]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Thu, 20 Nov 2008 22:40:08 +0000 (14:40 -0800)]
osd: clean up mkfs vs peek_super
Sage Weil [Thu, 20 Nov 2008 22:00:33 +0000 (14:00 -0800)]
osd: fix peek_whoami
Sage Weil [Thu, 20 Nov 2008 21:54:44 +0000 (13:54 -0800)]
osd: make peek_whoami verify fsid
Sage Weil [Thu, 20 Nov 2008 21:54:34 +0000 (13:54 -0800)]
osd: decode superblock properly
Sage Weil [Thu, 20 Nov 2008 21:42:06 +0000 (13:42 -0800)]
vstart: 3 osds
Sage Weil [Thu, 20 Nov 2008 21:31:13 +0000 (13:31 -0800)]
cstring: pre-terminate even if content unspecified
Sage Weil [Wed, 19 Nov 2008 19:19:28 +0000 (11:19 -0800)]
osd: feed type name into workqueue
Sage Weil [Wed, 19 Nov 2008 19:14:53 +0000 (11:14 -0800)]
osd: convert snap trimming to snap_trim_wq
Sage Weil [Wed, 19 Nov 2008 19:04:37 +0000 (11:04 -0800)]
osd: basic scrub works
Sage Weil [Wed, 19 Nov 2008 19:04:23 +0000 (11:04 -0800)]
osd: update stats when op is applied
Sage Weil [Wed, 19 Nov 2008 00:46:52 +0000 (16:46 -0800)]
osd: add scrub wq
Sage Weil [Thu, 20 Nov 2008 21:22:37 +0000 (13:22 -0800)]
os: use nstring instead of string for attrsets
Sage Weil [Thu, 20 Nov 2008 20:44:37 +0000 (12:44 -0800)]
ebofs: fix occasional bdev shutdown hang
Sage Weil [Thu, 20 Nov 2008 19:54:09 +0000 (11:54 -0800)]
osd: adjust merge_log
Object should only be marked missing if new entry is newer. If
they are the same, it may or may not be missing (depending on
whether it was before merge_log).
Sage Weil [Thu, 20 Nov 2008 19:31:12 +0000 (11:31 -0800)]
msgr: ref count message while they are owned by the messenger
Users still assume they hold the only reference, at least until
they call send_message.
One caveat is that ms_handle_failure is passed a message with an
unknown number of refs. The method should not try to free or
re-use the message.
Sage Weil [Thu, 20 Nov 2008 18:36:19 +0000 (10:36 -0800)]
msgr: reference count messenger
We want an explicit destroy() method, because the SimpleMessenger
needs to join the dispatch thread, and that can't happen just on
the last reference drop because that may happen in the dispatch
thread itself.
Sage Weil [Thu, 20 Nov 2008 18:32:42 +0000 (10:32 -0800)]
vstart: launch valgrind with --valgrind
just for cosd atm
Sage Weil [Thu, 20 Nov 2008 17:52:33 +0000 (09:52 -0800)]
msgr todo
Sage Weil [Thu, 20 Nov 2008 17:52:22 +0000 (09:52 -0800)]
Makefile: adjust link order (libcommon first _and_ last)
Sage Weil [Wed, 19 Nov 2008 19:00:25 +0000 (11:00 -0800)]
ebofs: add new objects to main collection
Sage Weil [Thu, 20 Nov 2008 16:15:13 +0000 (08:15 -0800)]
mds: remove session from xlist before deleting
Sage Weil [Thu, 20 Nov 2008 16:12:45 +0000 (08:12 -0800)]
client: remove xlist items before deleting
Sage Weil [Thu, 20 Nov 2008 16:09:27 +0000 (08:09 -0800)]
mds: remove Capability from session list before deleting
Sage Weil [Thu, 20 Nov 2008 06:01:14 +0000 (22:01 -0800)]
mds: pull scatterlock of xlist in destructor
Really, this should happen sooner, but for now this is equivalent to the
old xlist::item destructor.
Sage Weil [Thu, 20 Nov 2008 05:50:39 +0000 (21:50 -0800)]
os: clean up ObjectStore::Transaction interface
Also, fix attrset * thing.. that doesn't look safe!
Sage Weil [Thu, 20 Nov 2008 05:48:13 +0000 (21:48 -0800)]
msg: initialize footer
Sage Weil [Thu, 20 Nov 2008 05:41:26 +0000 (21:41 -0800)]
lockdep: turn lockdep off during shutdown
We can't function after the static items in lockdep.cc destruct. Disable
lockdep before that happens.
