test RmXattrPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:56 PM Revision 4781fa86 (ceph): test/rados-api/io.cc: test XattrsRoundTripPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 08:15 PM Revision de681da0 (ceph): client: use localized ldout macro
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:02 PM Revision fc5923a9 (ceph): mon: dump mdsmap in json
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 08:01 PM Revision 529c4d53 (ceph): test/rados-api/io.cc: add RemoveTestPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:59 PM Revision ceb8cd1e (ceph): test/rados-api/io.cc: AppendRoundTripPP,TruncPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:19 PM Revision a1895743 (ceph): test/rados-api/io.cc: test WriteFullRoundTripPP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:15 PM Revision fef049e1 (ceph): test/rados-api/io.cc: test RoundTripPP, etc.
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:40 PM Revision 05e7157c (ceph): test/rados-api/io.cc: test SimpleWritePP
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:40 PM Revision a1d8013a (ceph): test/rados-api/test.cc: add create_one_pool_pp,etc
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:40 PM Revision 2c1ea4a6 (ceph): librados: add IoCtx::close
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:06 PM Revision 009968ef (ceph): mds: don't wait for lock 'flushing' flag on replicas
- If we are a replica, the 'flushing' means that we had dirty scatterlock
data and are waiting for it to get flushed ou... - 05:06 PM Revision bbd4a26a (ceph): librbd: deduplicate sparse read interpretation
- AioBlockCompletions and read_iterate each had their own copy of this
code, leading to bugs when only one was changed.... - 05:06 PM Revision 4bf6f6a6 (ceph): librbd: make read_iterate mirror AioCompletion::complete
- This makes testlibrbd and testlibrbdpp pass again, and
now both I/O paths have the same logic.
Signed-off-by: Josh D... - 05:06 PM Revision 30c97d50 (ceph): testlibrbd: remove useless print statements
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 05:06 PM Revision a631c464 (ceph): testlibrbd: NUL-terminate orig_data
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 05:06 PM Revision ac0c4d8d (ceph): testlibrbd(pp): accept standard command line arguments
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 05:04 PM Revision 9d4f7487 (ceph): osd: don't request heartbeats from down peers
- We try to add sources for anything in peer_info, which can include down
OSDs. Skip them.
Signed-off-by: Sage Weil <... - 04:58 PM Revision d7455728 (ceph): cfuse: properly invalidate page cache data via callback
- The Client has a cache, but it is actually usually duplicating
what the kernel has in the page cache. Use the Client... - 04:53 PM Revision 5d40ed4c (ceph): client: add callback for cache invalidate
- This will get called any time an inode's cached file data is invalidated.
cfuse will use it.
Signed-off-by: Sage Wei... - 04:49 PM Revision 2a0a5b54 (ceph): client: use helper to invalidate/truncate cached file data
- We invalidate cached file data in several places; use a common helper.
Signed-off-by: Sage Weil <sage@newdream.net> - 04:49 PM rgw Bug #1332 (Resolved): Object creation with invalid Expect non-empty results in 417
- Fixed, commit:7045dd7f8ed1ca51843e2512f389b9a6ea963812. Also pushed fix to deployment tree.
- 04:49 PM rgw Bug #1331 (Resolved): Object creation with invalid Content-Length above actual results in BotoCli...
- Fixed, commit:09f41761dfd3a3329e851418e7dc5c475c34e735. Also pushed fix to deployment tree.
- 04:37 PM Revision 9e8f3482 (ceph): client: fix cached file data truncation
- We want to truncate our cached data when the file size changes, not
when truncate_size does. This aligns us with the... - 04:36 PM CephFS Bug #1393 (Resolved): cfuse failed 3 pjd tests
- Teuthology results are in teuthology:~teuthworker/archive/full_suite_coverage_20110811/35/teuthology.log
Nodes sepia... - 03:51 PM CephFS Bug #1390: MDS crash in function 'bool Locker::issue_caps(CInode*, Capability*)', in thread '0x7f...
- Will it be safe to cherry-pick this onto 0.32? Else I can try packaging up that branch and deploying it!
- 03:23 PM CephFS Bug #1390: MDS crash in function 'bool Locker::issue_caps(CInode*, Capability*)', in thread '0x7f...
- Pushed commit:26871eff1740d6ec5b9b287bf47e098db913fb27 (branch wip-needissue) that should fix this. Can you let me k...
- 03:36 AM CephFS Bug #1390 (Resolved): MDS crash in function 'bool Locker::issue_caps(CInode*, Capability*)', in t...
- Yesterday I was playing about with snapshots, which seemed to expose a bug in btrfs (delayed_inode.c) which I have si...
- 01:44 PM Bug #1392 (Resolved): osd: PGLog received in Initial state
- OSD crash on single OSD. Attached is the log and core files.
- 12:54 PM CephFS Bug #1391 (Resolved): client: crash on std::string in insert_trace()
- Random cfuse client crash. Sorry I don't have a core file for this. It only happened on
*** Caught signal (Segm... - 08:24 AM CephFS Bug #1389: re-created snapshot gets removed by mds journal replay
- I'm pretty sure I got it with both, before I understood why my snapshots were disappearing. Once I did, I only teste...
- 12:10 AM Revision 49a89c97 (ceph): fuse: allow big writes
- Easiest 500% performance improvement ever!
Signed-off-by: Sage Weil <sage@newdream.net> - 12:10 AM Revision e643e4e7 (ceph): fuse: enable atomic_o_trunc
- The MDS does this atomically.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:06 AM Revision 61a712dc (ceph): Add test/rados-api/misc.cc
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/10/2011
- 11:44 PM Revision 97a77c1c (ceph): Merge branch 'wip-pg-creation'
- 11:16 PM Revision 0139323e (ceph): Merge branch 'wip-nuke'
- Conflicts:
teuthology/task/kernel.py - 11:12 PM Revision 6938946a (ceph): manypools: remove commented-out code
- This accidentally got left in from my development.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 11:07 PM Revision d1728003 (ceph): librados: implement/document tmap_{get,put}
- These aren't strictly necessary now (you can just read the raw object or
do a writefull and get the same thing) but t... - 11:06 PM Revision 1ea825c6 (ceph): teuthology-nuke: split the big main function
- It was getting a bit big, but now all the functions fit on
one screen each.
Signed-off-by: Greg Farnum <gregory.farn... - 10:55 PM Revision 78f0f2d9 (ceph): test/rados-api/snapshots: add SelfManagedSnapTest
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:54 PM Revision 0e512314 (ceph): librados:fix name of rados_ioctx_snap_set_read/wr
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:40 PM Revision 287b0a12 (ceph): radosgw_admin: fix clitest
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:38 PM Revision af0d7c5e (ceph): teuthology-nuke: move it into its own file.
