]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Tue, 19 May 2009 19:43:04 +0000 (12:43 -0700)]
kclient: kill dput/d_drop debug cruft
Sage Weil [Tue, 19 May 2009 19:26:12 +0000 (12:26 -0700)]
kclient: remove bookkeeper
Sage Weil [Tue, 19 May 2009 18:23:35 +0000 (11:23 -0700)]
kclient: fix kbuild.patch for 2.6.30
Sage Weil [Tue, 19 May 2009 18:17:11 +0000 (11:17 -0700)]
changelog
Sage Weil [Tue, 19 May 2009 17:18:29 +0000 (10:18 -0700)]
todo'd
Sage Weil [Tue, 19 May 2009 17:18:16 +0000 (10:18 -0700)]
initscript: fix instance check
Sage Weil [Tue, 19 May 2009 16:58:15 +0000 (09:58 -0700)]
mon: send mount error to client on EPERM or protocol version mismatch
Sage Weil [Tue, 19 May 2009 16:57:53 +0000 (09:57 -0700)]
kclient: fix mount err reporting
No more delay
Sage Weil [Mon, 18 May 2009 23:51:29 +0000 (16:51 -0700)]
filer: make probe recover size and/or mtime; fix some bugs
This fixes up MDS recovery to find the file size. And makes
probe actually work.. it was pretty broken before.
Sage Weil [Mon, 18 May 2009 22:47:36 +0000 (15:47 -0700)]
osd: cleanly implement OSD_OP_STAT
Put stat result (size AND mtime) in read result bufferlist.
Sage Weil [Mon, 18 May 2009 21:48:54 +0000 (14:48 -0700)]
uclient: fix missing cap behavior
Ignore cap messages if we don't have the inode or cap.
Sage Weil [Mon, 18 May 2009 20:21:26 +0000 (13:21 -0700)]
kclient: style fixes
Sage Weil [Mon, 18 May 2009 20:21:58 +0000 (13:21 -0700)]
kclient: strip trailing whitespace
Sage Weil [Mon, 18 May 2009 19:01:21 +0000 (12:01 -0700)]
initscript: make sure cmdline instance exists
Sage Weil [Mon, 18 May 2009 18:28:49 +0000 (11:28 -0700)]
kclient: make bookkeeper compile when not in .config
Sage Weil [Mon, 18 May 2009 17:47:35 +0000 (10:47 -0700)]
osd: clean up output
Sage Weil [Mon, 18 May 2009 18:22:24 +0000 (11:22 -0700)]
mds: drop mut locks after file_recover_cow
Sage Weil [Mon, 18 May 2009 18:19:29 +0000 (11:19 -0700)]
mds: fix caps issued on snapped inodes
Do not adjust Capability on snapped inodes (otherwise cap state
gets out of sync with client).
Yehuda Sadeh [Mon, 18 May 2009 18:11:22 +0000 (11:11 -0700)]
kclient: keep page locked after page_mkwrite call
Yehuda Sadeh [Mon, 18 May 2009 18:03:10 +0000 (11:03 -0700)]
kclient: disable bookkeeper when including some kernel headers
Sage Weil [Thu, 14 May 2009 23:33:48 +0000 (16:33 -0700)]
mds: kill fugly arg to rdlock_path_pin_ref
The caller (readdir, in this case) can do it now, since it now does the
acquire_locks() call.
Sage Weil [Thu, 14 May 2009 23:31:29 +0000 (16:31 -0700)]
mds: rdlock filelock during readdir
This ensures any uncommitted creates/whatevers in the directory flush
(and we see them!).
Sage Weil [Thu, 14 May 2009 23:30:51 +0000 (16:30 -0700)]
mds: no rdlock in filelock LOCK state
Otherwise we get wrlocks AND rdlocks at the same time, which is clearly
problematic. Der.
Also fix _rdlock_kick to not simple_lock, since that won't help us.
Sage Weil [Thu, 14 May 2009 23:10:27 +0000 (16:10 -0700)]
mds: drain wrlocks before going from LOCK->SYNC in file_eval
This avoids excessive waits for the journal to flush on lock->sync
when a client request holds a wrlock. There's no reason to hurry.. if
someone needs it sync we can to the transition then; otherwise, it'll
happen when the wrlock is dropped.