Sage Weil [Thu, 20 Nov 2008 05:34:00 +0000 (21:34 -0800)]
osd: fix osd_reqid_t hash
blobhash is only safe on packed types.
Sage Weil [Thu, 20 Nov 2008 05:33:32 +0000 (21:33 -0800)]
filestore: pad with zeroed buffer
Shut up valgrind
Sage Weil [Thu, 20 Nov 2008 03:43:58 +0000 (19:43 -0800)]
xlist: enforce removal from xlist
We want to ensure that removal takes place in the correct locking context,
not whatever context the ::item is destroyed in.
Sage Weil [Thu, 20 Nov 2008 00:47:12 +0000 (16:47 -0800)]
mds: fix uninit value
Sage Weil [Thu, 20 Nov 2008 00:38:27 +0000 (16:38 -0800)]
osd: adjust missing in merge_old_entry
Our "old" entry may have been newer, and missing.. remove from missing, and re-add
"new" entry to ensure missing reflects the correct object version.
Sage Weil [Wed, 19 Nov 2008 18:30:10 +0000 (10:30 -0800)]
osd: type cleanup
Sage Weil [Thu, 20 Nov 2008 00:28:09 +0000 (16:28 -0800)]
osd: more merge_log updates
Sage Weil [Thu, 20 Nov 2008 00:28:49 +0000 (16:28 -0800)]
monclient: dont free messenger until races there are fixed
Sage Weil [Wed, 19 Nov 2008 21:12:57 +0000 (13:12 -0800)]
cmonctl: fix busy loop
Sage Weil [Wed, 19 Nov 2008 21:09:38 +0000 (13:09 -0800)]
mon: ignore 0-byte latest
Sage Weil [Wed, 19 Nov 2008 18:31:31 +0000 (10:31 -0800)]
osd: merge_log fix when logs abut but do not overlap
Yehuda Sadeh [Wed, 19 Nov 2008 00:54:21 +0000 (16:54 -0800)]
kclient: different handling of EIO, bad crc
Sage Weil [Wed, 19 Nov 2008 00:47:36 +0000 (16:47 -0800)]
dstart: smaller cluster
Sage Weil [Tue, 18 Nov 2008 23:03:27 +0000 (15:03 -0800)]
osd: track bytes/kb usage over clones too
Sage Weil [Tue, 18 Nov 2008 21:57:32 +0000 (13:57 -0800)]
osd: track per-pg bytes, kb utilization
For non-clones only.
Sage Weil [Tue, 18 Nov 2008 21:53:55 +0000 (13:53 -0800)]
osd: fix update_stats
Sage Weil [Tue, 18 Nov 2008 21:53:33 +0000 (13:53 -0800)]
osd: verify monmap.fsid matches superblock on startup
Sage Weil [Tue, 18 Nov 2008 21:53:13 +0000 (13:53 -0800)]
osd: include fsid in OSDPing, and verify
Sage Weil [Tue, 18 Nov 2008 05:16:33 +0000 (21:16 -0800)]
osd: maintain some per-pg stats (object counts, for now)
We'll add disk usage soon.
Sage Weil [Tue, 18 Nov 2008 04:42:38 +0000 (20:42 -0800)]
dstart: larger cluster
Sage Weil [Tue, 18 Nov 2008 23:52:54 +0000 (15:52 -0800)]
msgr: fix reconnect after error
Items in sent queue weren't being moved back to out queue, and
in_seq/out_seq weren't being set properly after an incoming
connection replaced an existing connection.
Yehuda Sadeh [Tue, 18 Nov 2008 18:09:07 +0000 (10:09 -0800)]
mount: typo
Sage Weil [Tue, 18 Nov 2008 04:37:05 +0000 (20:37 -0800)]
mds: add new directory to new_dirfrags list
This ensures the directory gets committed before the mkdir event
is trimmed from the journal. Fixes the failed assertion in
CDir::_fetched seen on mds recovery (due to a missing directory object).
Sage Weil [Tue, 18 Nov 2008 00:46:38 +0000 (16:46 -0800)]
mds: unqueue recovery on purging inodes
If an inode is queued for file size recovery when it is purged,
unqueue it. This catches the log replay case where client
reconnect queues up the inode.
Also, in eval_stray, skip inodes that are queued. This should
avoid a recovery running concurrently with the purge (which could
be problematic, as it would carry a pointer to *in).
