Activity
From 11/20/2010 to 12/19/2010
12/19/2010
- 02:03 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Sage Weil wrote:
> Stefan Majer wrote:
> > Hi,
> >
> > we see similar problems with kernel 2.6.37-rc6 and ceph b...
12/18/2010
- 02:49 PM Bug #662 (Resolved): update rbd qa scripts to use new sysfs interface
- ...and run them on one or more of the ceph* nodes!
see Documentation/ABI/testing/sysfs-bus-rbd in latest -rc to se... - 10:09 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Sage, like you asked:...
- 05:02 AM Revision b04b6f48 (ceph): mds: make nested scatterlock state change check more robust
- The predirty_journal_parents() calls wrlock_start() with nowait=true
because it has a journal entry open and we don't... - 12:52 AM Revision d64616b9 (ceph): mds: set layout on files created by MKNOD instead of CREATE
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:52 AM Revision 7d33830a (ceph): mds: set a writeable client range on regular files created via MKNOD
- If the client reexports ceph via nfs, file creations come through as
a MKNOD followed by OPEN. If it's a MKNOD on a ... - 12:52 AM Revision df7e3ffd (ceph): mds: issue caps on regular files created via MKNOD
- Give the client some caps to write with. This is similar to the current
MKDIR behavior, but with different bits.
Si... - 12:02 AM Revision 7df46936 (ceph): cephx: fix uninitialized value
- Signed-off-by: Sage Weil <sage@newdream.net>
12/17/2010
- 11:31 PM Revision 285f351b (ceph): mon: build_initial_monmap: fix mismatched alloc
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:30 PM Revision 3a235b0f (ceph): filestore: make OpSequencer::flush() work for writeahead journaling items
- It was only waiting for items in the op_queue to complete. The goal is
to wait for anything we've called queue_trans... - 11:26 PM Revision caa46093 (ceph): common: cleanups
- common_init: avoid (mismatched) heap allocation
ConfFile::_parse: avoid memory leak on error path
ConfFile: NULL fi... - 11:14 PM Revision 2f916086 (ceph): auth: avoid mismatched allocation
- Can't pair strdup and free.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 11:14 PM Revision 795811d6 (ceph): hadoop: fix a bunch of mismatched allocations
- Using array new means you need array delete.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 11:14 PM Revision f7dc1a92 (ceph): rgw: fix fd leak on error path
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:14 PM Revision 28bcf0bc (ceph): osd: PG::choose_acting: fix major iterator mistake
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 10:15 PM Revision 5184db44 (ceph): filestore: add per-sequencer flush operation
- Signed-off-by: Sage Weil <sage@newdream.net>
- 10:15 PM Revision 3c7d30f1 (ceph): osd: flush pg writes to disk before starting scrub scan
- This avoids two races:
- we just completed recovery by pushing objects to the replica, and the
replica starts sca... - 09:09 PM CephFS Bug #661 (Resolved): mds: mdlog recursion / scatter crash
- commit:b04b6f48237fcbd22716279726f827237a3eb4af
- 08:56 PM CephFS Bug #661 (Resolved): mds: mdlog recursion / scatter crash
- ...
- 08:51 PM Revision 2fb60daf (ceph): osd: debug scan_list and scrub a bit better
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:28 PM Revision c3a24fc5 (ceph): osd: refactor _get_pool / _put_pool a little bit
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 08:00 PM Revision 39aff221 (ceph): osd: OSD::_lookup_pool: avoid double lookup
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:52 PM Revision bf4529b9 (ceph): test: create test_pools.sh
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 06:59 PM Revision 1cfad2ea (ceph): osd: clear INCONSISTENT if scrub detects no errors
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:36 PM Revision b1908755 (ceph): osd: add assert that we're replica
- ar Fred saw a crash where we got into merge_log as a stray, which really
shouldn't ever happen! See #590.
Signed-of... - 04:59 PM CephFS Cleanup #660 (Closed): mds: use helpers in mknod, mkdir, openc paths
- Lots of not-quite-duplicate code here that can be cleaned up
- 04:43 PM Bug #531: Journaling Causes System Hang
- This is strange.
I tried a very large write.
root@cephnode01:/mnt# dd if=/dev/zero of=cephtest01 bs=1M count=1M... - 12:27 PM Bug #531: Journaling Causes System Hang
- Okay,
I got everything updated and it appears the cluster hung right about 180GB of data written. Which was double... - 11:41 AM Bug #531: Journaling Causes System Hang
- Just wanted to make an update.
I just got a chance to get back on the cluster today.
I have upgraded the cluste... - 04:31 PM Revision 1e291fc9 (ceph): debian: don't strip rados classes
- Signed-off-by: Laszlo Boszormenyi <gcs@debian.hu>
Signed-off-by: Sage Weil <sage@newdream.net> - 04:30 PM Revision 9c173bb4 (ceph): debian: rename ceph.lintian -> ceph.lintian-overrides
- Signed-off-by: Laszlo Boszormenyi <gcs@debian.hu>
Signed-off-by: Sage Weil <sage@newdream.net> - 04:15 PM Feature #659 (Resolved): direct io unit test
- For qa/workunits:
Test directio reads and writes where
- the file offset is 512-byte and not page aligned
- the ... - 01:29 PM Bug #629: cosd segfaults when deleting a pool containing degraded objects
- This shouldn't happen again commit:c3a24fc5d31d53e3db911be900b9067584f0e07e
It still might be interesting to see t... - 10:48 AM Bug #658 (Resolved): osd: scrub crash
- ...
- 10:40 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- ar Fred wrote:
> 2010-12-03 16:29:00.261788 7fe2f28a9710 osd1 1059 pg[3.70( v 660'2955057 lc 567'2954755 (643'295484... - 01:47 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- I spent some time yesterday trying the latest rc (914f6ddebd899667b1937dfe9d5f1a94537dc500), and I still hit the same...
- 09:33 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Stefan Majer wrote:
> Hi,
>
> we see similar problems with kernel 2.6.37-rc6 and ceph build from yesterday (29480... - 12:21 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Hi,
we see similar problems with kernel 2.6.37-rc6 and ceph build from yesterday (29480f42be8551f47d79282b7376a10a... - 09:02 AM Linux kernel client Bug #657 (Resolved): nfs-reexport null deref in ceph_init_dentry
- Al Viro observes:...
- 08:00 AM Revision b2746c94 (ceph): logging: eliminate out-of-range dout calls
- Eliminate calls to dout that use non-existent log levels, like negative
levels less than -1. Also trigger a compiler ... - 12:25 AM Revision c4b5c33b (ceph): common: more cerr -> derr conversions
- cmds: cerr -> derr
DoutStreambuf: primitive_log: just write to the stdout fd rather than cerr
assert: don't write o...
12/16/2010
- 11:31 PM Bug #656 (Closed): ceph
- i have 2 osd,[2.6.36,ceph-0.23.1]
#ceph osd down 0
#ls /mnt/ceph {ceph's mount dir}
hanging here,
after a while ... - 10:57 PM Revision 30f752cd (ceph): gceph,ceph: replace cerr->derr
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 10:05 PM Revision c76379fd (ceph): cosd: replace cerr with derr
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:06 PM Revision 73669d87 (ceph): PG.cc:
- sub_op_scrub must set finalizing_scrub on the replica
before waiting for last_update_applied to catch up to
info.la... - 06:32 PM Revision 4644247c (ceph): osd: FileJournal: use derr
- Use derr to announce errors in FileJournal.
Handle EINTR where necessary (still haven't fixed
read/write/pread/pwrit... - 06:32 PM Revision 4fc1af5e (ceph): logging: re-introduce derr
- Re-introduce derr as a special log level (level -1) which will show up
in all logs, and on stderr. These messages are... - 05:10 PM Tasks #653: get playground radosgw up and running again
- S3 gateway is up and running (I just recreated my user and tested it, it was working before). We still need to recrea...
- 03:02 PM Bug #655 (Resolved): class objects are being stripped (debian packages)
- When installing the latest debian packages, the resulting /usr/lib/rados-classes/*.so are completely stripped, thus w...
- 02:56 PM Tasks #654 (Resolved): get playground ladder0 mounted
- mounted, was just missing name=username on the mount command. Also created the users.
- 08:11 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- My rsync (196k files, 74GB data) finished succesfully, but the btrfs warning repeated itself twice.
- 02:09 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Compiled from the kernel GIT ( a4851d8f7d6351a395d36ae8fdcf41745a832d76 ) last night and then started a rsync this mo...
- 01:23 AM Revision 29480f42 (ceph): ReplicatedPG.cc:
- _scrub must set head when it encounters a head snap
curclone counts down, not up
Signed-off-by: Samuel Just <samuel... - 12:33 AM Revision 1e490eff (ceph): osd: timed out watcher is added to unconnected map
- 12:33 AM Revision c321620e (ceph): osd: send notify message only to unexpired watchers
- 12:26 AM Revision 619b45ad (ceph): logging: close file when reloading global config
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
12/15/2010
- 11:23 PM Revision 060fd428 (ceph): osd: fix watch timer, locking
- 09:39 PM Revision 914f6dde (ceph): filestore: detect final version of async ioctl SNAP_CREATE_V2
- Li's revised interface for the async snap ioctl is more flexible. Update
the ioctl call sites and detection code acc... - 09:15 PM Tasks #654 (Resolved): get playground ladder0 mounted
- Get it mounted (it was being weird on me). And recreate the users (probably servicectl ladder0:user config from yakko?)
- 09:14 PM Tasks #653 (Resolved): get playground radosgw up and running again
- I'm not sure what needs to be done to configure the radosgw pools..
Also we need to recreate the users/buckets fr... - 09:10 PM Bug #648 (Resolved): monclient: PGMap::apply_incremental
- trimming changed by commit:89d5c91e7d207d646651f8959ee37a15ea199d1b
- 09:10 PM Bug #631 (Won't Fix): OSD: FileJournal::committed_thru
- Sage Weil wrote:
> Okay, the first crash you saw is due to #645. I think it's a kernel bug causing that ioctl to fa... - 09:07 PM Revision 06a2d7a2 (ceph): mds: Save straydn in mdr so it's consistent across retry attempts.
- Otherwise, we could choose new stray dirs and fail to get all
the locks we needed (while leaving old strays locked fo... - 09:06 PM Cleanup #650 (Resolved): objecter: refactor request tracking to be per-osd instead of per-pg
- commit:d54a854811a51a5730b548da712d59761057fa58
- 08:44 PM Revision e31f0a47 (ceph): tools: don't start msgr thread before daemonize
- Calling messenger->add_dispatcher_head() has the side-effect of starting
the messenger thread. So we must not do it b... - 07:02 PM Revision d54a8548 (ceph): Merge branch 'objecter' into unstable
- 07:02 PM Revision 065cdf52 (ceph): objecter: track pending requests by osd, not pg
- This is a big cleanup. Also
- switch to keeping per-osd Connection *'s
- make requests time out independently (not... - 07:02 PM Revision f6dc5d9f (ceph): objecter: cleanup: rename op maps
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:02 PM Revision 32a8aed9 (ceph): objecter: add reopen_session helper
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:02 PM Revision 530083cc (ceph): objecter: check for pg mapping changes in each incremental; refactor mi...