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:38 PM Revision fb774ca3 (ceph): mds: don't wait for lock 'flushing' flag on replicas
- If we are a replica, the 'flushing' means that we had dirty scatterlock
data and are waiting for it to get flushed ou... - 10:28 PM Revision 4a960e4d (ceph): rgw: can specify time for intent log purging
- 10:27 PM Revision 8caef771 (ceph): test/rados-api/snapshots.cc: test SnapGetName
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:05 PM Revision 3f9c0d01 (ceph): test/rados-api/snapshots.cc: test Rollback
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:40 PM Revision 762e3e11 (ceph): rgw: reduce intent log obj name time resolution
- 09:37 PM Revision 453a0f99 (ceph): teuthology-nuke: identify and reboot machines with kernel mounts
- This includes untested code for just force-unmounting them
when that works again, but for now it does a full reboot-a... - 09:37 PM Revision 95660084 (ceph): teuthology-nuke: use a more robust cfuse mount finder
- This way it can remove cfuse mounts in any location on
the system.
Signed-off-by: Greg Farnum <gregory.farnum@dreamh... - 09:37 PM Revision 257d6313 (ceph): teuthology-nuke: split out different pieces into different loops
- This will let us behave more intelligently on things like
nuking kernel mounts.
Signed-off-by: Greg Farnum <gregory.... - 09:37 PM Revision b5859f87 (ceph): Move reconnect function from kernel task to misc.py
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 09:31 PM Revision 3dbf9d4e (ceph): test/rados-api/snapshots.cc: test SnapRemove
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:30 PM Revision 75e67318 (ceph): rgw: add 'temp remove' tool for rgw_admin
- 09:01 PM Revision 47ab2bab (ceph): qa/fsstress: be verbose
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:54 PM Revision ba3ac0fa (ceph): librados: document rados_pool_list
- Not sure exactly where this API markup comes from, but it seems as good
as any.
Signed-off-by: Sage Weil <sage@newdr... - 08:54 PM Revision c635c464 (ceph): librados: document tmap_update
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:46 PM Revision 48a2aa19 (ceph): Merge branch 'wip-librbd-read'
- 08:43 PM Revision e9d92013 (ceph): rgw: update radosgw man page
- 08:40 PM Revision 7fd798a3 (ceph): Configure grub to default to the right kernel, not the greatest install...
- This is sticky; that is, even if you install other kernels (manually/via fab/etc),
grub will keep booting up the one ... - 08:36 PM Revision e98b218d (ceph): restructure in terms of collections
- 08:22 PM Revision 39e22e4c (ceph): Handle socket.timeout when waiting for a reconnect.
- Now it gets ignored, just like the other harmless socket errors.
- 08:21 PM Revision 742109f4 (ceph): Wait up to 300 seconds for a reboot.
- At least sepia86 was reliably slower than the previous 180 second default.
- 07:55 PM Revision 23874712 (ceph): add workunits on top of rbd
- 07:47 PM Revision bf7b1dd4 (ceph): ceph: fix max_mds calculation
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:14 PM Revision 68bd54d1 (ceph): librados: support op flags
- 07:14 PM Revision 32711fc8 (ceph): rgw: remove original object in put_obj, instead of truncate
- 07:05 PM Revision 07837c9b (ceph): osd: don't request heartbeats from down peers
- We try to add sources for anything in peer_info, which can include down
OSDs. Skip them.
Signed-off-by: Sage Weil <... - 07:01 PM Revision 4260a58e (ceph): test/rados-api/snapshots.cc: test SnapList
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:24 PM Revision 3841e06f (ceph): Add test/rados-api/watch_notify.cc
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:24 PM Revision 5b6b5691 (ceph): test/rados-api: move TestAlarm into test.h
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:24 PM Revision d61f7ae1 (ceph): test/rados-api/stat.cc: test ClusterStat,PoolStat
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:15 PM CephFS Bug #1389: re-created snapshot gets removed by mds journal replay
- Is this with the kernel or fuse client?
- 06:10 PM CephFS Bug #1389 (Resolved): re-created snapshot gets removed by mds journal replay
- Say you enter a snapshot pseudo-dir then run:
mkdir test
rmdir test
mkdir test
then restart the mds, and run:... - 05:50 PM Revision 3e4a9f23 (ceph): Makefile: add rule for building coverity build bundle
- As per http://scan.coverity.com/self-build/
Signed-off-by: Sage Weil <sage@newdream.net> - 05:47 PM Revision b2a36238 (ceph): objecter: fix compile error
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:46 PM Revision b1539a22 (ceph): objecter: allow setting of op flags for previously added op
- This lets you set the per-op flags for the last op added to the compound
operation.
Signed-off-by: Sage Weil <sage@n... - 05:46 PM Revision 2a5467d9 (ceph): osd: don't try to remove head if it doesn't exist
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:46 PM Revision 45e7ac78 (ceph): osd: add CEPH_OSD_OP_FLAG_FAILOK
- Setting this flag on an individual op in a compound operation will make
the OSD continue with the operation despite t... - 05:04 PM Revision 5107813c (ceph): Add test/rados-api/stat.cc: test rados_stat
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 04:53 PM Bug #1370 (Resolved): pgmon: PGMap::creating_pgs is not checked on leader election
- Well, I haven't seen any more errors from here, so I guess I misread something or fixed it elsewhere. Merged all the ...
- 04:46 PM Bug #1388 (Won't Fix): teuthology: ceph task doesn't clean up on failure
- If there's a problem in your config file, you get /tmp/cephtest directories sitting around.
- 04:30 PM Bug #1278 (Resolved): teuthology: make nuke nuke kernel mounts
- Merged into master in commit:0139323e51058a6b8d96d15af99a46ba1be2b6b8
- 02:50 PM Bug #1278: teuthology: make nuke nuke kernel mounts
- Finally got this to be successful, waiting to push until I've done a few other things with this.
- 04:30 PM Bug #1386 (Resolved): teuthology-nuke: make own file
- Oooh, it's so pretty! Merged in commit:0139323e51058a6b8d96d15af99a46ba1be2b6b8
- 03:26 PM Bug #1386 (In Progress): teuthology-nuke: make own file
- 09:23 AM Bug #1386: teuthology-nuke: make own file
- I'll do this while I'm messing about with nuke.
- 04:18 PM Feature #1387 (Closed): teuthology-nuke: don't fail on down nodes
- Right now teuthology-nuke fails if it can't ssh into a node. While it's good to get a notification that not all the n...
- 04:07 PM Revision a1f3cac0 (ceph): kernel: comment reconnect task, clean up reporting
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 03:46 PM Revision 7dcae356 (ceph): auth: use set<> instead of map<,bool>
- Value is always true. Be paranoid about the bool default value.
Signed-off-by: Sage Weil <sage@newdream.net> - 03:30 PM rgw Feature #1290 (Resolved): rgw atomic GETs
- Done, as of commit:d48bebeed74e584268b81d169e3c8f69d79d7005.
- 03:29 PM rgw Feature #1371 (Resolved): rgw: clean up temporary atomic get clones
- implemented, as of commit:4a960e4d56e29bdfb1da750e6bd435f2651091f9. Usage example:
radosgw_admin temp remove --dat... - 03:23 PM Linux kernel client Bug #1347: forced unmount kernel bug
- Jojy- do you have more detail about the workload that ran prior to the umount? We need to be able to reproduce this ...
- 02:46 PM Bug #1372 (Resolved): testlibrbd and testlibrbdpp fail
- Fixed in commit:f7ec3786c6faed5140ac3e14589c11d1e4486516.
- 01:40 PM Bug #1364 (Closed): teuthology: kernel task won't downgrade
- commit 7fd798a3474a7c453bba97c78fd3080687718386
Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Date: 2011-... - 01:08 PM Bug #1356 (Can't reproduce): OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Let's keep an eye out for future appearances of 0-byte pginfo files, as per our conversation on irc yesterday.
- 07:48 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- The thing is, none of the PGs were assigned to these OSDs....
- 07:34 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Great! You were lucky that it didn't affect all replicas for any of the PGs.
- 03:29 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- I couldn't resist the urge to try, so I removed all the empty pginfo files and their corresponding *_head directories...
- 01:06 PM Feature #1305 (Resolved): qa: add a bunch of rbd+ext34 tests to suite
- 12:15 PM rgw Bug #1352 (Resolved): rgw meta is persistent across object instances
- Fixed, commit:32711fc8f9c243e446c377b9f2fec809e7a12555.
- 11:27 AM CephFS Bug #1360: mds crash during pjd workunit on cfuse
- Machines are nuked and unlocked.
- 12:31 AM Revision 446f89e1 (ceph): librbd: deduplicate sparse read interpretation
- AioBlockCompletions and read_iterate each had their own copy of this
code, leading to bugs when only one was changed.... - 12:30 AM Revision f7ec3786 (ceph): librbd: make read_iterate mirror AioCompletion::complete
- This makes testlibrbd and testlibrbdpp pass again, and
now both I/O paths have the same logic.
Signed-off-by: Josh D... - 12:12 AM Revision d48bebee (ceph): librados, rgw: integrate stats categories
08/09/2011
- 11:53 PM Revision 663bbf8b (ceph): manypools: remove commented-out code
- This accidentally got left in from my development.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 11:25 PM Revision 1ccdcb98 (ceph): Make rbd task use mnt.N not mnt.client.N as mountpoint.
- Everything else expects this, so e.g. workunits wouldn't work with rbd.
- 11:20 PM Revision a3219c9a (ceph): test/rados-api/pool.cc:test PoolCreateWithCrushRule
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:11 PM Revision 780ebcdf (ceph): Make sure workunit task does not create mnt.N by itself.
- This used to hide a bug in the rbd task, where rbd
created the mountpoint with the wrong name. The workunits
ended up... - 11:07 PM Revision fd0d26d4 (ceph): test/rados-api/pool.cc: test rados_ioctx_get_id
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 271ed93b (ceph): Add test/rados-api/aio: SimpleWrite
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 03382bef (ceph): test/rados-api/aio: add WaitForSafe, RoundTrip
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision ed199461 (ceph): test/rados-api/aio.cc: add RoundTripAppend
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision b4590703 (ceph): test/rados-api/aio.cc: use rados_aio_release
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision e9b433e9 (ceph): test/rados-api/aio.cc: test IsComplete
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision a5077125 (ceph): test/rados-api/aio.cc: test IsSafe
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision c2f7adfe (ceph): test/rados-api/aio.cc: add ReturnValue
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 3090322d (ceph): test/rados-api/aio.cc: test Flush
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision d0d76918 (ceph): test/rados-api/aio.cc: fix shadowed class variable
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 571608d4 (ceph): test/rados-api/aio.cc: add RoundTripWriteFull
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision bb3114cd (ceph): rados/librados.h: remove rados_aio_get_obj_ver
- Remove the unused and unimplemented rados_aio_get_obj_ver from the
header file.
Signed-off-by: Colin McCabe <colin.m... - 11:02 PM Revision 0ad5500b (ceph): Add test_rados_api_list
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision b9348c20 (ceph): Add test/rados-api/pool.cc
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 9935e9fa (ceph): create_one_pool: improve error reporting
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 9e8a8b2d (ceph): .gitignore: add test_rados_api_*
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 54b9441f (ceph): test/rados-api/pool.cc: test PoolLookup,PoolDelete
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 5a06f0c5 (ceph): test/rados-api/pool.cc: add AuidTest1, AuidTest2
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision e442a2c3 (ceph): Makefile.am: build gtest in debug builds
- Formerly, we only built gtest when doing 'make check'. However, the
librados-api tests, which are built in debug mode... - 11:02 PM Revision 87f2960f (ceph): test/rados-api/io: add roundtrip, overlappingwrite
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 6349a132 (ceph): test/rados-api/io: add WriteFullRoundTrip
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 35093987 (ceph): test/rados-api/io: add AppendRoundTrip
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 5bee453b (ceph): test/rados-api/io: add TruncTest
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 039c5fca (ceph): test/rados-api/io: add RemoveTest
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 0a29e63a (ceph): test/rados-api/io: add XattrsRoundTrip
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:02 PM Revision 291c3c26 (ceph): test/rados-api/io.cc: add RmXattr, XattrIter
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:43 PM Revision 41f484d9 (ceph): objecter: allow requesting specific maps in maybe_request_map
- Use this capability so that wait_for_new_map can specify a specific
map it wants go get. I am extending the sins of t... - 10:42 PM Revision 4e2ec6fd (ceph): Add interactive-on-error, to pause and explore on error.
- Closes: http://tracker.newdream.net/issues/1291
- 09:28 PM Revision 5cb2bd27 (ceph): testlibrbd: remove useless print statements
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:28 PM Revision eee1d9a9 (ceph): allow s3tests.create_users defaults be overridden
- 09:28 PM Revision d7d995e8 (ceph): allow s3tests.create_users defaults be overridden
- 09:19 PM Revision 52b90b75 (ceph): testlibrbd: NUL-terminate orig_data
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:13 PM Revision 60b45882 (ceph): testlibrbd(pp): accept standard command line arguments
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 08:40 PM Revision 4241185a (ceph): Add simple unit test for get_clients.