Sage Weil [Tue, 18 Nov 2008 00:03:47 +0000 (16:03 -0800)]
Yehuda Sadeh [Mon, 17 Nov 2008 23:37:32 +0000 (15:37 -0800)]
kclient: limit incoming messages size
Yehuda Sadeh [Mon, 17 Nov 2008 23:37:05 +0000 (15:37 -0800)]
mount: remove extra log
Yehuda Sadeh [Mon, 17 Nov 2008 23:32:56 +0000 (15:32 -0800)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Mon, 17 Nov 2008 23:32:07 +0000 (15:32 -0800)]
created mount helper, can resolve mount addresses
Sage Weil [Mon, 17 Nov 2008 22:42:48 +0000 (14:42 -0800)]
osd: rewrite merge_log
Sage Weil [Mon, 17 Nov 2008 22:21:52 +0000 (14:21 -0800)]
cmonctl: include original command in command reply to avoid weirdness
Sage Weil [Mon, 17 Nov 2008 22:08:30 +0000 (14:08 -0800)]
osd: fix build_prior any_up logic again
last_epoch_started may be anywhere in the interval.
Sage Weil [Mon, 17 Nov 2008 22:00:23 +0000 (14:00 -0800)]
filestore: fix btrfs detection based on latest btrfs-unstable
Look for working CLONE_RANGE ioctl.
Sage Weil [Mon, 17 Nov 2008 21:59:55 +0000 (13:59 -0800)]
todos
Sage Weil [Mon, 17 Nov 2008 20:57:11 +0000 (12:57 -0800)]
cmonctl: pick new mon on timeout
Sage Weil [Mon, 17 Nov 2008 20:54:53 +0000 (12:54 -0800)]
mon: fix get_latest
Sage Weil [Mon, 17 Nov 2008 22:42:30 +0000 (14:42 -0800)]
msg: non-destructively copy data buffers in set_data()
Sage Weil [Mon, 17 Nov 2008 21:23:03 +0000 (13:23 -0800)]
mds: use last_sent (not last_open) to untangle cap release races
If we use last_open, the client has to be smart about ignoring
MDS revocations after it sends a release request. (Or, the MDS has
to somehow know the ack is for an old cap.) Instead, just
serialize release over all cap messages sent to the client. It may
make for a slightly chattier cap release in some cases, but those
cases should be very rare, and this is simpler.
Sage Weil [Sat, 15 Nov 2008 00:54:32 +0000 (16:54 -0800)]
mds: be more forgiving on EPurgeFinish
Inode may not be in cache because of purge_stray() avoiding
journaling it's intent to purge. If that changes down the line,
add the assertion back.
Sage Weil [Mon, 17 Nov 2008 18:47:30 +0000 (10:47 -0800)]
osd: adjust build_prior any_up logic
We mark the pg 'down' unless there is at least one osd alive specifically
from the last epoch started.
Yehuda Sadeh [Mon, 17 Nov 2008 18:38:16 +0000 (10:38 -0800)]
kclient: silence down some warning
Yehuda Sadeh [Mon, 17 Nov 2008 18:37:30 +0000 (10:37 -0800)]
mds: fix an erroneous assertion (sage)
Sage Weil [Mon, 17 Nov 2008 17:15:44 +0000 (09:15 -0800)]
osd todos
Sage Weil [Mon, 17 Nov 2008 17:13:08 +0000 (09:13 -0800)]
osd: fix deadlock on map_lock vs peer_stat_lock
Sage Weil [Mon, 17 Nov 2008 17:03:47 +0000 (09:03 -0800)]
osd: remember past intervals instead of recalculating each time
This _vastly_ improves the speed of build_prior (and thus activate_map).
There is no need to recalculate this information each time as it is fully
dependent on _old_ OSDMaps, not current cluster state.
Sage Weil [Mon, 17 Nov 2008 17:01:30 +0000 (09:01 -0800)]
msgr: adjust mark_down locking to avoid possible race
Sage Weil [Mon, 17 Nov 2008 17:01:03 +0000 (09:01 -0800)]
cmonctl: reprobe every second
Sage Weil [Sat, 15 Nov 2008 05:14:13 +0000 (21:14 -0800)]
osd: clear_map_cache at end of activate_map
after we're done with it
Sage Weil [Sat, 15 Nov 2008 04:50:35 +0000 (20:50 -0800)]
osd: introduce map_lock RWLock, take read lock during heartbeat
This prevents a race between handle_osd_map updating the map while
heartbeat() is using it to ping peers.
Currently we take a write ref over the entirely to handle_osd_map; we may
be able to push that down a bit.
Sage Weil [Sat, 15 Nov 2008 00:58:37 +0000 (16:58 -0800)]
msgr: small cleanup
Sage Weil [Sat, 15 Nov 2008 00:58:17 +0000 (16:58 -0800)]
lockdep: force backtraces on specific mutexes
Maintaining backtraces is expensive to do for every acquisition. Make a
per-mutex flag so that specific deadlocks can be tracked down.