- We need to detect when a pg mapping changes but the primary stays the same.
That means we can't just look at the fina... - 07:01 PM Revision 5d44d599 (ceph): msgr: mark down by Connection*
- Signed-off-by: Sage Weil <sage@newdream.net>
- 07:01 PM Revision 07e593c4 (ceph): mds: fix inode ancestor attr encoding
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:15 PM Revision fdbd85e4 (ceph): automake: ignore rmdir errors during uninstall
- We don't want to fail "make distcheck" for a silly reason.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 06:01 PM Bug #652 (Resolved): ReplicatedPG _scrub missing clone
- ReplicatedPG.cc:4016 = handle the missing clone case
- 03:18 PM Linux kernel client Bug #552 (Resolved): Samba with kernel oplocks=on produces lots of corrupt mds entries in dmesg
- Closing this out unless we hear about more issues.
- 01:38 PM Feature #643 (Resolved): filestore: update btrfs ioctl interface for soon-to-be-pushed SNAP_CREAT...
- commit:914f6ddebd899667b1937dfe9d5f1a94537dc500
- 01:01 PM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- I took the for-linus branch, since both next-rc and master wouldn't compile against 2.6.37-rc5. Due to this error:
... - 10:39 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- I've just hit the bug again, this while I was running a rsync to my Ceph cluster....
- 09:48 AM Feature #95: mon: adjust overload based on osd disk utilization
- Implemented reweight-by-utilization in the overload branch.
C. - 01:00 AM Revision 7b5e923c (ceph): osd: send pending notification for reconnected watcher
- 12:28 AM Revision f9694648 (ceph): automake: add osd/Watch.h to noinst_HEADERS
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
12/14/2010
- 09:55 PM Linux kernel client Bug #651 (Resolved): osd_client: need to recalculate request mapping for every osdmap incremental
- Currently if we get an osdmap message with multiple incrementals, and a request maps to a different osd and then back...
- 08:47 PM Revision c8d9b20c (ceph): Merge branch 'sync2' into unstable
- 07:50 PM Revision 89d5c91e (ceph): mon: trim pgmap less aggressively
- This will make observer crashes due to missed states (#648) much harder to
hit. Eventually the pgmap state trim prob... - 07:02 PM Revision 056e91e0 (ceph): librados: drop watch_lock
- Use the existing lock to do protect all of this.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:00 PM Revision d4420a8a (ceph): objecter: drop linger_info_mutex
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:51 PM Revision b989087d (ceph): crypto: catch cryptopp decrypt/encrypt exceptions
- 06:48 PM Revision 215f3320 (ceph): objecter: simplify linger register
- Drop single-use helper; make unregister_linger part of the public
interface.
Signed-off-by: Sage Weil <sage@newdream... - 06:47 PM Revision b60a9abf (ceph): objecter: fix up linger ack/commit to trigger first time only
- We only want the user-provided ack/commit callbacks to trigger the first
time we register the lingering op. Same goe... - 06:35 PM Revision 8a75086d (ceph): objecter: clean up linger interface
- Put LingerOp on heap. Use xlist to attach to PGs. Add in/out bufferlists.
Signed-off-by: Sage Weil <sage@newdream.... - 05:55 PM Revision 96b32382 (ceph): Merge remote branch 'origin/unstable' into sync2
- Conflicts:
src/auth/Crypto.cc
src/osd/ReplicatedPG.cc
src/osd/ReplicatedPG.h
src/osd/osd_types.h - 03:03 PM Feature #562 (Closed): separate gui into separate binary, package
- 02:56 PM Revision 3e076c39 (ceph): logging: use Mutex::Locker
- Use Mutex::Locker to make logging exception-safe. That is, if you are
doing "dout() << foo() << dendl;" and foo throw... - 02:20 PM Revision bf31f3f1 (ceph): logger: Fix DoutStreambuf::create_rank_symlink
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:13 PM Revision 4c377199 (ceph): cephtool: rename tools files
- Rename tools files to be more consistent. For example, the main()
function for ./ceph should be in ceph.cc.
Signed-o... - 12:50 PM rbd Feature #341: libvirt bindings
- A "network" disk type was introduced in "036ad5052b43fe9f0d197e89fd16715950408e1d":http://libvirt.org/git/?p=libvirt....
- 11:44 AM Bug #646: [OOPS] mount.ceph monip:6789: /mnt/ceph ,ceph-client-standalone.git + ceph 0.23
- I'm going to get the -standalone.git repo updated to the latest code (it's a bit out of date) so we can confirm this ...
- 11:36 AM Cleanup #650 (Resolved): objecter: refactor request tracking to be per-osd instead of per-pg
- 10:40 AM Bug #649 (Resolved): OSD: CryptoPP::StreamTransformationFilter::LastPut
- Fixed in b989087ddf8775588ddbb6234d099398a2e18072. CryptoPP threw an exception when failed to decode message (probabl...
- 02:34 AM Bug #649 (Resolved): OSD: CryptoPP::StreamTransformationFilter::LastPut
- This morning on my test machine (noisy.ceph.widodh.nl, 1 MON, 1 MDS, 3 OSD) all three OSD's died at exact the same mo...
- 09:58 AM Revision a3fcf908 (ceph): logging: Fix use-before-access in debug.cc
- Signed-off-by: Vangelis Koukis <vkoukis@cslab.ece.ntua.gr>
Signed-off-by: Constantinos Venetsanopoulos <cven@cslab.ec... - 09:53 AM Revision 3932f084 (ceph): osd: PG::prior_set_affected: const cleanup
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:14 AM Bug #648: monclient: PGMap::apply_incremental
- This is a known issue, caused by the pg state trimming. It'll go away eventually with #647. In the meantime, I'll m...
- 01:31 AM Bug #648 (Resolved): monclient: PGMap::apply_incremental
- I left my laptop on last night with a 'ceph -w' on one of my test machines, this morning I saw:...
12/13/2010
- 08:23 PM Feature #647 (Duplicate): mon: refactor paxos interaction
- We currently have a paxos instance per state machine, which is silly for a bunch of reasons. The big one is that a m...
- 07:28 PM Bug #646: [OOPS] mount.ceph monip:6789: /mnt/ceph ,ceph-client-standalone.git + ceph 0.23
- master-backport ,reproduce it.
unstable-backport,can't reproduce it ,
i'm not sure whether this issue had been fixed. - 07:00 PM Bug #646: [OOPS] mount.ceph monip:6789: /mnt/ceph ,ceph-client-standalone.git + ceph 0.23
- /ceph-client-standalone$ git branch
master
* master-backport
- 06:45 PM Bug #646 (Can't reproduce): [OOPS] mount.ceph monip:6789: /mnt/ceph ,ceph-client-standalone.git +...
- 1. ceph-client-standalone: git from git://ceph.newdream.net/git/ceph-client-standalone.git
2. ceph: ceph-0.23
3.OS:... - 10:18 AM Feature #640: support log rotation
- I guess when I filed this I was thinking of a setup where there was a small tmpfs partition where the logs went to, w...
12/12/2010
- 10:40 PM Revision 9add26be (ceph): mds: fix replay/resent vs completed request check
- If it is a _replayed_ request, we should always send a simple ack if it is
completed, because the client doesn't not ... - 09:51 PM rbd Tasks #421 (Resolved): get rbd support into qemu upstream
- 09:50 PM Linux kernel client Bug #473 (Can't reproduce): Kernel panic: ceph_pagelist_append
- 09:49 PM Linux kernel client Bug #304 (Can't reproduce): GPF in writepages_finish
- 09:47 PM Linux kernel client Feature #642 (Rejected): fill in s_uuid on superblock
- nevermind, i misread the nfs thread.. this is an extN thing.
- 02:33 PM CephFS Cleanup #638 (Resolved): mds: verify open+create resent/replayed event exception
- commit:9add26b
12/11/2010
- 09:58 PM Revision 0e08cb0f (ceph): osd: return ENOSPC for non-mds if full flag is set in osdmap
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:09 PM Revision 239b7677 (ceph): Merge remote branch 'origin/syslog' into unstable
- Conflicts:
src/mon/Paxos.cc
src/osd/PG.cc - 04:04 PM Revision 46242586 (ceph): Merge branch 'gceph' into unstable
- 08:20 AM Feature #562: separate gui into separate binary, package
- merged everything so far in commit:46242586eddcc948f71260f8c1ea2e8b1845a9f8
- 07:48 AM Feature #562: separate gui into separate binary, package
- Looks good! The one thing I'd change is to rename ceph.cc tool-common.cc or something along those lines, and cmd.cc ...
- 08:20 AM Feature #245 (Resolved): Logging to syslog
- merged in commit:239b7677e7a9df86d35cbfb25226c3f1a06771c5
- 08:09 AM CephFS Feature #630: release caps on inodes unlinked by other clients
- Sage Weil wrote:
> Or, the MDS needs to delete file data as soon as a stray's wanted drops to 0.
That won't work,... - 08:06 AM RADOS Feature #433: improve osd reweighting
- I think the thing to do here is extend the CrushWrapper interface (probably by wrapping something in mapper.c or buil...
- 04:34 AM Revision 292414c5 (ceph): gceph: Add gceph to rpm, deb
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 04:30 AM Revision 71a19a94 (ceph): gceph: run shutdown functions at exit
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 04:25 AM Revision bb82fd3d (ceph): gceph: fix compile
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 04:15 AM Revision 1a201f85 (ceph): gceph: add -h argument
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 04:12 AM Revision b4ceb194 (ceph): ceph tool: Create gceph
- Put the gui into a separate binary.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
12/10/2010
- 08:25 PM Feature #562: separate gui into separate binary, package
- I implemented this in the gceph branch.
cheers,
Colin - 04:28 PM Bug #631: OSD: FileJournal::committed_thru
- Okay, the first crash you saw is due to #645. I think it's a kernel bug causing that ioctl to fail.
This is actuall... - 03:18 AM Bug #631: OSD: FileJournal::committed_thru
- To answer your question, I never saw a crash with 2.6.37, I just rebooted back into 2.6.32
Just rebooting into 2.6... - 04:25 PM Bug #645 (Closed): intermittent failure of snap ioctl
- We're occasionally getting back an EINVAL from teh snap create ioctl.
extra debugging in place on the sepia test... - 04:15 PM Bug #644: rsync can be sloooow
- Copy of my notes file:
FS->FS
gregf@kai:~/ceph/src$ time rsync -r /btrfs/gregf/ceph-client/ mnt/ceph-client
skippi... - 04:15 PM Bug #644 (Closed): rsync can be sloooow
- This is probably due to metadata ops being fairly expensive, but we should inspect an rsync run over Ceph to make sur...
- 04:02 PM Feature #643 (Resolved): filestore: update btrfs ioctl interface for soon-to-be-pushed SNAP_CREAT...
- Chris is going to push Li's revision of the async ioctl for the next -rc.
12/09/2010
- 11:57 PM Revision 49844738 (ceph): librados, objecter: fix unwatch operation
- 10:38 PM Revision 346a2aac (ceph): rpm: update changelog
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 10:35 PM Revision e23d6200 (ceph): rpm: fix ceph.spec to work with gcephtool
- Don't try to package gui_resources unless we are building the GUI.
Get GUI dependencies correct.
Signed-off-by: Coli... - 09:55 PM Revision e5769b06 (ceph): objecter: resend_linger copies ops
- 09:45 PM Revision 83612ef7 (ceph): Fix overflow in FileJournal::_open_file()
- [ The following text is in the "iso-8859-7" character set. ]
[ Your display is set for the "iso-8859-1" character... - 09:09 PM Revision 329ae1bc (ceph): ReplicatedPG: snap_trimmer now acquires a read lock on the osd map
- before calling share_pg_info.
Signed-off-by: Samuel Just <samuelj@hq.newdream.net> - 09:09 PM Revision d0fbc30a (ceph): ReplicatedPG.cc: Fixes a bug in snap_trimmer where a pointer to a stack
- Cond is left in the mode.waiting_cond list.
Signed-off-by: Samuel Just <samuelj@hq.newdream.net> - 07:43 PM Bug #631: OSD: FileJournal::committed_thru
- Okay, pretty sure this was caused by a bug in 2.6.37-rc that was doing an async commit even for the sync snap/subvol ...
- 01:45 PM Bug #631: OSD: FileJournal::committed_thru
- Okay, confirmed it was the same bug you were seeing:...
- 07:18 PM Revision f68e6e7d (ceph): rpm: don't try to package radosacl
- radosacl is just a test binary, so unless we build with --with-debug, we
won't get it.
Signed-off-by: Colin McCabe <... - 07:18 PM Revision 6722b0c8 (ceph): rpm: add pkgconfig to BuildRequires
- You can't build without pkgconfig.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 06:28 PM Revision 9df18d19 (ceph): rpm: set files-attr for radosgw
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:13 AM Bug #639: RHEL6 beta packaging breaks on 'libcls_rbd.so.1.0.0.debug'
- The purpose of touching it is to prevent the installed .so from being stripped. Which means we don't need it in the ...
- 09:11 AM Linux kernel client Feature #642 (Rejected): fill in s_uuid on superblock
- 09:09 AM Feature #640 (Closed): support log rotation
- see logrotate.conf. it doesn't cap logs by size, but follows the standard scheme used by just about everything else ...
- 08:48 AM Feature #640 (Closed): support log rotation
- Here I'm talking about the log messages generated by dout() and friends.
We should allow users to set up log rotat... - 08:50 AM Feature #641 (Rejected): allow logs to be piped to an external program
- We already support sending logs to syslog, to stdout, or to a file. We could pretty easily support a fourth option, w...
- 02:10 AM Revision b4264fbb (ceph): filejournal: reset last_commited_seq if we find journal to be invalid
- If we read an event that's later than our expected entry, we set read_pos
to -1 and discard the journal. If that hap... - 12:02 AM Revision cc78bbf1 (ceph): objecter: create a new op for resending lingering requests
12/08/2010
- 09:51 PM Revision 027d5bfd (ceph): logger: tweak cmon log output a bit
- Make the output of cmon on stderr a little bit less verbose.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 08:25 PM Revision fdc7414e (ceph): logging: DoutStreambuf: handle daemonizing better
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:36 PM Revision 5cba1e63 (ceph): objecter: a few lingering fixes
- 07:12 PM Revision a9c098df (ceph): mon: use helper for clock drift check; log relative instead of absolute...
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:54 PM Revision 986c2af4 (ceph): logging: debug.h: use DoutStreambuf
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 03:52 PM Cleanup #299: catch std::bad_alloc and die with helpful error in log on ENOMEM
- This is kind of a tricky thing to really handle correctly.
In practice, most people run with memory overcommit tur... - 01:56 PM Feature #245: Logging to syslog
- 11:19 AM Bug #639 (Resolved): RHEL6 beta packaging breaks on 'libcls_rbd.so.1.0.0.debug'
- /usr/lib/debug/usr/lib64/rados-classes/libcls_rbd.so.1.0.0.debug is
installed but unpackaged.
In spec file I see ... - 11:11 AM CephFS Cleanup #638 (Resolved): mds: verify open+create resent/replayed event exception
- These look a bit fishy to me.
- 11:11 AM CephFS Bug #637 (Resolved): mds: check replica scatterlock flush on rejoin
- This needs to behave consistent with the start_flush/finish_flush hooks.
- 11:06 AM Bug #636 (Can't reproduce): RHEL6 beta packaging breaks on 'gui_resources'
- 'gui_resources' not getting into BUILDROOT correctly.
ceph.spec.in:128
%{_datadir}/ceph_tool/gui_resources/*
Co... - 10:31 AM Bug #635 (Resolved): RHEL6 beta packaging breaks unexpected characters
- ceph.spec.in uses @VERSION@ directly, which chokes rpmbuild when ''Version:' uses certain characters (for example 0.2...
- 07:27 AM Revision c53ffafb (ceph): logging: Remove _dout_check_log
- _dout_check_log is unneeded, since every invocation of dout makes the
same check.
Signed-off-by: Colin McCabe <colin... - 07:17 AM Revision 8fdd0f44 (ceph): logging: debug.h: minor cleanup
- Don't put std::ostream into the global namespace. Copyright update.
Signed-off-by: Colin McCabe <colinm@hq.newdream.... - 06:52 AM Revision aeba6bca (ceph): logging: eliminate dbeginl
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 03:09 AM Revision 0f0cb46a (ceph): logging: Implement rank symlinks
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:54 AM Revision 6c4a7d58 (ceph): logging: Support isym_path
- Support instance symlinks, which are activated when we are using
g_conf.log_per_instance.
Signed-off-by: Colin McCab... - 02:51 AM Revision e597d02d (ceph): logging: rename_output_file -> handle_pid_change
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:49 AM Revision 116478a3 (ceph): logging: _calculate_opath: use g_conf.log_dir
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:47 AM Revision ea3414d4 (ceph): logging: DoutStreambuf: better debug output
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:45 AM Revision 627399f7 (ceph): logging: create_symlink:sometimes use rel symlinks
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:43 AM Revision b00baab1 (ceph): logging: implement get_dirname, move get_basename
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 02:41 AM Revision ef223664 (ceph): logging: fix normalize_relative
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 12:44 AM Revision 2000f69e (ceph): mds: no not choose lock state on replicas
- The lock state has already been set during rejoin.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:44 AM Revision fe103003 (ceph): mds: sync->mix replica state is sync->mix(2)
- When auth first moves to sync->mix,
- auth sends AC_MIX to replicas
- replicas go to sync->mix
- replicas finish g... - 12:44 AM Revision 4f643994 (ceph): mds: introduce rejoin_invent_dirfrag() helper
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:44 AM Revision f97660ff (ceph): mds: fix LOOKUPHASH to avoid creating bogus replica CDir
- We can't create the CDir if we are non-auth.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:44 AM Revision 681b010f (ceph): mds: clear EXPORTINGCAPS on export_reverse
- We need to reverse the effects of encode_export_inode_caps(), which is just
the pin and state bit.
The original prob... - 12:44 AM Revision 9bbb33b4 (ceph): mds: send LOCKFLUSHED to trigger finish_flush on replicas
- Since f741766a we have triggered start_flush and finish_flush on replicas.
The problem is that the finish_flush didn'... - 12:44 AM Revision c681ed75 (ceph): mds: explicitly pass scatterlock dirty flag to auth on gather
- This ensures that if the replica is thinks it is flushing something the
auth will always do a scatter_writebehind.
S... - 12:44 AM Revision 39c5933d (ceph): mds: add missing try_clear_more() to scatterlock
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:44 AM Revision b5fd2e4d (ceph): mds: open undef dirfrags during rejoin
- Any invented dirfrags have a version of 0. This will cause problems later
if we pre_dirty() anything in that dir bec... - 12:44 AM Revision 2ea9b2d7 (ceph): mds: fix replay of already-journaled requests
- Check for already-completed tids for both retried and replayed requests.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:44 AM Revision 9b9b8693 (ceph): mds: rev mds cluster internal protocol
- The lock encoding changed with the dirty bit on scatterlocks.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:44 AM Revision 3825c4b8 (ceph): mds: small rejoin cleanup
- Signed-off-by: Sage Weil <sage@newdream.net>
12/07/2010
- 11:47 PM Revision 42464fb7 (ceph): logging: Add symlink helper functions
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 10:00 PM Revision e2ba601b (ceph): logger: fix EINTR handling
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:57 PM Revision bacdd493 (ceph): logging: rename_output_file: fix bug
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:55 PM Revision d70851ef (ceph): logging: DoutStreambuf: Implement log-to-file
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:46 PM Revision 95211145 (ceph): logging: Add log_to_file option
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:40 PM CephFS Bug #15 (Resolved): mds rejoin: invented dirfrags (MDCache.cc:3469)
- commit:b5fd2e4d4ee4bf02a993e75b756a3775b2d566e5
- 08:11 PM Revision df5d4e62 (ceph): logging: DoutStreambuf improvements
- Write to stdout_fileno directly rather than using a buffer, which we
would then have to flush. Fix a bug in the buffe... - 06:56 PM Revision 1e2e4aa0 (ceph): automake: in scripts, use sysconfdir as-is
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 06:48 PM Revision 10b6887e (ceph): automake: in deb pkg, use --syconfdir=/etc
- When building the debian packages, use --sysconfdir=/etc.
Also, don't fudge sysconfdir in the init-ceph script.
Sig... - 03:25 PM CephFS Feature #91: mds: up:shadow mode
- Okay, this seems to be working now. Had to adjust how the Journaler treated read_pos and to fix a few of my new re-re...
- 08:49 AM Linux kernel client Bug #634: Kernel client takes too long to recover after a MDS restart
- It's also possible (though unlikely) that the client isn't getting an updated MDSMap quickly enough or that the MDS t...
- 07:58 AM Linux kernel client Bug #634: Kernel client takes too long to recover after a MDS restart
- The client doesn't 'reconnect' until the MDS reaches the up:reconnect state. That's preceeded by up:replay (journal ...
- 07:53 AM Linux kernel client Bug #634 (Can't reproduce): Kernel client takes too long to recover after a MDS restart
- [208292.940934] libceph: mds0 192.168.1.11:6800 socket closed
[208293.050282] libceph: mds0 192.168.1.11:6800 connec... - 07:46 AM Revision d4043e81 (ceph): logging: add DoutStreambuf::set_prio
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:14 AM Revision 6c7735f6 (ceph): logging: DoutStreambuf must handle stdout + stderr
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:03 AM Revision 12544a49 (ceph): logging: Add log_to_syslog option
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 06:18 AM Revision 57bcdc54 (ceph): mkcephfs: require -k; update man page
- Force users to specify keyring location; update man page accordingly.
Signed-off-by: Sage Weil <sage@newdream.net>
12/06/2010
- 11:57 PM Revision 5ac581df (ceph): Rename SyslogStreambuf -> DoutStreambuf
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:53 PM Bug #631: OSD: FileJournal::committed_thru
- Oh yes, there seems to be a issue with the IPv6 connectivity where 'noisy' is at.
I added debug journal = 20 (I in... - 10:30 PM Bug #631: OSD: FileJournal::committed_thru
- I updated the above comment but I suspect you only looked at the email notification? In any case, can you reproduce ...
- 03:13 AM Bug #631: OSD: FileJournal::committed_thru
- Yes, simply starting the OSD's again gave me the same crash on one OSD.
I've attached the log, but here are the la... - 11:38 PM Revision 9811fbd0 (ceph): logging: Replace derr with dout
- derr was really just an alias for STDERR. Unfortunately, after we call
daemonize, STDERR is connected to /dev/null. S... - 11:38 PM Revision c94e0d2d (ceph): logging: optimize with likely/unlikely macros
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:38 PM Revision d1e0a2ae (ceph): logging: debug.h: move some debug functions
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:38 PM Revision ab18aaec (ceph): logging: add g_conf.clog_to_syslog
- Add a new configuration option that allows you to send central log
messages to syslog.
Signed-off-by: Colin McCabe <... - 11:35 PM Revision ab61823e (ceph): logging: LogEntry: don't pass enums by reference
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:30 PM Revision 4ef069c3 (ceph): logging:Move LogEntry.h into common with LogClient
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:30 PM Revision 82fa7f2d (ceph): logging: LogClient: refactor handle_log_ack
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:30 PM Revision fcae8a7a (ceph): logging: MLog.h: const cleanup
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:29 PM Revision f2ead26e (ceph): logging: better syntax for LogClient
- Rather than having to write logclient.log(LOG_ERROR, ss), coders can now
write clog.error() << "str". Auto-flushing, ... - 11:25 PM Revision 87545d06 (ceph): configure: detect crypto++ library
- 10:01 PM Revision ebcc9395 (ceph): osd: drop not-quite-copy constructor for object_info_t
- Making a copy-like constructor that doesn't actaully copy is confusing
and error prone. In this case, we initialized... - 08:49 PM Revision d69f3dd3 (ceph): MDS: Encode a full ancestor trace on inodes, not just the immediate par...
- 07:17 PM Revision 11c7dc03 (ceph): librados: fix the C++ interface init
- 07:17 PM Revision b1afea51 (ceph): librados: fix error path in rados_deinitialize
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:16 PM Revision aa3dda61 (ceph): librados: fix the C++ interface init
- 06:31 PM Revision 9a604816 (ceph): librados: fix C interface error handling in init code
- 06:28 PM Revision 130b8b3f (ceph): librados: fix C interface error handling in init code
- 05:59 PM Revision bf030ca2 (ceph): client: resync ioctl header from ceph-client.
- Previous change to the CEPH_IOCTL_MAGIC in fbbf448 was incorrect!
Signed-off-by: Greg Farnum <gregf@hq.newdream.net> - 12:42 PM CephFS Feature #600 (Resolved): mds: store full trace on directories
- Done in commit:d69f3dd327730a61b614c9f41f6155626bc07686. Just loops through the parents and encodes them sequentially...
- 11:04 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- If you're not running btrfs, you can also copy a pg directory to another disk and symlink it. Just be sure to preser...
- 11:01 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- Hi Fred,
We don't support deleting things manually from the object store while cosd is running.
Are you running... - 12:52 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- At the time, when I said "by hand", it was even worse:
As the journal was full, doing a rm -Rf while the osd was s... - 11:03 AM CephFS Bug #451 (Closed): mds: replay error
- It look like the cluster is running pretty old code (0.22~rc). This particular problem was fixed by 1c934ebd (0.23).
- 10:22 AM CephFS Bug #451 (In Progress): mds: replay error
- 03:59 AM CephFS Bug #451: mds: replay error
- <removed sensitive info>
- 10:25 AM Bug #633 (Resolved): librados crashes when init failed
- Fixed with commit:9a60481681d86065979d4353305cdaad74fe1a01
- 10:25 AM Bug #633 (Resolved): librados crashes when init failed
- When the librados init fails, specifically when using the C interface, the subsequent call to deinitialize() crashes.
- 06:20 AM Revision 4e3c2011 (ceph): Tune Debian packaging for the upcoming v0.24 release.
- Including switch OpenSSL dependency to Crypto++ as its being used instead of
the former; remove radosacl as its not c...
12/05/2010
- 09:03 PM Bug #631: OSD: FileJournal::committed_thru
- Wido, any chance you can reproduce this with 'debug filestore = 20' and 'debug journal = 20'?
- 01:55 AM Bug #631 (Won't Fix): OSD: FileJournal::committed_thru
- On a small cluster (1 MDS, 1 MON, 3 OSD's) I just saw 2 OSD's crashing with the same backtrace:...
- 08:32 PM CephFS Feature #601: mds: order directory commits after rename
- What if another file is renamed in the other direction? Do a partial commit first? Tricky.
- 06:46 PM Bug #632 (Won't Fix): init script won't stop an instance that's been removed from config
- If an instance of a Ceph daemon is removed from the config file, then the Ceph init script no longer knows how to kil...
- 05:29 AM Revision 27b70eb5 (ceph): osd: search for unfound on osds in might_have_unfound
- We were looking at 'up', which is just the set of OSDs we should be on in
the current epoch; nothing to do with where... - 04:45 AM Revision 8aa7b391 (ceph): Makefile: make radosacl build WITH_DEBUG only
- Signed-off-by: Sage Weil <sage@newdream.net>
12/04/2010
- 08:53 PM Tasks #616 (Rejected): radosacl needs a man page
- 08:52 PM Bug #627 (Resolved): replace openssl with crypto++
- 08:38 PM CephFS Feature #630 (Resolved): release caps on inodes unlinked by other clients
- If client A writes a file, and client B unlinks it, client A needs to drop the inode sooner rather than later.
O... - 03:34 AM Revision 15d8bdf3 (ceph): crypto: use crypto++ for aes instead of openssl
- need to implement it more efficiently, currently going through a string object
- 03:34 AM Revision 58f3ce4a (ceph): crypto: test for allocation failure, cleanup
- 03:34 AM Revision 6ec622c0 (ceph): common: use ceph_armor instead of openssl based functions
- also modify ceph_[un]armor to get dest buffer length
- 03:34 AM Revision 7fa9426c (ceph): makefile.am: most binaries (except rgw_*) don't link with openssl
- 03:34 AM Revision e135e924 (ceph): crypto: remove old openssl implementation
- 03:34 AM Revision 76e02c71 (ceph): common: remove base64.c
- 03:34 AM Revision 88213770 (ceph): crypto: change include
- 03:34 AM Revision a28b4494 (ceph): configure: check for the presence of libcrypto++ header files
- 03:34 AM Revision f2424dfb (ceph): rgw: get rid of openssl altogether
- 03:34 AM Revision e0059259 (ceph): rgw: null terminate armor result
- 03:34 AM Revision 23f37043 (ceph): ceph.spec.in: update dependency
12/03/2010
- 06:02 PM Revision a457cbb9 (ceph): mon: fix typo
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:02 PM Revision 378d13df (ceph): osd: remove poid/soid from ScrubMap::object; clean up callers
- The soid is in the key in the map; no need to store it in the value.
Update the scrub code appropriately.
Signed-off... - 05:35 PM Revision a4cc929c (ceph): make: create log directories and tmp directories
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 05:10 PM Revision a5297388 (ceph): msgr: Correctly handle half-open connections.
- If poll() says a socket is ready for reading, but zero bytes
are read, that means that the peer has sent a FIN. Hand... - 11:11 AM Bug #590 (In Progress): osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- 11:11 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- Hi Fred,
When you say you removed the PGs "by hand"... does that mean you used "rm -rf" on the object store while ... - 07:33 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- this is with ceph rc 39b42b21e9805b3ec838f8682420166fede719f2
I tried to solve the ENOSPC problem by removing PGs ... - 09:36 AM CephFS Bug #623 (Resolved): MDS: MDSTable::load_2
- 12:32 AM CephFS Bug #623: MDS: MDSTable::load_2
- Yes, tried with the latest rc, works!
MDS starts and recovers, als mounting and using the FS goes fine. - 09:35 AM Bug #625 (Resolved): make install should create dirs
- implemented by commit:a4cc929cedb0ee773a2fa68d691a9951221ae31a and commit:39b42b21e9805b3ec838f8682420166fede719f2
C. - 01:35 AM Revision 39b42b21 (ceph): make: create /etc/ceph if it doesn't exist
- make: create /etc/ceph if it doesn't exist. On uninstall, remove the
directory if it's empty. (Never remove a user's ... - 12:56 AM Revision da5ab7c9 (ceph): ost: object_info_t: decode old versions correctly
- object_info_t has one constructor that initializes everything from a
bufferlist. This means that the decode function ... - 12:18 AM Revision 03eb4e7a (ceph): man: add man page for cephfs
- Add to Makefile, debian, and ceph.spec.in bits
12/02/2010
- 07:52 PM Revision 6518fae3 (ceph): watch: some more linger fixes
- 06:16 PM CephFS Feature #91: mds: up:shadow mode
- I have yet to implement trimming, but the basic restarting-replay bits are now in place along with hooks to make it s...
- 05:14 PM Bug #479: ceph/mount crash badly when writing
- Hi all:
Ok, so I gitted again, original/unstable,
- Linux ss1 2.6.36-02063601-generic #201011231330 SMP Tue Nov 2... - 05:07 PM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- 05:07 PM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- Hi Fred,
I think the assertion you're seeing here was fixed very recently by commit:78a14622438addcd5c337c4924cce1... - 05:02 PM Bug #629: cosd segfaults when deleting a pool containing degraded objects
- Looks like some kind of lifecycle issue related to deleting pools.
OSD::_remove_pg does a _put_pool, and that does... - 04:54 PM Bug #629 (Resolved): cosd segfaults when deleting a pool containing degraded objects
- started a 4 node osd cluster. created some pools with some objects in them. killed one osd node. waited for it to be ...
- 04:52 PM CephFS Bug #623: MDS: MDSTable::load_2
- I think that commit:da5ab7c9a49f8996b41783175683d4b8b13ece4d should fix this issue.
wido, can you re-run with the ... - 04:44 PM CephFS Bug #623: MDS: MDSTable::load_2
- root@noisy:/var/log/ceph# grep mark_all_unfound_as_lost *
[ no results ]
So we're not marking things as lost in... - 11:52 AM CephFS Bug #623: MDS: MDSTable::load_2
- actually -23 is NFILE, which is I think coming from the LOST code...but that should never trigger unless the admin ha...
- 05:12 AM CephFS Bug #623 (Resolved): MDS: MDSTable::load_2
- On a small test machine I have a Ceph RC cluster running (Which was running a old unstable before), after my upgrade ...
- 04:13 PM Tasks #617: cephfs needs a man page
- Already had it in the Makefile, put it in the other bits and updated the commit.
- 03:56 PM Tasks #617: cephfs needs a man page
- need to add filename to debian/ceph.install and ceph.spec.in too.
and to man/Makefile.am - 02:07 PM Tasks #617 (Resolved): cephfs needs a man page
- Done in commit:6cdaa2f6a7670357313401ddbd322bdf529a1547 on the rc branch.
- 03:29 PM Bug #622 (Resolved): crushtool useless parse error
- Resolved-- the crushmap.txt was bad.
I created #628 for getting better error messages from crushtool. - 01:19 PM Bug #622: crushtool useless parse error
- There is a more advanced error handling API for spirit described at:
http://www.boost.org/doc/libs/1_41_0/libs/spiri... - 11:35 AM Bug #622: crushtool useless parse error
- Reposting the diff; hopefully clearer this time.
--- crushmap.txt.1 2010-12-02 11:38:43.816441440 -0800
++... - 11:33 AM Bug #622: crushtool useless parse error
- I was able to get the crushmap.txt to work by deleting the word "domain" in the gb1 region.
We should definitely h... - 03:07 AM Bug #622 (Resolved): crushtool useless parse error
- I can't decide whether this is a bug in crushtool or a bug in my crushmap but whichever it is, the error message isn'...
- 03:27 PM RADOS Feature #628 (New): crushtool: better error messages when parsing a crushmap.txt
- There is a more advanced error handling API for spirit described at:
http://www.boost.org/doc/libs/1_41_0/libs/spiri... - 03:22 PM Bug #625: make install should create dirs
- Should be pretty straightforward. The only question is, should we remove those directories on an uninstall?
- 11:11 AM Bug #625 (Resolved): make install should create dirs
- /var/log/ceph
/var/lib/ceph/tmp
?
check debian/ceph.dirs to see what else gets created... - 11:57 AM Bug #627 (Resolved): replace openssl with crypto++
- https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/684011
- 11:25 AM CephFS Feature #626 (Closed): qa: add IOR, rompio, or other parallel workloads suite
- We've had reports that rompio is just terrifically unstable, and shows serious scaling issues.
IOR is a more commo... - 09:39 AM Feature #624: radostool: make 'put' write large objects in chunks
- Can we be able to set the chunk size with an argument, for testing this kind of thing in future?
- 09:38 AM Feature #624 (Resolved): radostool: make 'put' write large objects in chunks
- otherwise a put on a large (100mb+) file can fail because it exceeds the size of the osd journals. it's also clearly...
12/01/2010
- 11:40 PM Revision 78a14622 (ceph): osd: fix log tail vs last_complete assert on replica activation
- The last_complete may be below the log tail IFF we have a backlog.
Fixes 756918be3b24d8164699da301ddfbc8e6fd6b751.
... - 11:11 PM Revision 63fab458 (ceph): rados_bencher.h:
- bench_write and bench_seq will now wait on any write/read
rather than the one least recently started.
bench_write ... - 11:00 PM Revision 0ea601ab (ceph): Create SyslogStreambuf
- SyslogStreambuf is a kind of stream buffer that allows you to output
characters from an ostream to syslog. Most stand... - 09:48 PM Revision a3d8c527 (ceph): filestore: call lower-level do_transactions() during journal replay
- We used to call apply_transactions, which avoided rejournaling anything
because the journal wasn't writeable yet, but... - 09:46 PM Revision 9ecbc300 (ceph): filestore: do journal mode autodetect and sanity check _before_ replay
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:25 PM Tasks #617: cephfs needs a man page
- I'll get this tomorrow. I wrote the tool and have had a task in my private manager to do this ever since then.
- 07:05 PM Revision f9fa855a (ceph): filestore: fix journal locking on trailing mode
- We're already holding journal_lock due to the surrounding
op_submit_{start,finish}.
Signed-off-by: Sage Weil <sage@n... - 06:20 PM Revision 0897edaf (ceph): Merge branch 'testing' into rc
- Conflicts:
configure.ac - 06:20 PM Revision cbb56208 (ceph): rbd: use MIN instead of min()
- Not even sure where min() was coming from, but it seems to be missing on
i386 lucid.:
g++ -DHAVE_CONFIG_H -I. -W... - 06:20 PM Revision 792b04ba (ceph): client: connect to export targets on cap EXPORT
- Also unconditionally connect on reconnect, even when there aren't any
outstanding requests.
Signed-off-by: Sage Weil... - 06:03 PM Revision bde0c721 (ceph): filestore: do not autodetect BTRFS_IOC_SNAP_CREATE_ASYNC until interfac...
- Li has proposed an alternative V2 ioctl that looks nicer, so wait until
that is finalized.
Signed-off-by: Sage Weil ... - 06:03 PM Revision 5bdae2af (ceph): ceph v0.23.2
- 05:44 PM Revision 4592c220 (ceph): client: fix cap export handler
- An EXPORT cap msg can race with a cap release; deal with that (realigning
this code with the kclient).
Signed-off-by... - 05:24 PM Revision 15c272e8 (ceph): man: fix monmaptool man page
- I've found the manpage problem that I've noted before. It's about
monmaptool, the CLI says it's usage:
[--print] [--c... - 03:17 PM Bug #611 (Resolved): OSD: OSDMap::get_cluster_inst
- 03:17 PM Bug #612 (Resolved): OSD: Crash during auto scrub
- 02:58 PM Linux kernel client Bug #564 (Resolved): Configuration via configfs instead of sysfs
- acked by greg kh, yay
- 02:58 PM rbd Bug #391 (Can't reproduce): snap create/delete caused corruption
- this is old
- 02:47 PM Bug #550 (Can't reproduce): mon: PGMonitor::update_from_paxos()
- haven't been able to reproduce this. commit:62716aa7 gives us useful error messages. if/when it comes up again we'l...
- 02:28 PM Linux kernel client Bug #436 (Can't reproduce): cmon: basic_string::_S_construct NULL not valid
- 02:28 PM Bug #460 (Can't reproduce): OSD crash: ReplicatedPG::push_to_replica / Rb_tree
- 09:46 AM Bug #621 (Resolved): error building unstable branch, rbd.cc:837: error: no matching function for ...
- should be fixed by commit:307404231ecb09fdd2f6dd6e50677e746bba4236
- 07:08 AM Bug #621 (Resolved): error building unstable branch, rbd.cc:837: error: no matching function for ...
- Building on i386 Ubuntu Lucid, it fails building rbd.
This is a build of unstable at commit bf784cdb4f605c467eb094... - 09:14 AM CephFS Bug #344 (Resolved): cfuse should pass all qa tests
- Sage asked me to mark this resolved. I ran the bonnie test yesterday and it eventually crashed when the disk ran out ...
- 02:22 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- I just tried the latest unstable: fe9fad7bea
osd log attached...
osd/OSD.cc: In function 'void OSD::_process_pg... - 12:50 AM Revision 6d96104e (ceph): osd: simplify scrub sanity checks
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:50 AM Revision 76b55c8a (ceph): osd: only adjust osd scrub_pending if pg was reserved
- If for some reason we enter scrub() without scrub_reserved == true, don't
adjust the osd->scrubs_pending or we'll scr... - 12:38 AM Revision 260840f5 (ceph): mds: fix import_reverse re-exporting of caps
- Make the import_reverse() set the pin/state before it clears them by using
the helper that sets them.
Signed-off-by:... - 12:25 AM Revision fe9fad7b (ceph): v0.25~rc
- 12:25 AM Revision 109e3f18 (ceph): mds: turn off mds_bal_frag until resolve vs split/merge is fixed
- See #594
Signed-off-by: Sage Weil <sage@newdream.net> - 12:11 AM Revision f216b020 (ceph): Merge remote branch 'origin/lost' into unstable
- Conflicts:
src/osd/osd_types.h
11/30/2010
- 11:48 PM Revision 0cc8d34e (ceph): osd: refactor object_info_t constructor a bit
- Create a copy constructor for object_info_t, since we often want to copy
an object_info_t and would rather not try to... - 11:48 PM Revision c281e1e0 (ceph): osd: mark_all_unfound_as_lost: wake waiters
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:48 PM Revision d5e6cae2 (ceph): radostool: fix memleak in error path
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:48 PM Revision 55f7e567 (ceph): osd: mark_all_unfound_as_lost: set lost attr
- In mark_all_unfound_as_lost, we need to set the lost bit in the objects'
object_info_t.
Signed-off-by: Colin McCabe ... - 11:48 PM Revision 5e243f3e (ceph): osd: create lost2 test
- This one verifies:
1. Client asks for an unfound object and gets put to sleep
2. Object gets declared lost
3. Client ... - 11:48 PM Revision b46f847c (ceph): osd: mark_obj_as_lost: don't assume we have obj
- In PG::mark_obj_as_lost, we have to mark a missing object as lost. We
should not assume that we have an old version o... - 11:48 PM Revision c29fbb12 (ceph): osd: mark_all_unfound_as_lost: bugfix, refactor
- mark_all_unfound_as_lost: just delete items from the rmissing set as we
find them, rather than using a multi-pass sys... - 11:48 PM Revision e9ccd7eb (ceph): osd: mark_obj_as_lost: fix oloc init, eversion
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:48 PM Revision cee3cd51 (ceph): osd: share_pg_log: update peer_missing
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:48 PM Revision ad4e5f36 (ceph): osd: ReplicatedPG::do_op: error on read-from-lost
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:48 PM Revision b15a97c7 (ceph): test_lost: add lost1 test
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:47 PM Revision 136dfdeb (ceph): osd: don't mark objs as lost unless we're active
- We don't have enough information to mark objects as lost until we
activate the PG. might_have_unfound isn't even buil... - 11:43 PM Revision 08bd4ead (ceph): mds: fix resolve for surviving observers
- Make all survivors participate in resolve stage, so that survivors can
properly determine the outcome of migrations t... - 11:43 PM Revision fb4734be (ceph): (re)add mechanism for marking objects as lost
- In activate_map, we now mark objects that we know are unfindable as
lost. This relies on the might_have_unfound set i... - 11:43 PM Revision 80f3ea10 (ceph): Add ./ceph dump pg debug degraded_pgs_exist
- ./ceph dump pg debug degraded_pgs_exist returns TRUE if some pgs are
degraded; false otherwise.
tests: move start_re... - 11:43 PM Revision de094224 (ceph): osd: object_info_t: add lost field
- We can now permanently mark objects as lost by setting the lost bit in
their object_info_t. Rev the object_info_t str... - 11:43 PM Revision e555899c (ceph): osd: active replicas process logs from primaries
- In _process_pg_info, if the primary sends us a PG::Log, a replica should
merge that log into its own.
mark_all_unfou... - 11:43 PM Revision c0e60afe (ceph): test: dump_osd_store: sort dump output
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:21 PM Revision 1123b5c5 (ceph): osd, librados: misc fixes, linger related issues
- 08:57 PM Revision bf784cdb (ceph): osd: fix object_info_t() initialization of oloc
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:56 PM Revision 91a75590 (ceph): mds: add debug output to make completions easier to track
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:48 PM Revision ba1f3cb9 (ceph): osd: fix misuses of OLOC_BLANK
- Commit 6e2b594b fixed a bunch of bad get_object_context() calls, but even
with the parameter fixed some were still br... - 08:23 PM Revision 2ad901b3 (ceph): Revert "mds: resolve cleanup"
- This reverts commit cd53719f3ce712a060e4ac80cab934c597531a5e.