Sage Weil [Thu, 14 May 2009 21:23:40 +0000 (14:23 -0700)]
.gitignore update
Sage Weil [Thu, 14 May 2009 20:57:25 +0000 (13:57 -0700)]
mds: break CAP_RDCACHE into CAP_SHARED, CAP_CACHE
FILE_CAP_RDCACHE was being used to mean both read access to the
file attributes (size, mtime) and permission to retain cached
data. That lead to an incorrect definition of the filelock in the
mds, and in turn bugs with multiple client access. These are now
CAP_*_SHARED (all locks) and CAP_FILE_CACHE (filelock only).
The main observed symptom was a client creating files in a
directory and a second client not seeing them, due to RDCACHE not
being revoked and rdcache_gen thus not incrementing, allowing a
dcache readdir to proceed.
Sage Weil [Thu, 14 May 2009 20:47:31 +0000 (13:47 -0700)]
kclient: don't skip EXPIREABLE caps
EXPIREABLE is obsolete. We need to reconnect _all_ caps!
Sage Weil [Thu, 14 May 2009 00:02:44 +0000 (17:02 -0700)]
kclient: fix crush decoding for recent changes
Sage Weil [Wed, 13 May 2009 23:49:05 +0000 (16:49 -0700)]
rev osd protocol, disk format to reflect crush changes
Sage Weil [Wed, 13 May 2009 23:46:20 +0000 (16:46 -0700)]
crush: fix crush_perm_choose; optimize r=0 case.
This was misbehaving for x=0, among other things.
Avoid filling in perm array for the initial (p)r=0 call. We only
need to do a full permutation for subsequent r.
Sage Weil [Wed, 13 May 2009 21:56:28 +0000 (14:56 -0700)]
psim: count result set sizes
Sage Weil [Wed, 13 May 2009 21:55:07 +0000 (14:55 -0700)]
crush: fall back to exhaustive bucket search for any bucket type
If we don't get a bucket-specific choice in 5 tries, do an
exhaustive search (based on a random permutation). Only then give
up on the bucket and retry descent.
Note that the search-based fallback does not honor weighting at
all.
Sage Weil [Wed, 13 May 2009 21:07:45 +0000 (14:07 -0700)]
crush: ditch prime number theorem; generate random permutation on the fly
Sage Weil [Wed, 13 May 2009 19:02:45 +0000 (12:02 -0700)]
crush: improve uniform selection a bit
Shift to a new prime (and thus permutation) every few r.
Sage Weil [Mon, 11 May 2009 21:33:17 +0000 (14:33 -0700)]
mds todo
Sage Weil [Mon, 11 May 2009 21:00:52 +0000 (14:00 -0700)]
kclient: fail connection when s_addr==0 and port/nonce don't match
Even if s_addr is 0, the port and nonce should match. We were
previously going ahead with the connection when we shouldn't have
been.
Sage Weil [Mon, 11 May 2009 20:54:03 +0000 (13:54 -0700)]
mds: maintain capid across mds restart and client reconnect
Sage Weil [Mon, 11 May 2009 20:21:26 +0000 (13:21 -0700)]
mon: don't replace standby mds
Sage Weil [Mon, 11 May 2009 20:05:38 +0000 (13:05 -0700)]
mds: handle MMonMap
Sage Weil [Mon, 11 May 2009 18:51:21 +0000 (11:51 -0700)]
journaler: fix replay
Broken by commit
497ade3b90 .
Use objecter to write header (not just to read it).
Also reset the prefetch values each time the layout is set.
Sage Weil [Mon, 11 May 2009 18:34:37 +0000 (11:34 -0700)]
mds: fix loner drops
This was broken way back by commit
0c3becdf .
Only drop loner in eval().
Sage Weil [Mon, 11 May 2009 18:15:44 +0000 (11:15 -0700)]
rev protocols
Just for good measure, since I missed a few changes earlier.
Sage Weil [Mon, 11 May 2009 18:01:50 +0000 (11:01 -0700)]
debian: new rules file. don't strip.
Sage Weil [Mon, 11 May 2009 17:44:20 +0000 (10:44 -0700)]
osd: fix accounting .snap thing
Sage Weil [Sat, 9 May 2009 18:25:27 +0000 (11:25 -0700)]
conf: improved sample
Sage Weil [Fri, 8 May 2009 20:37:06 +0000 (13:37 -0700)]
mon: check for osd exists before up/down
Sage Weil [Fri, 8 May 2009 21:04:03 +0000 (14:04 -0700)]
osd: maintain up_epoch AND boot_epoch; revise OSDSuperblock accordingly
In order to make the superblock clean interval meaningful after we
are marked down and then up again (over the life of a single
cosd process insance), we track both boot_epoch and up_epoch,
and keep [boot_epoch,clean_thru] in the superblock.
This avoids seeing crashed pgs when and osd is wrongly marked down
and the osd marks itself up again.
Sage Weil [Fri, 8 May 2009 19:55:37 +0000 (12:55 -0700)]
osd: adjust heartbeat peer lock
Need it to protect heartbeat_from_stamp.
Sage Weil [Fri, 8 May 2009 19:55:16 +0000 (12:55 -0700)]
osd: reset heartbeat peer set on osd down
This clears out the timers.
Sage Weil [Fri, 8 May 2009 04:42:19 +0000 (21:42 -0700)]
osd: adjust snap collection memberships during pg split
Sage Weil [Fri, 8 May 2009 04:38:26 +0000 (21:38 -0700)]
osd: adjust pg_stats during pg split
Sage Weil [Fri, 8 May 2009 04:32:53 +0000 (21:32 -0700)]
osd: generate correct child pg when doing pg split
Sage Weil [Fri, 8 May 2009 04:13:39 +0000 (21:13 -0700)]
ceph: don't choke on unexpected MMonMap
Sage Weil [Thu, 7 May 2009 21:39:47 +0000 (14:39 -0700)]
kclient: recalculate pgid each time request is sent
The pg calculation depends on osdmap parameters that are transient. In
contrast, the rest of calc_layout is concerned with file striping, which
is fixed (at least over the lifetime of the request).
Sage Weil [Thu, 7 May 2009 21:14:29 +0000 (14:14 -0700)]
mds: include CAP_PIN in caps_liked()
Among other things, this prevents confusion in the uclient.
Sage Weil [Thu, 7 May 2009 20:56:14 +0000 (13:56 -0700)]
osd: fix bug in recovery ops accounting
Adjust osd global counter too!
Yehuda Sadeh [Thu, 7 May 2009 17:47:32 +0000 (10:47 -0700)]
kclient: a few sparse fixes
Sage Weil [Wed, 29 Apr 2009 23:50:02 +0000 (16:50 -0700)]
mds: only issue sensible caps on directories
Sage Weil [Wed, 29 Apr 2009 23:27:24 +0000 (16:27 -0700)]
mds: make eval take a mask; kill try_*_eval
This will eventually help us deal with all locks as a group.
Also, simplify the eval_subtree_root bits.
Sage Weil [Fri, 1 May 2009 14:27:32 +0000 (07:27 -0700)]
start v0.8
Sage Weil [Fri, 1 May 2009 14:11:46 +0000 (07:11 -0700)]
msgr: kill static instance 'rank' of SimpleMessenger
Sage Weil [Fri, 1 May 2009 13:55:54 +0000 (06:55 -0700)]
msgr: clean up refs to static 'rank'
Sage Weil [Thu, 30 Apr 2009 22:24:43 +0000 (15:24 -0700)]
uclient: clean up mount wrt osdmap, mdsmap
Sage Weil [Thu, 30 Apr 2009 18:14:31 +0000 (11:14 -0700)]
journaler: store layout in journal header
This means we won't fall over if the 'default' layout changes. Specify
the layout during journal creation, as you'd expect.
Sage Weil [Thu, 30 Apr 2009 17:42:26 +0000 (10:42 -0700)]
mds: clean up layouts
Sage Weil [Thu, 30 Apr 2009 16:44:39 +0000 (09:44 -0700)]
osd: more pg_pool, pg changes
Remove 'type' from pg_t. It's implied based on the pool.
Clean out file_layout. The pool implies a type and namespace.
Sage Weil [Thu, 30 Apr 2009 14:12:07 +0000 (07:12 -0700)]
kclient: fix page_mkwrite for 2.6.30
This will change again soon when Nick's other fix gets merged. Also, we
probably need to stop doing write_begin/end due to some differences in
behavior (e.g., no extending file size? see btrfs as a reference).
Sage Weil [Wed, 29 Apr 2009 22:37:39 +0000 (15:37 -0700)]
osd: typo
Sage Weil [Wed, 29 Apr 2009 19:50:07 +0000 (12:50 -0700)]
config: fix pg pools in default file layouts
Sage Weil [Wed, 29 Apr 2009 22:33:31 +0000 (15:33 -0700)]
osd: fix up recovery op accounting
Fix varous rop accounting bugs. Add assertions. Log recovery
ops.
Sage Weil [Wed, 29 Apr 2009 19:06:42 +0000 (12:06 -0700)]
todo
Sage Weil [Wed, 29 Apr 2009 18:47:13 +0000 (11:47 -0700)]
kclient: retry 2x on ESTALE; fix up mds reply error paths
Also fix error paths, which weren't handling the snap_rwsem
properly.
Sage Weil [Wed, 29 Apr 2009 18:44:03 +0000 (11:44 -0700)]
kclient: fix symbol names without 'ceph_' prefix
Yehuda Sadeh [Wed, 29 Apr 2009 19:00:08 +0000 (12:00 -0700)]
kclient: add missing parameter
Yehuda Sadeh [Wed, 29 Apr 2009 18:41:44 +0000 (11:41 -0700)]
conf: added missing allocations
Sage Weil [Wed, 29 Apr 2009 18:38:11 +0000 (11:38 -0700)]
mon: track per-pg_pool usage
Sage Weil [Wed, 29 Apr 2009 18:13:13 +0000 (11:13 -0700)]
config: make uninit var warning go away
Sage Weil [Wed, 29 Apr 2009 18:13:01 +0000 (11:13 -0700)]
osd: push pg_num etc into pg_pool
Sage Weil [Tue, 28 Apr 2009 23:44:12 +0000 (16:44 -0700)]
osd: clean up assemble_backlog
Yehuda Sadeh [Tue, 28 Apr 2009 23:45:10 +0000 (16:45 -0700)]
kclient: don't allow to pass max file size
Sage Weil [Tue, 28 Apr 2009 23:30:34 +0000 (16:30 -0700)]
todo
Sage Weil [Tue, 28 Apr 2009 23:30:12 +0000 (16:30 -0700)]
osd: make osd mark itself back up if it is marked down
We may still want a way to tell an osd to exit.
Sage Weil [Tue, 28 Apr 2009 23:29:15 +0000 (16:29 -0700)]
mon: optionally dump specific osdmap epoch
Sage Weil [Tue, 28 Apr 2009 21:08:57 +0000 (14:08 -0700)]
uclient: update dir inode, cap from reply
Sage Weil [Tue, 28 Apr 2009 20:42:49 +0000 (13:42 -0700)]
vstart: --nodaemon to run via crun and -f (so that gprof works)
Sage Weil [Tue, 28 Apr 2009 20:22:29 +0000 (13:22 -0700)]
uclient: take inode ref when dirtying inode
Sage Weil [Tue, 28 Apr 2009 19:12:05 +0000 (12:12 -0700)]
initscript: default btrfs mount options notreelog,flushoncommit
Sage Weil [Mon, 27 Apr 2009 20:54:23 +0000 (13:54 -0700)]
cosd: use \ in dev list in .conf
Sage Weil [Mon, 27 Apr 2009 20:54:10 +0000 (13:54 -0700)]
osd: fix merge_log split point
Look at full eversion, not just .version (part 2).
Sage Weil [Tue, 28 Apr 2009 20:10:50 +0000 (13:10 -0700)]
mon: order log entries as they're committed
This doesn't fully order them (across commits), but it helps.
Sage Weil [Tue, 28 Apr 2009 05:37:09 +0000 (22:37 -0700)]
mon: adjust log levels
Sage Weil [Tue, 28 Apr 2009 05:34:52 +0000 (22:34 -0700)]
todo
Sage Weil [Tue, 28 Apr 2009 19:54:02 +0000 (12:54 -0700)]
assert: turn off tls in assert macro
About 3% cpu time
Sage Weil [Tue, 28 Apr 2009 05:16:35 +0000 (22:16 -0700)]
kclient: mark caps sync in check_caps writeback path too
We are already doing so in try_flush_caps().
Sage Weil [Tue, 28 Apr 2009 05:01:52 +0000 (22:01 -0700)]
kclient: added missing include
Sage Weil [Mon, 27 Apr 2009 23:32:43 +0000 (16:32 -0700)]
todo
Sage Weil [Mon, 27 Apr 2009 23:31:50 +0000 (16:31 -0700)]
osd: make pg pools resizeable
Move the pg size from the pgid to the pg_pool descriptor, so that
entire pools of pgs can be resized.
Sage Weil [Mon, 27 Apr 2009 20:35:21 +0000 (13:35 -0700)]
todo
Sage Weil [Mon, 27 Apr 2009 20:32:50 +0000 (13:32 -0700)]
config: -v or --version to print version info
Sage Weil [Mon, 27 Apr 2009 19:13:54 +0000 (12:13 -0700)]
mon: make down->out osd occur despite mon restarts
Also add a bit more sanity checking to make sure we don't do the
wrong thing.