- 08:23 PM Revision b5ba155c (ceph): Revert "fix get_clients"
- This reverts commit 83b6678e79904793bf31e82bbecad7bf16c1b2b5. The bug I was
hitting was actually fxied by 06e3e69c29... - 05:53 PM Revision 5e9bdf89 (ceph): librados: add missing method that was lost in merge
- 05:46 PM Revision 9f764db3 (ceph): Merge branch 'wip-stats' into wip-atomic-get
- Conflicts:
src/librados.cc - 05:41 PM Revision 8942d2bb (ceph): rgw: add hour to intent log name for higher resolution
- 03:58 PM Linux kernel client Feature #27: ACLs
- this just got simpler with the acl changes in 3.1. fs/btrfs/acl.c is a good model, although we can skip the ceph-spe...
- 03:51 PM Bug #1383 (Resolved): objecter: under some circumstances it requests the wrong map
- Yep. Have the simple fix in my local branch, will push with the rest of my changes!
- 03:22 PM Bug #1383 (Resolved): objecter: under some circumstances it requests the wrong map
- While testing manypools, I had a hang on pool creation. Turns out the monitor was creating the pool, but the client w...
- 03:43 PM Feature #1291 (Closed): teuthology: optionally pause on error
- commit 4e2ec6fd8be6816d677a008694ec75c970849257
Author: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Date: 2011-... - 03:37 PM Linux kernel client Bug #1380 (Resolved): msgpool osd_op_reply alloc failed
- In and of themselves these are harmless. There were mempool bugs fixed in commit:5185352c163a72cf969b2fbbfb89801b398...
- 07:54 AM Linux kernel client Bug #1380 (Resolved): msgpool osd_op_reply alloc failed
- ...
- 03:33 PM Bug #1386 (Resolved): teuthology-nuke: make own file
- Now is the time to move nuke() into teuthology/nuke.py main().
Also, teuthology-nuke shouldn't hijack parse_args()... - 03:32 PM Feature #1385 (Closed): teuthology-nuke should support nuking "all hosts i have locked"
- Right now, teuthology-nuke forces me to search through the logs from my last run, and construct a yaml file with the ...
- 03:29 PM Feature #1384 (Resolved): teuthology-lock --list should list only locks i own, by default
- That is, --owner user@host should be the default; that's the only think I'm interested in 99% of the time, and listin...
- 03:28 PM Linux kernel client Bug #1381 (Resolved): kernel BUG at mm/mempool.c:186!
- 07:56 AM Linux kernel client Bug #1381 (Resolved): kernel BUG at mm/mempool.c:186!
- ...
- 03:28 PM Linux kernel client Bug #1136 (Resolved): mempool_destroy failure on umount
- 12:52 PM Linux kernel client Bug #1140 (Resolved): balance_dirty_pages makes Fw cap revocation slow
- 12:43 PM Bug #1359: fsstress workunit hang on cfuse
- This looks just like #1369 with the ~15 minute interval resets.
- 12:40 PM CephFS Bug #1368: mds crash after blogbench on cfuse
- unlocked the nodes
- 12:38 PM CephFS Bug #1368: mds crash after blogbench on cfuse
- The crash was on shutdown. Have core file but gitbuilder binaries were expired.
Running in a loop to reproduce. - 12:28 PM Bug #1369: ffsb hang on cfuse (messenger?)
- The logs show regular ms_handle_resets (exactly 15 minute intervals). In config.cc is ms_tcp_read_timeout is 900 sec...
- 11:09 AM Bug #1382: kclient: crash on resending osd ops
- It looks like the issue stems from having a bunch of osds going out.
I'm not certain why these osds fail, but this b... - 10:34 AM Bug #1382 (Resolved): kclient: crash on resending osd ops
- Under performance testing with the SCST iSCSI driver on top of RBD (we'll switch to LIO at some point in the future, ...
- 03:57 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Sage Weil wrote:
> Wido den Hollander wrote:
> > Notice the fact that it's about the same osdmap, 8974!
>
> Does... - 03:52 AM Revision 2c20a6cd (ceph): Merge branch 'wip-client'
- Conflicts:
src/Makefile.am
08/08/2011
- 11:44 PM Revision 7cc208bb (ceph): rgw: log shadow objects removal in intent log
- 10:13 PM Revision 137f36d5 (ceph): teuthology: add task manypools
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 10:08 PM Revision 66050ad7 (ceph): Add test-rados directory, I/O test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:19 PM Revision 901a7a46 (ceph): Merge branch 'wip-atomic-get'
- 09:19 PM Revision 98ee76be (ceph): pgmon: use pool.get_last_change whenever creating new PGs
- We maintain last_change properly now, so we can use it at any time.
It may still be possible that we can get PGs wit... - 09:17 PM Revision 5b260cd2 (ceph): rgw: return ENOENT after get_obj() in some cases
- 09:11 PM Revision 84462a15 (ceph): osdmon: maintain ceph_pg_pool::last_change correctly
- We were making a few changes that didn't update this value, which
can cause problems when creating new PGs.
Signed-o... - 08:48 PM Revision fca79085 (ceph): pgmon: call check_osd_map via a new on_active implementation
- Previously it was possible to lose PG creations if a monitor election
happened at the right time. The issue would get... - 08:33 PM Revision 43a5cd70 (ceph): pgmon: fix comments to output the right function
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 08:21 PM Revision e27b4d17 (ceph): Merge branch 'debian-deps'
- 07:16 PM Revision 6d0192cb (ceph): debian: explicitly bind library users to matching version
- We are cheating with the shared libs by making small API changes without
bumping the soname. Bind users to a matchin... - 07:12 PM Revision a9da6db6 (ceph): Merge branch 'wip-heartbeats'
- Conflicts:
src/osd/OSD.cc - 05:25 PM Revision 258d7435 (ceph): rgw: ignore hidden objects for bucket deletion, fix namespace
- 05:06 PM Bug #1379 (Resolved): osd segfault during end of recovery
- From huang jun's email titled 'osd down after adding OSDs':...
- 05:00 PM Messengers Bug #1378 (Resolved): connection race - existing connection not open or connecting
- From huang jun's email titled 'osd down after adding OSDs':...
- 04:55 PM Revision 3cda58dd (ceph): Merge branch 'wip-config'
- 03:56 PM Feature #1377 (Resolved): teuthology: use btrfs to back osds
- each sepia node has a scratch /dev/sdb. (re)format it with btrfs on each run and put the osd data there.
- 03:50 PM Bug #1376 (Resolved): errant scrub stat mismatch logs after upgrade
- upgraded from git commit #394537092d to git commit #68cbbf42c42, and after restarting the cluster I immediately saw m...
- 03:41 PM rgw Feature #1375 (Rejected): Actually find pool limits
- DHO will create lots of buckets. With radosgw, bucket=pool=16 pgs (or 48 in-memory ones, due to replication across OS...
- 03:41 PM CephFS Bug #1366: mds segfault
- IIRC, I was testing out metadata. Create a bunch of directories (1000), and then create 1000 files in each directory...
- 03:38 PM CephFS Bug #1366: mds segfault
- Hmm, can you describe the workload?
Were you doing the write a bunch of data then add osds type of test? (That wo... - 03:28 PM CephFS Bug #1366: mds segfault
- Sorry, there probably was a core file, but its gone now. This was with ceph stable commit: dcaca3e358f7f42c7c826d0b6...
- 03:10 PM CephFS Bug #1366: mds segfault
- Do you have a core for this? Which commit were you running?
My guess is that its related to the full -> not full ... - 03:30 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Wido den Hollander wrote:
> Notice the fact that it's about the same osdmap, 8974!
Does that osdmap file size mat... - 03:29 PM Bug #1297 (Resolved): write a manypools teuthology task
- So each pool is actually 8 PGs each by default, so the task just writes 16 1-byte objects to each pool using rados be...
- 03:08 PM Bug #1358 (Can't reproduce): misc workunit hang on cfuse
- chalking this up to an ancient kernel, and/or perhaps the rename regression fixed last week. cleaned up the nodes an...
- 01:26 PM Bug #1358 (In Progress): misc workunit hang on cfuse
- 02:36 PM Bug #1370: pgmon: PGMap::creating_pgs is not checked on leader election
- 01:53 PM Bug #1370: pgmon: PGMap::creating_pgs is not checked on leader election
- Ah, not quite what I thought it was. creating_pgs is actually maintained via stat_pg_[add|sub], and it's done correct...
- 02:35 PM Bug #1365 (Resolved): pgmon: cannot guarantee new PGs have correct OSDMap creation epochs
- Done. Also it turns out this won't conflict with snapshot tracking at all; there's a separate snap_epoch for that.
- 01:26 PM Bug #1354 (Resolved): rgw debian package should depend on librados
- 01:02 PM Bug #1363 (Resolved): osd: heartbeat problems
- merged my changes. slang's issues were actually due to multiple ips/interfaces, though.
we may be able to impro... - 09:23 AM Bug #1363 (In Progress): osd: heartbeat problems
- 11:45 AM CephFS Bug #1114: NFS export extreme slowdown
- 11:33 AM Feature #1374 (Resolved): teuthology: email suite run results
- - failures
- summary of entire run
- specify target addr when you schedule a run. e.g. nightly run to list, indivi... - 11:22 AM Bug #1345 (Resolved): rgw: removing a bucket logs bad bucket name
- 09:55 AM Bug #1372 (Resolved): testlibrbd and testlibrbdpp fail
- On master at 45ca94054b191f57f48d0de64b212569e5a58d57, these tests both segfault.
Looks like an invalid use of memor... - 09:33 AM Bug #516 (Closed): filestore: handle large xattrs on ext3
- we already do what we can given the extN limits.
- 09:30 AM CephFS Bug #1064 (Can't reproduce): all mds's dies one by one after restart
- 09:23 AM rgw Feature #1371 (Resolved): rgw: clean up temporary atomic get clones
- rgw will make an intent log, and a background process will periodically clean up temporary objects.
- 04:27 AM Revision 857f2adc (ceph): mds: fix Migrator::audit() for IMPORT_ABORTING
- Make audit match reality.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 04:27 AM Revision 68cbbf42 (ceph): osd: missing ! in ifs on bad map after osd boot
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
08/06/2011
- 12:22 AM Revision 8d42c389 (ceph): rgw: clone_obj can set mtime
- 12:22 AM Revision 265d29a4 (ceph): librados: can set mtime for operate (mutate)
08/05/2011
- 11:33 PM Revision 45ca9405 (ceph): objecter: halt compound operation only if result < 0
- 10:49 PM Revision ad31db9e (ceph): Remove unused dyn_sprintf
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:44 PM Revision 6ad7f46b (ceph): config.cc: clean up includes
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:35 PM Revision 83b6678e (ceph): fix get_clients
- Only return the clients that are listed (not _all_ clients). There might
be a combination of cfuse and kclient (or o... - 09:35 PM Revision 3f2ad30a (ceph): cfuse, kclient: print remote host
- 09:35 PM Revision 01fac3e2 (ceph): new gitbuilder ref/branch naming
- no origin_ prefix
- 09:35 PM Revision ef2b8091 (ceph): use coverage_dir
- 09:35 PM Revision 06e3e69c (ceph): tasks/kclient: don't clobber remote
- 09:28 PM Revision 3d258798 (ceph): client: send renames to the correct node
- AFAICS we got this wrong in b794aeb39, which is annoying given I remember
thinking reasonably carefully about it.
Th... - 09:28 PM Revision 7a3e7502 (ceph): client: whitespace cleanup
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:28 PM Revision 822d28b5 (ceph): mds: chain rename subtree projections
- We can have two renames for the same file in flight to the journal. Stack
them up in a list. The old project_subtre... - 09:05 PM Revision 618b0745 (ceph): rgw: multiple fixes
- 07:49 PM Revision 9f608c72 (ceph): config: more thread-safety stuff
- * Don't allow parse_argv, parse_env, or parse_config_files to be used
after threads have been started.
* Don't allow... - 07:18 PM Revision d99333d0 (ceph): injectargs: forbid unsafe string modifications
- Don't allow string-valued configuration items to be changed using
injectargs unless they have observers. Otherwise, w... - 06:25 PM Revision c409e701 (ceph): test/daemon_config: add injectargs test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:20 PM Revision 2d4f5493 (ceph): Add daemon_config, libceph_config, etc.
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:17 PM Revision f38c3697 (ceph): kernel: install in parallel
- 06:08 PM Revision f66c010e (ceph): kernel: debug weird socket exceptions
- 06:07 PM Revision 6df0d71a (ceph): kernel: reboot immediately after installing
- This hides the latency of rebooting when installing on many machines.
- 06:07 PM Revision 7be4b6dc (ceph): pgmon: create ALL pgs in new pools with correct creation epoch
- 5bb07df6aa4684ebd2e70437081dea170464d8ee tried to do this, but it
only set them properly for localized PGs. Whoops!
A... - 06:04 PM Revision 6dcf379e (ceph): add config unit test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:59 PM Revision 3e6b17f1 (ceph): Down machines shouldn't be considered free.
- 05:47 PM Revision a33b8378 (ceph): librados.h: fix out-of-date comment
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 05:46 PM Revision fa2a4067 (ceph): debian: don't explicitly list deps; let debian figure it out
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:32 PM Revision 92e976a5 (ceph): config: more cleanup
- * any place where we know a set_val can't fail should be set_val_or_die
* in private md_config_t functions that need... - 05:12 PM Bug #1370 (In Progress): pgmon: PGMap::creating_pgs is not checked on leader election
- 05:09 PM Bug #1370 (Resolved): pgmon: PGMap::creating_pgs is not checked on leader election
- So, state is lost when the leader changes. This needs a little bit of thought to handle leaders updating their maps p...
- 05:07 PM Bug #1365 (In Progress): pgmon: cannot guarantee new PGs have correct OSDMap creation epochs
- Solving this by keeping last_changed data member updated the way it's supposed to be.
- 11:13 AM Bug #1365 (Resolved): pgmon: cannot guarantee new PGs have correct OSDMap creation epochs
- Following on from #1357:
Right now, pools in the OSDMap track their "last_changed" epoch. This is currently set when... - 05:01 PM Bug #1369 (Can't reproduce): ffsb hang on cfuse (messenger?)
- This is from teuthology run 218 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110805/218/).
The rel... - 02:33 PM CephFS Bug #1360 (Resolved): mds crash during pjd workunit on cfuse
- fixed by commit:822d28b55836e8013f1ccd542d17ed3d1a19727d
- 02:28 PM CephFS Bug #1368 (Can't reproduce): mds crash after blogbench on cfuse
- This is from teuthology run 216 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110805/216/).
The rel... - 02:26 PM CephFS Bug #1367 (Resolved): cfuse and mon crash after dbench
- This is from teuthology run 212 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110805/212/).
The rel... - 11:39 AM CephFS Bug #1366 (Can't reproduce): mds segfault
- I have 4 mds's running in the following setup:
[mds.alpha]
host = 192.168.101.12
[mds.bravo]
... - 11:13 AM Bug #1357 (Resolved): Making many pools and writing to them eventually hangs
- Ah, I didn't have quite the right fix in for the PGMonitor's PG creation (I was only setting it properly for the loca...
- 09:23 AM Bug #1122 (Resolved): kclient: async readahead
- I'm seeing significantly better performance on sepia. E.g. for 8 osds/1 client, kclient reads saturate the interface...
- 08:59 AM Bug #1364 (Closed): teuthology: kernel task won't downgrade
- update-grub sorts kernels in descending order, so it will always boot the first one. Unless you put the default=N in...
- 01:32 AM Revision 68e6f2b7 (ceph): Make scheduled tasks leave some machines free.
- 01:25 AM Revision 4e399da7 (ceph): Log connections to targets
- This way you can tell which machines have problems in case of an
error. - 12:16 AM Revision 66c3d8ff (ceph): libceph.h: fix C compilation
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/04/2011
- 11:49 PM Revision 2c9949e0 (ceph): config: add proper locking, fix comments
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 11:44 PM Revision f53cc374 (ceph): rgw: mark ops as atomic
- 11:28 PM Revision d305a5e2 (ceph): osd: remove unnecessary history creation.
- The history is coming from an MOSDPGNotify, but that is covered
by a require_same_or_newer_map check, so we're all go... - 11:23 PM Revision 8bb4c0a7 (ceph): osdmon: remove default values for prepare_new_pool
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 11:21 PM Revision 3b459efc (ceph): rgw: conditional clone, depends on src_cmpxattr result
- 11:12 PM Revision 454d7de4 (ceph): librados: modify src_cmpxattr interface
- 11:11 PM Revision 5bb07df6 (ceph): pgmon: create PGs with a creation epoch that matches the pool's
- Previously, the PGs were created with a creation date of the current
OSDMap. However, under some circumstances the PG... - 11:07 PM Revision 84ee468b (ceph): librados: get it compiled
- 10:58 PM Revision 10dda208 (ceph): rgw: fix decoding
- 10:58 PM Revision b03930fa (ceph): rgw: atomic get, various fixes
- 10:58 PM Revision bcf25bd5 (ceph): rgw: a few more operations are using atomic infrastructure
- 10:58 PM Revision a2bcbf26 (ceph): librados: restructure ObjectOperation
- 10:58 PM Revision 128e948d (ceph): rgw: adapt some more rados calls to the new atomic structure
- 10:58 PM Revision 476f5c64 (ceph): rgw: prepare for conditional clone_range
- 10:58 PM Revision 8d63e140 (ceph): rgw: introduce context to specific access calls
- 10:58 PM Revision 71a64818 (ceph): librados, objecter: add operations to support rgw atomic get
- 10:58 PM Revision 70cdd5da (ceph): rgw: atomic get, basic flow for severl rados ops
- not working yet
- 10:33 PM Revision 7ec6909b (ceph): Merge branch 'rgw-formatter'
- 09:48 PM Revision 76880e3e (ceph): In RGW, use the common/Formatter
- Implement a new subclass of Formatter, RGWFormatter_Plain, using code
from the old RGWFormatter and RGWFormatter_Plai... - 09:35 PM Revision 0bc7e28f (ceph): filestore: remove unused var
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:34 PM Revision 22e25247 (ceph): osd: Fix last_epoch_started initialization on new PGs
- This used to be safe by virtue of assigning same_acting_since
to osdmap->get_epoch(), but since we fixed bugs by hand... - 09:32 PM Revision 9baa0096 (ceph): osd: rename variables in project_pg_history.
- These are always the current sets; I don't know why they would
ever be called lastup and it confused me when I starte... - 09:27 PM Revision 0413e2ac (ceph): Formatter: add open_object_section_in_ns, etc.
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:11 PM Revision 041a22cc (ceph): test/formatter: test stream clearing
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:07 PM Revision 434be6c0 (ceph): Formatter: must use str("") to clear ostreams
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:05 PM Revision e2cbfb62 (ceph): More changes towards using common/Formatter in rgw
- * rename RgwFormatter::init to RgwForamtter::reset.
Roll old reset() into base_reset().
* add write_raw_data to Form... - 08:48 PM Revision 72b0851f (ceph): osd: expect heartbeats from anyone peering depends on
- We were getting heartbeats from just acting replicas. That's really not
enough if we want to be sure to detect failu... - 08:35 PM Revision cd01d14a (ceph): osd: restructure heartbeats
- Primary sends explicit message to replicas asking for heartbeats to start,
or to stop. Replicas never send heartbeat... - 07:44 PM Revision b4b61d62 (ceph): osd: Initialize new PGs with correct info.history.same_primary_since
- Previously we were initializing based on the local osdmap epoch, which
is often correct, but if we process the MOSDPG... - 07:44 PM Revision a8f3602d (ceph): osd: fix comment for project_pg_history
- Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
- 07:44 PM Revision 2f0d3b13 (ceph): osd: refactor PG creation slightly.
- We want to carefully set up the PG History. In most cases this will
do the same thing as previously, since it's a bra... - 07:44 PM Revision 92f02378 (ceph): messages: add print functions to the MOSDPG messages
- This should help with debugging any future issues.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> - 04:52 PM Bug #1357 (In Progress): Making many pools and writing to them eventually hangs
- Well, I've found several different races and bugs. My current progress is in the wip-pg-creation branch, but there's ...
- 09:26 AM Bug #1357: Making many pools and writing to them eventually hangs
- 09:13 AM Bug #1357 (Resolved): Making many pools and writing to them eventually hangs
- Discovered while testing #1297. Looping pool creates eventually hang when the OSD rejects writes for being sent earli...
- 03:39 PM Revision 309d1579 (ceph): osdmon: always send in the CRUSH rule.
- Not sure why this wasn't there before...looks like the CRUSH option
is newer than the code and it didn't all get upda... - 03:39 PM Revision 9bdc43fa (ceph): osdcaps: allow the default UID to have full caps on its own pools.
- The explicit skip in the check was just a wrongheaded attempt to
stay secure, I think.
Signed-off-by: Greg Farnum <g... - 03:39 PM Revision cf88ce5e (ceph): messages: change = to result = for less confusion!
- Every time I see "ondisk = 0" I think it means that the message
is an ack, not a commit. Now I won't, and neither wil... - 03:25 PM CephFS Bug #1360 (In Progress): mds crash during pjd workunit on cfuse
- 11:27 AM CephFS Bug #1360 (Resolved): mds crash during pjd workunit on cfuse
- This is from teuthology run 164 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110803-2/164/).
The r... - 03:24 PM Bug #1361 (Closed): filelayout workunit failed on kclient
- this passes with a recent kernel.
- 12:22 PM Bug #1361 (Closed): filelayout workunit failed on kclient
- This is from teuthology run 169 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110803-2/169/).
The r... - 02:04 PM Bug #1363 (Resolved): osd: heartbeat problems
- see slang's thread on ceph-devel.
- 01:48 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Just to sum up what we said on IRC:
osd.9 has 47 pginfo* files which are empty in the current dir and the two snap... - 01:22 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- can you look in the snap_* directories and see if that pglog file is 0 in those too?
- 01:15 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- I just started the OSD again with debug osd and filestore on 20, got a different backtrace:...
- 01:02 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Yes, that's like it's going. In the process where they are all starting again they start bouncing up and down, in thi...
- 12:58 PM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- This looks like it's crashing during cosd start-up... can you confirm what's going on here? Basically,
- osds ar... - 11:52 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- I'm doing something very wrong here... Attached is osdmap 8974 from the monitor (again).
- 11:50 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- I also attached the osdmap from osd.9.
- 11:47 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- The osdmap is attached, picked it from the mon out of osdmap_full.
I took a look at the coredump (/core.atom2.3816... - 11:16 AM Bug #1356: OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- do you have a core file? in frame 10, can you look at bl->_len ?
also, can you attach the osdmap_full_8974 from t... - 07:45 AM Bug #1356 (Can't reproduce): OSD crashes during recovery with OSDMap::decode(ceph::buffer::list&)
- Hi,
Like I said some time ago, I've been seeing these kind of crashes lately.
I just tried to start my cluster ... - 12:38 PM Feature #1362 (Resolved): collectd cephtool plugin: only spawn one cephtool process
- implemented in ssh://ceph.newdream.net/git/collectd-4.10.1.git by commit:c5a79bd6b869f43ff5c7920774b2336763ac16c9
- 12:37 PM Feature #1362 (Resolved): collectd cephtool plugin: only spawn one cephtool process
- Be more clever about spawning cephtool processes. This requires the new ...
- 11:24 AM Bug #1359 (Can't reproduce): fsstress workunit hang on cfuse
- This is from teuthology run 162 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110803-2/162/).
The r... - 11:22 AM Bug #1358 (Can't reproduce): misc workunit hang on cfuse
- This is from teuthology run 157 (logs in teuthology:~teuthworker/archive/full_suite_coverage_20110803-2/157/).
The r... - 08:51 AM Feature #1218 (Resolved): perfcounter collectd plugin
- 07:41 AM Linux kernel client Feature #1355 (New): use SKB destructor in messenger?
- http://lwn.net/SubscriberLink/452866/379edec80d09e4cb/
Currently we block message revocation if the page is being ... - 01:04 AM Revision ab537f1e (ceph): Revert "qa: Remove fsx workunit."
- This reverts commit a66da4af2571f27f55e8ee819f22e70355fb42b0. Still useful
to have this around. - 01:04 AM Revision 1c1d92ec (ceph): qa: move fsx.sh
- 01:04 AM Revision 7078c34c (ceph): mds: add hooks to test multimds link/unlink failures
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:56 AM Revision 6a920bb9 (ceph): os/FileStore: print out BTRFS_IOC_SNAP_CREATE errs
- Print out the errno for BTRFS_IOC_SNAP_CREATE errors if they happen.
Signed-off-by: Colin McCabe <colin.mccabe@dream...
08/03/2011
- 11:24 PM Revision efc1566c (ceph): osd: add CATEGORIES incompat feature
- Care should be taken to upgrade the monitors first, then osds.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com> - 10:28 PM Revision ac0a469b (ceph): teuthology-worker: log to a file with timestamps
- 10:01 PM Revision a71adf1b (ceph): librados: add create in specific category to c++ api
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:55 PM Revision fdd9332c (ceph): osd: allow 'category' to be set on object creation
- Allow a string category to be set on object creation. Object stats will
be included in the category total (in additi... - 09:52 PM Revision 5897d7b9 (ceph): teuthology-nuke: run in parallel, and print each node being nuked
- 09:35 PM Revision ee4460d0 (ceph): osd: segregate stats into categories
- Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
- 09:03 PM Revision 30a8dac3 (ceph): Set success at the beginning of a run.
- This way internal tasks like locking can tell whether the run
succeeded, and unlock nodes if it did. - 06:39 PM Revision 8981a232 (ceph): rgw_rest: don't leak formatters
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 06:21 PM Revision e8676ce0 (ceph): teuthology-nuke: reset rsyslog config
- 04:06 PM Bug #1297: write a manypools teuthology task
- Ah, nope, it hung again. Without debugging enabled, naturally...
- 03:59 PM Bug #1297: write a manypools teuthology task
- Running final tests now, but this appears to be working properly. Which is odd since it seemed broken before, but may...
- 03:51 PM Bug #1314 (In Progress): teuthology: tasks for file lock behavior on process exit
- I've got this mostly done I think but am having some issues with configuring it (odd issues in ceph task setup, proba...
- 03:51 PM Feature #1348 (Resolved): osd: categorize object stats
- 01:30 PM Bug #1354 (Resolved): rgw debian package should depend on librados
- 09:45 AM Bug #1316 (Rejected): HAVE_SYNCFS, acconfig.h, #define HAVE_SYNCFS 1
- broken libc header install
- 09:41 AM CephFS Bug #1350 (Closed): mds: dup lookups on scp workload
- this was fixed by commit:5dc09dd6b81c622960f628acdabda9eac8af1ceb on the server and commit:2f90b852e3ae73889d7f6de6ec...
- 12:59 AM Revision d079d510 (ceph): teuthology-worker: keep machines locked on error
- This prevents a failure to clean up in one case from affecting the
rest of the tests. - 12:05 AM Revision cec04ca9 (ceph): XmlFormatter: add DTD option
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
08/02/2011
- 11:15 PM Revision c7b62fea (ceph): teuthology-lock: update usage
- 11:11 PM Revision ede86e7d (ceph): Formatter.cc: add get_len method
- Add get_len method because rgw needs it.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 11:04 PM Revision 24a7b5f7 (ceph): RGWFormatter: get rid of one flush variant
- This flush variant won't exist in common/Formatter.cc.
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 10:56 PM Revision cd044235 (ceph): teuthology-lock: allow list of locks to be filtered by owner and status
- 10:42 PM Revision dbbce937 (ceph): rgw_formats: introduce dump_unsigned, dump_int,etc
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:29 PM Revision 7b239448 (ceph): rgw: rename dump_value_str to dump_format
- rename dump_value_str to dump_format for compatibility with Formatter.cc
Signed-off-by: Colin McCabe <colin.mccabe@d... - 10:26 PM Revision 9dbeeaaf (ceph): rados tool: fix cmdline parsing for --block-size
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 10:26 PM Revision 77fed7b2 (ceph): Formatter: add XMLFormatter::reset
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:54 PM CephFS Bug #1350: mds: dup lookups on scp workload
- It happens only for the directories that are subdirs of root. Might be something to do with the code in dir.c(ceph_mk...
- 05:46 PM CephFS Bug #1350: mds: dup lookups on scp workload
- Running kernel 2.6.39
server version:
ceph version 0.29.1-479-g7dd919c (commit:7dd919c2b4c90c51758062c32accf22fff... - 11:22 AM CephFS Bug #1350: mds: dup lookups on scp workload
- what kernel version are you running?
your server version looks quite old. can you try the latest? - 10:49 AM CephFS Bug #1350: mds: dup lookups on scp workload
- Adding dmesg and mds logs for scp use case.
- 09:42 PM Revision 8b5cda57 (ceph): Formatter.cc: whitespace fix
- whitespace fix to conform to the coding standard
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 09:41 PM Revision 8c4f729a (ceph): rgw:rename open_obj_section to open_object_section
- For compatibility with common/Formatter.cc
Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com> - 09:30 PM Revision 8003b772 (ceph): Formatter: remove protected data, methods
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:30 PM Revision 70f585c8 (ceph): Add XMLFormatter and unit test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 09:20 PM Revision 7401feaf (ceph): objecter, osd: fix up cmpxattr
- Objecter stubs, useful debug output.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:20 PM Revision 4fbdb996 (ceph): osd: fix cmpxattr string construction
- The bufferlist may not contain a terminating '\0'.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:20 PM Revision 23a46f28 (ceph): testradospp: test cmpxattr (eq, str)
- No coverage for the other tests yet.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:20 PM Revision acc8600c (ceph): osd: change src_oid encoding -- FLAG DAY
- The old encoding was mutually exclusive with putting any data payload on
the operation. That was stupid.. we can't, ... - 09:20 PM Revision 97b68221 (ceph): osd, objecter, librados: add SRC_CMPXATTR operation
- and test.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:17 PM Revision 945cdd4b (ceph): pg: remove do_complete_notify
- This method has no dependence on the pg.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com> - 09:17 PM Revision 50d7684f (ceph): osd: document remove_watchers race avoidance
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision 73c8e81c (ceph): librados: fix notify deadlock
- The success of the notify call needs to checked before waiting to
receive a notification. If we try notifying on an o... - 09:17 PM Revision e6ff13d8 (ceph): systest: move pool deletion into its own file
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision b8d7c223 (ceph): systest: make pool configurable
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision aa00e0dd (ceph): systest: add st_rados_watch building block for watch/notify tests
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision cbaddc59 (ceph): systest: add st_rados_notify building block for watch/notify tests
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision ee447618 (ceph): systest: add object deletion to mirror object creation
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision 602365e8 (ceph): systest: st_rados_create_pool: optionally wait before creating the pool
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision 83ad583c (ceph): systest: add watch/notify tests with pool and object deletion
- Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
- 09:17 PM Revision 3a6d8ebb (ceph): osd, pg: clean up watchers on pg deletion and shutdown
- Watchers and their object contexts need to be cleaned up so
they aren't used after the pg is gone. This happened if t... - 09:17 PM Revision 8a8d9bf5 (ceph): osd: put_object_context: tolerate pgs being deleted
- PGs that are queued for deletion won't be in the osdmap,
and may not be in the pg_map, but if they are, it's safe to
... - 06:40 PM Revision 43daec44 (ceph): thread: detach()
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:40 PM Revision 37dc931d (ceph): mds: detach replay thread
- Since we don't join it.
This fixes a leak of per-thread state.. namely, an ~8MB chunk of virtual
memory (and a handf... - 06:08 PM Revision 6f39bb93 (ceph): common/Formatter: add unit test
- Signed-off-by: Colin McCabe <colin.mccabe@dreamhost.com>
- 02:31 PM Feature #1353 (Resolved): qa: full librados c++ api gtest coverage
- the testrados[pp] apps cover a very very very small portion of the api, are interactive (testradospp is at least), an...
- 01:59 PM Feature #1274 (Resolved): qa: test watch/notify races
- 01:39 PM rgw Bug #1352 (Resolved): rgw meta is persistent across object instances
- i.e., PUT obj with meta, PUT obj w/o meta, HEAD obj will return the old meta.
- 11:51 AM CephFS Bug #1349 (Resolved): mds: standby-replay leaks memory
- 11:51 AM CephFS Bug #1349: mds: standby-replay leaks memory
- fixed by commit:37dc931d2b715070e9ea806620cea9bdc22e85b3
- 10:51 AM CephFS Bug #1349: mds: standby-replay leaks memory
- I don't know much of anything about memory address management, but my assumption is that this is just a result of som...
- 10:48 AM CephFS Bug #1349: mds: standby-replay leaks memory
- Yeah, it's VIRT not RSS:...
- 11:38 AM Bug #1351 (Resolved): rados bench should report errors
- Right now if you get eg a perm error on write, rados bench just keeps looping waiting for the write to complete. That...
- 12:59 AM Revision 09a35767 (ceph): mdsmon: send commands to all MDSes, not just the in&up ones.
- Now we can send messages to standbys via broadcast, even if we
can't yet single them out.
Signed-off-by: Greg Farnum... - 12:23 AM Revision 3c77a456 (ceph): cephtool: accept semicolons in commandline args
- Semicolons can now be used to give multiple arguments to cephtool in one
invocation. So a command like this is now po...
Also available in: Atom