We need this on surviving nodes too to resolve ambiguo... - 08:19 PM Revision b39f0425 (ceph): Merge branch 'testing' into unstable
- Conflicts:
src/os/FileJournal.cc - 07:43 PM Revision 1b06332d (ceph): osd: make recovery_oids debug list per-pg
- Otherwise we hit bad asserts if an object of the same name in different
pools is getting recovered simultaneously.
S... - 06:56 PM Revision 05ad97b6 (ceph): client: Set the DirResult buffer to NULL when deleting it.
- This should fix a crash exposed by our bonnie workunit. Previously
the client would keep trying to read out of the (d... - 05:22 PM Revision 559d4d20 (ceph): ceph.spec.in: include gui files
- Signed-off-by: Sage Weil <sage@newdream.net>
- 05:13 PM Revision 93601269 (ceph): debian: many many cleanups
- Signed-off-by: Laszlo Boszormenyi <gcs@debian.hu>
- 04:55 PM Revision 5eb8ef7f (ceph): filejournal: fix throttle vs FULL behavior
- We don't want to add to the throttler if we aren't going to queue the
write, or else we'll never take it off again.
... - 04:45 PM Bug #612: OSD: Crash during auto scrub
- this should be fixed by commit:76b55c8a121acd4e5e8b6f5dbb83c25926ac9f76
- 04:32 PM Revision 132f74c5 (ceph): Merge branch 'osd_journaling' into unstable
- 04:30 PM Revision 7af9ffdf (ceph): filestore: make sure blocked op_start's wake up in order
- If they wake up out of order (which, theoretically, they could before) we
can screw up journal submitting order in wr... - 04:24 PM Revision fac7266d (ceph): filestore: assert op_submit_finish is called in order
- Verify/assert that we aren't screwing up the submission pipeline ordering.
Namely, we want to make sure that if op_ap... - 04:20 PM CephFS Bug #594: mds: frag split/merge vs replay
- disabled in v0.24
- 04:01 PM Tasks #539 (Resolved): wiki: document pg expansion
- Documented on:
http://ceph.newdream.net/wiki/Changing_the_number_of_PGs - 03:54 PM Revision 5e391db0 (ceph): filejournal: rework journal FULL behavior and fix throttling
- Keep distinct states for FULL, WAIT, and NOTFULL.
The old code was more or less correct at one point, but assumed th... - 03:51 PM Revision 79419c33 (ceph): filestore: refactor op_queue/journal locking
- - Combine journal_lock and lock.
- Move throttling outside of the lock (this fixes potential deadlock in
parallel j... - 03:22 PM Revision 0df9dd6e (ceph): filestore: do not throttle op_queue in queue_op()
- In parallel mode, queue_op is called while holding the journal lock, so it
is not okay to throttle there. Instead, t... - 12:25 PM Feature #620 (Resolved): objecter: (optionally) read from replica if on localhost and primary is not
- This can either compare the ip address, or possibly have a netmask (set in g_conf) to determine 'locality' (where 255...
- 12:23 PM Feature #619 (Resolved): objecter: optionally read from replicas
- Add a read flag to allow reads to come from a random replica. If a replica replies with EAGAIN, retry the request, b...
- 12:21 PM Feature #618 (Resolved): osd: allow reads from replicas
- Allow osd to handle reads on a replica. If the replica is missing the object in question, reply with -EAGAIN to the ...
- 11:41 AM Bug #613 (Resolved): OSD crash: FAILED assert(recovery_oids.count(soid) == 0)
- this was actually a problem with the debug sanity checks. fixed by commit:1b06332de69b332092d115451efbd29afec79269
- 10:04 AM Tasks #617 (Resolved): cephfs needs a man page
- 10:04 AM Tasks #616 (Rejected): radosacl needs a man page
- 08:36 AM Bug #615 (Resolved): osd: improve op+journal throttling
- Currently we block first, then take locks, then update the throttle accounting. This makes things racy, because a bu...
- 08:34 AM Bug #598 (Resolved): osd: journal reset in parallel mode acts weird
- fixed as of commit:132f74c56064fdb3c47943679c48aa2a6b98f4eb, along with a ton of other related issues with the io que...
- 02:49 AM Revision 8003915b (ceph): Makefile: add bloom_filter.hpp to noinst_HEADERs
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 01:16 AM Revision 62075f34 (ceph): Makefile: Fix VPATH builds
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 12:41 AM Revision 0bcdc84a (ceph): osd: osd_types.h: const cleanup
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 12:40 AM Revision 7ee50add (ceph): osd: don't try to load a PG in a nonexistent pool
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 12:38 AM Revision 6ab17236 (ceph): filestore: simplify apply_transactions
- Always use queue_transactions, even in no-journal case.
Signed-off-by: Sage Weil <sage@newdream.net>
11/29/2010
- 11:52 PM Revision c9f864a0 (ceph): osd: PG::trim: fix inverted conditional in assert
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:12 PM Revision b2bcf4b3 (ceph): common: prevent infinite recursion on SIGSEGV
- Install SIGSEGV / SIGABORT handlers with sigaction using SA_RESETHAND.
This will ensure that if the signal handler it... - 10:12 PM Revision 85191813 (ceph): osd: Create pg_split test
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:35 PM Revision fb60e114 (ceph): logger: Fix a crash when the MDS shuts down cleanly.
- We weren't holding the lock on the logger_timer before calling shutdown.
- 09:35 PM Revision b4db4100 (ceph): Timer: add some asserts to catch certain errors.
- 08:56 PM Revision adbb5459 (ceph): osd: some notify simplifications and FIXMEs
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:56 PM Revision ec15c465 (ceph): osd: track unconnected_watchers and when they expire
- - set up an initial expiration when we load the obc off disk
- remove expiration when we connect to an existing watch... - 08:55 PM Revision 376870fa (ceph): osd: add timeout to watch_info_t
- Allow the watch timeout be set on a per-watch basis. Still need to figure
out where that comes from.. the client? A... - 08:55 PM Revision 239c0a12 (ceph): rbd: fix version renaming
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:55 PM Revision b3051531 (ceph): osd: fix up WATCH
- Separate various paths: registering new watch, reconnecting to existing
watch, removing watch, etc.
Signed-off-by: S... - 08:55 PM Revision 2563905b (ceph): osd: some cleanup
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:55 PM Revision b722662e (ceph): osd: use pg_t to find PG's again
- The ceph_object_layout is approaching obsolete. Also, use a more general
lookup_lock_raw_pg() helper that doesn't ta... - 08:54 PM Revision a61f6b5e (ceph): osd: add missing Watch.cc
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:54 PM Revision 0e62c421 (ceph): osdc: spell out version
- Cosmetic
Signed-off-by: Sage Weil <sage@newdream.net> - 08:51 PM Revision 15ffbc8d (ceph): makefile: add missing MWatchNotify.h
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:50 PM Revision 4dca64b2 (ceph): osd: drop unused fields
- Signed-off-by: Sage Weil <sage@newdream.net>
- 08:18 PM Revision 463d624d (ceph): Makefile: Add --as-needed to LDFLAGS
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:51 PM Revision a77eb6bd (ceph): vstart.sh: don't specify journaling mode
- Let the autodetection kick in, or let the dev specify via -o '...'.
Signed-off-by: Sage Weil <sage@newdream.net> - 07:41 PM Revision e0b927b2 (ceph): osd: PG::trim: add assert
- Assert that we're not trimming the PG log past last_complete.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 05:48 PM Revision 756918be (ceph): osd: _process_pg_info: add assert for replicas
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 05:06 PM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- Fred, can you see if this reproduces on the latest unstable? Thanks.
-C
- 11:14 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- I added the PG::trim assert. It seems to cause problems immediately with test_unfound.sh
The plot thickens... - 10:36 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- Argh yeah I was all wrong here. The recovery code looks ok.. I think the problem is that _before_ this the log was t...
- 09:21 AM Bug #590: osd/PG.cc:1645: FAILED assert(info.last_complete >= log.tail || log.backlog)
- > The replicas only ever get messages from the primary, and the primary
> sends a log to activate. Never anything e... - 04:51 PM Bug #614: SEGV loop on _open_lock_pg after rmpool
- Er, by that I mean:
load_pgs shouldn't try to load a PG that is in a nonexistent pool. This could only happen aft... - 04:49 PM Bug #614 (Resolved): SEGV loop on _open_lock_pg after rmpool
- In OSD::load_pgs, we weren't checking to make sure that the pool existed when going through all the collections.
F... - 02:23 PM Bug #614 (Resolved): SEGV loop on _open_lock_pg after rmpool
- discovered my cosd processes at 100%, possibly following some "rados rmpool" commands to delete some pools. Stopped ...
- 04:41 PM Bug #598: osd: journal reset in parallel mode acts weird
- bunch of problems here, not all related to a full journal.
- 12:18 PM Feature #568 (Resolved): debian: build with --as-needed?
- Implemented!
before:
cmccabe@flab:~/src/ceph2/src$ ldd .libs/rados
linux-vdso.so.1 => (0x00007fff4eff... - 11:13 AM Bug #575 (Resolved): monmaptool terminates when input file is not a monmap
- 10:49 AM Bug #479 (Can't reproduce): ceph/mount crash badly when writing
- 10:15 AM CephFS Subtask #547 (Resolved): mds: define fsck strategy, required metadata
- 10:13 AM CephFS Bug #594: mds: frag split/merge vs replay
- needs to be fixed in 0.24, or g_conf.mds_frag needs to be disabled.
- 10:06 AM Bug #595 (Won't Fix): Autogen: not a literal
- seems to go away with latest automake
- 07:12 AM Bug #613 (Resolved): OSD crash: FAILED assert(recovery_oids.count(soid) == 0)
- I'm running a script that reads and writes random objects using librados (creating a new pool once in a while). Runn...
11/25/2010
- 07:36 AM Revision 3ab60091 (ceph): osd: dump_missing: also dump missing_loc
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 07:35 AM Revision da087e47 (ceph): osd: discover_all_missing fix
- Don't request information from an OSD unless it is up and part of the
might_have_unfound set. Add more logging.
Sign... - 12:18 AM Bug #611: OSD: OSDMap::get_cluster_inst
- commit:da087e47c21190f9cbde4d24182b7dfe581cd069 should resolve this
11/24/2010
- 10:54 PM Bug #611: OSD: OSDMap::get_cluster_inst
- I'll take a look
- 10:18 PM Bug #611: OSD: OSDMap::get_cluster_inst
- Okay, I somehow commented/set this bug backwards with another one. Whoops, sorry guys!
This looks like the OSD is as... - 10:38 AM Bug #611: OSD: OSDMap::get_cluster_inst
- Sam said he'd look at this since it's in the background scrubbing bits that he and Josh did.
- 05:11 AM Bug #611 (Resolved): OSD: OSDMap::get_cluster_inst
- After upgrading to the latest unstable, one OSD crashed. Before the upgrade, 10 of the 12 OSD's were online.
When ... - 10:18 PM Bug #612: OSD: Crash during auto scrub
- Dunno how, but somehow commented/assigned this and another bug backwards. Meant to say:
Sam said he'd look at this s... - 10:38 AM Bug #612: OSD: Crash during auto scrub
- This looks like the OSD is assembling a list of missing queries and then sending them out without bothering to check ...
- 05:28 AM Bug #612 (Resolved): OSD: Crash during auto scrub
- After I saw #611 my cluster started to crash. One after the other, the OSD's started to go down, all with a message a...
- 10:09 PM Feature #453 (Resolved): osd: return error (instead of blocking) on lost objects
- It's passing the lost1 and lost2 unit tests now.
- 09:41 PM rgw Bug #353: Handle non-ascii filenames
- Yeah, I agree with Amazon's approach here. UTF-8 makes sense. I think we could continue to use std::string internally...
- 02:03 AM Revision d6e8e8d1 (ceph): gui: some cleanup
- Rather than vectors of pointers, use vectors of NodeInfo structures.
This avoids the problem of freeing the NodeInfo ... - 12:56 AM Revision 1b1e040e (ceph): osd: add a map for lingering messages
- 12:55 AM Revision 99e1e4de (ceph): librados: assert_version on sync operations
- 12:55 AM Revision c4b97953 (ceph): librados: last_objver is set on the pool, and not per thread
- 12:55 AM Revision 454ea06e (ceph): rbd: notify about header changes
- 12:55 AM Revision 520b523b (ceph): librados: fix unnecessary locking
- 12:55 AM Revision 4c8bdc53 (ceph): osd: don't notify notifier
- 12:54 AM Revision a76de3b2 (ceph): librados: complete C interface for watch/notify
- 12:54 AM Revision 38c8e383 (ceph): librados: rename cookie to handle in api
- 12:54 AM Revision 2954799a (ceph): librados: notify waits for completion
- 12:50 AM Revision e7184e6d (ceph): librados: start implementing watch/notify
- 12:50 AM Revision a4864bd8 (ceph): librados: enable object versioning
- 12:50 AM Revision f36677f8 (ceph): librados: update C api
- 12:49 AM Revision f8af4f2c (ceph): osd: add watch/notify timeout
- 12:49 AM Revision cc62f2eb (ceph): osd: fix bad mutex lock
- 12:49 AM Revision e0c548ad (ceph): osd: fix ms_handle_reset
- 12:49 AM Revision d5cc6732 (ceph): osd: some notify related cleanups
- 12:49 AM Revision 7272bfec (ceph): osd: send notify response from reset handler if needed
- 12:49 AM Revision d66b52e1 (ceph): osd: watch infrastructure
- third attempt
- 12:49 AM Revision 2b5e61ca (ceph): osd: send notification id
- 12:49 AM Revision 59e61d0e (ceph): osd: discard of disconnected watchers
- still need to add a timeout
- 12:49 AM Revision f5f33822 (ceph): osd: send notify reply if there are not watchers
- 12:49 AM Revision 9437ea84 (ceph): osd: add user_version field in obect_info_t
- 12:49 AM Revision 7bda45a1 (ceph): osd: reply with either user_version or at_version, depends on the op
- 12:49 AM Revision f7b7d67a (ceph): osd: check requested watch version number
- send appropriate status code if needed
- 12:47 AM Revision 2bce34e7 (ceph): osd: handle watch op, register client on object xattr
- 12:47 AM Revision 3110e361 (ceph): osd: basic watch/notify handling
- 12:47 AM Revision e493c7ae (ceph): osd: handle notify-ack
11/23/2010
- 11:39 PM Revision 2f13dd8e (ceph): gui: more reindenting
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 11:37 PM Revision 66a78c23 (ceph): gui: reindent a bunch of code
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 10:40 PM Revision d8652de6 (ceph): mdcache: in trim_non_auth, only print out path if it has a parent dentry.
- This should only occur with the root inode, but caused a segfault for
anybody running more than one MDS who restarted... - 10:04 PM Revision 8768b52d (ceph): mds: Reply checking_lock while reading filelock
- Use checking_lock to repalce lock_state in extra buffer list to let client can get correct file lock reply.
- 09:59 PM Revision 4041bf0d (ceph): mds: fix set_state_rejoin auth_pin check
- We carry an auth pin IFF !stable AND auth.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:59 PM Revision 5ed06ffc (ceph): client: remove inode from flush_caps list when auth_cap changes
- Avoid confusing other code (e.g. kick_flushing_caps) by staying on the mds
flushign_caps list when we don't even have... - 09:52 PM Revision 285cc946 (ceph): osd: fix is_all_uptodate()
- This should only return true when recovery is done, i.e., no more missing
objects. Nothing to do with unfound.
Sign... - 09:52 PM Revision 36f703e1 (ceph): osd: removing unused variable, fix warning
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:52 PM Revision 413ecb0b (ceph): osd: only search_for_missing if there are unfound objects
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:52 PM Revision 671b1c09 (ceph): osd: add get_num_unfound() helper
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:52 PM Revision 7ea7a435 (ceph): osd: only discover_all_missing if unfound
- Signed-off-by: Sage Weil <sage@newdream.net>
- 09:52 PM Revision 5452dae6 (ceph): osd: recover_primary() until primary has all found objects
- The logic in that if was effectively reversed.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:52 PM Revision 5498c467 (ceph): osd: fix recover_replicas() unfound check
- missing_loc.count(soid) == 0 only means unfound if it's not missing on the
primary.
Signed-off-by: Sage Weil <sage@n... - 09:52 PM Revision e97eae15 (ceph): init-ceph: tolerate failure in cleanallogs
- Otherwise /var/log/ceph/stat makes rm -f error out and we fail.
Signed-off-by: Sage Weil <sage@newdream.net> - 09:52 PM Revision 84612286 (ceph): Build might_have_unfound set at activation
- The might_have_unfound set is used by the primary OSD during recovery.
This set tracks the OSDs which might have unfo... - 09:52 PM Revision 0e15da8d (ceph): Rename peer_summary_requested to peer_backlog_req
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 09:52 PM Revision c0c301d5 (ceph): osd: PG::read_log: don't be clever with lost xattr
- Formerly, we had a special case in read_log for dealing with objects
whose objects were present on the disk, but not ... - 09:52 PM Revision 55570baf (ceph): osd: fix PG::is_all_uptodate
- In PG::is_all_uptodate, don't try to look for peer_missing[osd->whoami].
The primary keeps that in PG::missing!
Sign... - 08:26 PM Revision 36c6569c (ceph): monmaptool: Return a non-zero error code and print a useful error
- message if unable to read the monmap file.
Signed-off-by: Samuel Just <samuelj@hq.newdream.net> - 06:14 PM Feature #610 (Resolved): gui: make PG view prettier
- The ceph -g GUI should display PGs in a list, rather than as icons that have to be clicked on. We should get rid of t...
- 06:13 PM Bug #604 (Resolved): Compiler warning: 'status' may be used uninitialized in this function
- Fixed by commit:d6e8e8d15d22b51ec86bc5687336c3d50d9b3a5d
We should change PG view on the GUI to be a list view at ... - 05:43 PM Revision fc212548 (ceph): mds: allow for old fs's with stray instead of stray0
- New fs's get stray0, but we want to still behave with old ones.
Signed-off-by: Sage Weil <sage@newdream.net> - 05:37 PM Revision de61991a (ceph): Merge branch 'testing' into unstable
- Conflicts:
configure.ac - 03:00 PM Bug #531: Journaling Causes System Hang
- Awesome, thanks for the help. I will give these patches a shot towards the end of the week.
Thanks - 02:43 PM Bug #599 (Resolved): recover_master_log, doesn't
- There were two problems here:
1) we were restarting the osds before the monitors, which in this case prevented a f... - 02:01 PM Linux kernel client Bug #552: Samba with kernel oplocks=on produces lots of corrupt mds entries in dmesg
- Our friends at Tcloud just submitted patches for this today, which I've applied to the unstable branch of our kernel ...
- 11:46 AM CephFS Feature #593 (Rejected): mds: fsck: anchor table repair
- dup
- 11:42 AM Feature #609 (Resolved): osd: query pool/pg for objects with given xattr
- This will probably take the form of a pool class plugin?
It could start as just a hack, for now.
- 11:03 AM Bug #595: Autogen: not a literal
- This problem does not seem to occur using 2.68 on my local machine. Slider et al. seem to be using 2.67.
- 09:39 AM CephFS Bug #608 (Resolved): mds: MDCache::create_system_inode()
- this should be fixed by commit:fc212548aea1d7f001b56ba096a79ba54b8a92c3
Thanks! - 07:09 AM CephFS Bug #608 (Resolved): mds: MDCache::create_system_inode()
- On a small test cluster I saw that my MDS was not coming up after a fresh mkcephfs, this is what the log showed:
<... - 09:33 AM Tasks #584: do throughput scaling tests on sepia
- What was the variance in per-node throughput? Did we have one node dominating?
- 09:22 AM Tasks #584 (In Progress): do throughput scaling tests on sepia
- There's definitely a problem here; the total throughput should be scaling more or less linearly until we hit a bottle...
- 07:44 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- I'll have to rebuild, since I didn't look at the messages that closely.
- 07:02 AM Revision 868665d5 (ceph): v0.23.1
- 06:41 AM Revision c327c6a2 (ceph): mon: always use send_reply for auth replies
- Signed-off-by: Sage Weil <sage@newdream.net>
- 06:41 AM Revision 61dd4f03 (ceph): mon: simplify send_reply code
- No need to specify destination in send_reply, as we always have the request
for reference.
Simplify MRoute construct... - 01:37 AM Revision 2c71bd33 (ceph): osd: add assert to _process_pg_info
- When activating an inactive replica, assert that we are doing so based
on a message from the primary.
Signed-off-by:... - 01:35 AM Revision a70943fd (ceph): osd: re-indent some code in _process_pg_info
- Re-indent the code and add a comment.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net> - 12:12 AM Revision 71369541 (ceph): msgr: tolerate 0 bytes from tcp_read_nonblocking
- This can happen, I belive when we get a signal or something.
Signed-off-by: Sage Weil <sage@newdream.net> - 12:12 AM Revision 7ec0034b (ceph): init-ceph: fix (and test!) cleanlogs and cleanalllogs
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:03 AM Revision 7b4a801f (ceph): mds: fix rejoin_scour_survivor_replicas inode check
- We want to remove replicas that we don't ack, but those don't appear in
the strong_inode map; they're appended to the...
11/22/2010
- 11:08 PM Revision 8d95b5b6 (ceph): messenger: init rc to -1, removing compiler warning.
- This actually is initialized before all uses, but compilers tend to
have trouble with assignment in if-else branches,... - 11:08 PM Revision dd11fe27 (ceph): types: Allow inodeno_t structs to alias.
- This removes a compiler warning that appeared in a gcc upgrade and
is apparently erroneous, about its usage violating... - 10:56 PM Bug #540 (Resolved): CephxClientHandler::handle_response
- couldn't reproduce this, but fixed two smallish things that may have been responsible for this:
commit:61dd4f03e6e15... - 10:35 PM Linux kernel client Bug #552: Samba with kernel oplocks=on produces lots of corrupt mds entries in dmesg
- From the reply dump, it looks like a ceph_mds_reply_head, a length 0 tracebl, a length 1 extrabl (containing a u8 == ...
- 09:25 PM Revision ac6b018a (ceph): Causes the MDSes to switch among a set of stray directories when
- switching to a new journal segment.
MDSCache:
The stray member has been replaced with strays, an array of inodes
r... - 09:16 PM Revision 3f8f5905 (ceph): Timer must be initialized in Client::init and shutdown in
- Client::shutdown.
Signed-off-by: Samuel Just <samuelj@hq.newdream.net> - 06:47 PM Revision 8eb4de9e (ceph): generate_past_intervals:generate back to lastclean
- PG::generate_past_intervals needs to generate all the intervals back to
history.last_epoch_clean, rather than just to... - 06:07 PM Revision 80f28235 (ceph): vstart.sh: 'init-ceph stop' instead of 'stop.sh'
- This just makes it easier to run multiple vstart sessions as the same user
on the same host.
Signed-off-by: Sage Wei... - 05:55 PM Revision 53d0650a (ceph): Merge branch 'osd_msgr' into unstable
- 05:55 PM Revision cd53719f (ceph): mds: resolve cleanup
- Only track ambiguous imports and such if we get a resolve message while in
the resolve state.
Signed-off-by: Sage We... - 05:55 PM Revision c0c81d53 (ceph): mds: trim exported subtree _after_ adjusting auth
- We need to set the subtree bounds before trimming it away, or else we may
throw out things we're still auth for.
Sig... - 05:55 PM Revision 9e15ade8 (ceph): mds: do not eval subtree root when replay|resolve
- This is nonsensical. And can lead to scatter_writebehind, which breaks
horribly.
Signed-off-by: Sage Weil <sage@new... - 05:55 PM Revision 27c6f217 (ceph): mds: remove bogus assert
- Causes problems during resolve finish.
Signed-off-by: Sage Weil <sage@newdream.net> - 05:49 PM Revision 924b1fcb (ceph): osd: bind to new cluster address when wrongly marked down
- If we come back up on the same address, there is a possible race. Other
nodes will mark_down when they see us go dow... - 05:45 PM Revision 19409763 (ceph): msgr: implement rebind() to pick a new port
- Closes out all old connections and binds to a _different_ port. This
ensures that someone doing mark_down on our old... - 05:09 PM Revision f7170f95 (ceph): client: only encode_cap_releases once per request.
- Accomplish this by making a list of cap releases in the (permanent)
MetaRequest, and then copying that into the (pote... - 04:36 PM Bug #607 (Rejected): osd: ReplicatedPG: sub_op_modify: fix creation of ObjectState
- There's a part of the ReplicatedPG::sub_op_modify code that goes like this:
> // do op
> ObjectStat... - 04:29 PM CephFS Feature #91: mds: up:shadow mode
- Updated Journaler to make new interface options asynchronous.
Presently working on how to disambiguate between a one... - 03:48 PM Tasks #584 (Resolved): do throughput scaling tests on sepia
- Results of running rados -p bench bench 20 write on <Nodes>. <Average Throughput> is the average of the Bandwidth st...
- 01:24 PM CephFS Feature #88 (Resolved): mds: change stray commit strategy to avoid rolling stray dir commits
- commit:ac6b018acbeaf8670f8c268db164cfb8a12c171d
- 12:59 PM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Is the stack trace you're getting now identical, or different? The FileStore.cc change _should_ have avoided the asy...
- 09:28 AM Bug #563: osd: btrfs, warning at inode.c ( btrfs_orphan_commit_root )
- Just to update the issue, Sage asked me to change something in FileStore.cc, tried that for some days, but that didn'...
- 12:47 PM CephFS Feature #606 (Duplicate): mds: optionally store parent attr on file objects
- The goal is to be able to find files contained in rebuilt directories (#603). We can store the same attrs we do for ...
- 12:45 PM CephFS Feature #605 (Rejected): mds: verify/repair anchor table
- - Make sure every item we encounter while traversing the that is anchored correctly appears in the anchor table.
- M... - 12:44 PM Bug #604 (Resolved): Compiler warning: 'status' may be used uninitialized in this function
- In gui.cc
The warning's location references are a bit off, but the function gen_node_info_from_icons declares a "sta... - 12:43 PM CephFS Feature #603 (Resolved): mds: repair directory hierarchy
- The goals are
- rebuild missing/corrupt directories
- repair multiple primary links to directories
We'll do so... - 12:40 PM CephFS Feature #602 (Resolved): mds: handle corrupt/missing journals
- This probably means
- shutting down current instances, resetting cluster membership
- throwing out journals (or m... - 12:37 PM CephFS Feature #601 (New): mds: order directory commits after rename
- When we rename something between directories, we should try to commit the target directory _before_ the source direct...
- 12:34 PM CephFS Feature #600 (Resolved): mds: store full trace on directories
- Currently we only store the immediate parent; store a full trace up to the root. This is CInode::encode_parent_mutat...
- 12:17 PM Bug #599: recover_master_log, doesn't
- Also, I have verified that osd3 and osd9 did NOT crash. They're still running, and they did receive the messages from...
- 12:13 PM Bug #599 (Resolved): recover_master_log, doesn't
- This is another peering bug. We found it on wido's cluster. Basically, peering never completes.
I just examined PG... - 09:52 AM Bug #592 (Resolved): osd: rebind cluster_messenger when wrongly marked down
- commit:53d0650a42cbfd2f02db2c708a570b6d9e116bb4
- 09:14 AM CephFS Bug #596 (Resolved): crash during mds reconnect
- Well, that seems to fix it. I added a releases vector to the MetaReqest so it will only encode the releases once, and...
- 08:49 AM Bug #598 (Resolved): osd: journal reset in parallel mode acts weird
- from ML:...
- 04:52 AM Revision 51abcaa2 (ceph): mon: clean up cluster_addr code a bit, better debug output
- Signed-off-by: Sage Weil <sage@newdream.net>
- 04:52 AM Revision 20313644 (ceph): osdmap: fix cluster_addr encoding; printing
- The cluster addrs were getting lost because we were checking v instead of
ev.
Signed-off-by: Sage Weil <sage@newdrea... - 04:52 AM Revision 28498a00 (ceph): osd: send correct ip addrs to monitor for cluster_, hb_addr
- Signed-off-by: Sage Weil <sage@newdream.net>
- 03:59 AM Revision ec434eda (ceph): osd: unconditionally set up separate msgr instance for osd<->osd msgs
- Always set up cluster_messenger (before we would only do so if there was
an explicit address configured for it). The... - 12:16 AM Revision 0dddf453 (ceph): filestore: only warn about disk write cache on kernels <2.6.33
- Signed-off-by: Sage Weil <sage@newdream.net>
- 12:15 AM Revision 0856f57e (ceph): osd: fix search_for_missing: old last_update implies object not present
- For example, if an osd sends an empty PG::Info (last_update = 0'0) and
empty missing, we should not conclude that the... - 12:09 AM Revision 6ef5c2f3 (ceph): init-ceph: fix cleanlogs for no log_sym_dir case
- Signed-off-by: Sage Weil <sage@newdream.net>
11/21/2010
- 07:55 PM Linux kernel client Bug #549 (Resolved): bonnie++ file stat failure
- commit:3105c19c450ac7c18ab28c19d364b588767261b3
- 03:50 PM Bug #592: osd: rebind cluster_messenger when wrongly marked down
- I think the cleanest solution here is to re-bind the cluster_messenger to a new port when we are marked down and go b...
- 03:38 PM Linux kernel client Bug #597 (Closed): Reproducible crash mounting multiple directories from a pool
- This bug was fixed in v2.6.36, commit:ca04d9c3ec721e474f00992efc1b1afb625507f5. Thanks for the report though! :)
- 03:34 PM Linux kernel client Bug #597: Reproducible crash mounting multiple directories from a pool
- Should have mentioned - this is with the Ubuntu 10.10 desktop kernel, which is 2.6.35-22, I think.
- 03:33 PM Linux kernel client Bug #597 (Closed): Reproducible crash mounting multiple directories from a pool
- When trying to mount a pool multiple times (with different subdirectories) I get a consistent system hang.
Steps t...
11/20/2010
- 05:06 PM Bug #531: Journaling Causes System Hang
- Please try out the patches in the filestore_throttle branch, commit:b28c0bf82ac28ded4fe85573d32fdc111c66e50b
It lo... - 03:15 AM Revision fc9b0976 (ceph): OSDMap: const cleanup
- Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
- 03:14 AM Revision 2a5c3893 (ceph): mds-dumper: Define Dumper::~Dumper()
- To fix compile error.
Signed-off-by: Colin McCabe <colinm@hq.newdream.net>
Also available in: Atom