Sage Weil [Sat, 15 Nov 2008 00:56:59 +0000 (16:56 -0800)]
osd: maintain a cache of past osd maps during repeering
It's expensive and stupid to load and reparse them for each PG.
Sage Weil [Sat, 15 Nov 2008 00:09:09 +0000 (16:09 -0800)]
osd: pause/unpause recovery thread while processing map
Otherwise bad things happen (everyone assumes *osdmap is static and
readable).
Sage Weil [Sat, 15 Nov 2008 00:31:53 +0000 (16:31 -0800)]
mds: journal updates _after_ predirty_parents (which adds parent context)
This ensures the dirlumps occur in an order that can be replayed
to reconstruct the hierarchy (ancestors first).
Sage Weil [Sat, 15 Nov 2008 00:02:15 +0000 (16:02 -0800)]
mds: remove bad assertion
Inode may still be dirty. bah.
Sage Weil [Fri, 14 Nov 2008 23:52:58 +0000 (15:52 -0800)]
mds: mark inode clean only when purge is complete
Otherwise we confuse CDir dirty vs commit rules.
Sage Weil [Fri, 14 Nov 2008 23:39:00 +0000 (15:39 -0800)]
mds: only mark clean if dirty
Sage Weil [Fri, 14 Nov 2008 23:31:07 +0000 (15:31 -0800)]
mds: adjust purge_stray sequence; include explicit ino destroy
First purge the inode content. Don't bother journaling our intent,
as that's implied by the fact that it's an unused stray.
Once purged, journal an event that destroys the inode and unlinks
the dentry. Don't remove null dentry itself, as we still need to
update the stray dir... it will get removed when that is committed.
Sage Weil [Fri, 14 Nov 2008 23:26:01 +0000 (15:26 -0800)]
mds: avoid unnecessary issue_caps in file_eval
Sage Weil [Fri, 14 Nov 2008 23:25:46 +0000 (15:25 -0800)]
mds: fix placement of eval_stray call on caps release
Sage Weil [Fri, 14 Nov 2008 23:02:34 +0000 (15:02 -0800)]
mds: restructure purge_stray to remove inode objects, _then_ dentry
This ensures that any inode we are purging is referenced in the
hierarchy, since we do not destroy the stray dentry until it is
completely gone.
Sage Weil [Fri, 14 Nov 2008 22:31:24 +0000 (14:31 -0800)]
mds: mark and pin dentries while purging, so they don't get trimmed out from under us
Aslo avoid purging more than once.
Previously it was possible to drop the dentry from the cache while
it was being purged.
Sage Weil [Fri, 14 Nov 2008 21:48:30 +0000 (13:48 -0800)]
mon: commit large numbers of state values quickly
Write them all, then sync once at the end.
Also include some infrastructure for using the latest stashed value
to recover. Don't use it yet, though. The interaction with
keeping last_committed and latest stashed values in sync wrt a
failure between the two is a bit tricky.
Sage Weil [Fri, 14 Nov 2008 20:42:56 +0000 (12:42 -0800)]
mon: use generic stash mechism to manage latest version of paxos-managed object
Sage Weil [Fri, 14 Nov 2008 20:08:55 +0000 (12:08 -0800)]
kclient: use generic timeout/retry code for various monitor request types
Sage Weil [Fri, 14 Nov 2008 19:14:30 +0000 (11:14 -0800)]
kclient: pick new mon if statfs is unresponsive; clean up other retry code
Sage Weil [Fri, 14 Nov 2008 01:45:37 +0000 (17:45 -0800)]
streamtest: fix recursive locking
Sage Weil [Fri, 14 Nov 2008 00:48:15 +0000 (16:48 -0800)]
journal: detect size of raw block devices properly
Sage Weil [Fri, 14 Nov 2008 00:37:56 +0000 (16:37 -0800)]
osd: only trim pg log if pg contains complete set of osds
Eventually we may want to also impose some maximum pg log size. At
some point the cost of the long log will approach the cost of
building a backlog...
Sage Weil [Thu, 13 Nov 2008 23:45:36 +0000 (15:45 -0800)]
osdmap: fix type conversions
Sage Weil [Thu, 13 Nov 2008 23:22:24 +0000 (15:22 -0800)]
crush: mention license. minor cleanup
Sage Weil [Thu, 13 Nov 2008 22:38:08 +0000 (14:38 -0800)]
be quiet
Yehuda Sadeh [Thu, 13 Nov 2008 22:56:43 +0000 (14:56 -0800)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable