]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
16 years agokclient: sync writes use page cache
Yehuda Sadeh [Mon, 1 Dec 2008 23:50:36 +0000 (15:50 -0800)]
kclient: sync writes use page cache

16 years agomds: do not purge until leases expire
Sage Weil [Mon, 1 Dec 2008 22:01:45 +0000 (14:01 -0800)]
mds: do not purge until leases expire

16 years agolockdep: lockdep_dump_locks()
Sage Weil [Mon, 1 Dec 2008 18:41:29 +0000 (10:41 -0800)]
lockdep: lockdep_dump_locks()

Handy for dumping held locks in gdb w/ 'p lockdep_dump_lock()'

16 years agorwlock: try_get_read, try_get_write
Sage Weil [Mon, 1 Dec 2008 18:13:00 +0000 (10:13 -0800)]
rwlock: try_get_read, try_get_write

16 years agoosd: optionally avoid zeroing trimmed log on disk
Sage Weil [Mon, 1 Dec 2008 15:13:30 +0000 (07:13 -0800)]
osd: optionally avoid zeroing trimmed log on disk

This is a half-hearted attempt to keep old PG log content around.  It'll
still be lost if a PG moves to another node or the entire log is written
to disk for some other reason.

16 years agokclient: fix bad check
Yehuda Sadeh [Mon, 1 Dec 2008 20:36:37 +0000 (12:36 -0800)]
kclient: fix bad check

16 years agokclient: slient down some log message
Yehuda Sadeh [Mon, 1 Dec 2008 19:39:00 +0000 (11:39 -0800)]
kclient: slient down some log message

16 years agoosd: skip peer_info on down osds
Sage Weil [Mon, 1 Dec 2008 04:42:44 +0000 (20:42 -0800)]
osd: skip peer_info on down osds

We don't clean old/down OSDs out of peer_info map, since we may not
restart peering when strays go up/down.  That's fine... just make sure
we ignore them later.

16 years agoosd: remove bad PG::put() assertion
Sage Weil [Mon, 1 Dec 2008 04:41:50 +0000 (20:41 -0800)]
osd: remove bad PG::put() assertion

16 years agoosd: fix lock inversion on workqueue shutdown
Sage Weil [Sun, 30 Nov 2008 21:06:37 +0000 (13:06 -0800)]
osd: fix lock inversion on workqueue shutdown

Simplify PG get/put vs lock/unlock.  Since we are reference counting with
an atomic_t, we don't need to re-use PG::lock to protect the reference
count.

16 years agotodo
Sage Weil [Thu, 27 Nov 2008 16:34:52 +0000 (08:34 -0800)]
todo

16 years agoworkqueue: deliberately leak string with lock name
Sage Weil [Thu, 27 Nov 2008 16:34:43 +0000 (08:34 -0800)]
workqueue: deliberately leak string with lock name

Lockdep assumes strings are statically allocated.

16 years agoMerge branch 'unstable' of ssh://yehudasa@ceph.newdream.net/git/ceph into unstable
Ariela [Thu, 27 Nov 2008 01:17:05 +0000 (17:17 -0800)]
Merge branch 'unstable' of ssh://yehudasa@ceph.newdream.net/git/ceph into unstable

16 years agowireshark: update for win32
Ariela [Thu, 27 Nov 2008 01:16:43 +0000 (17:16 -0800)]
wireshark: update for win32

16 years agoosd: fix bad sub_op_push assertion; only write data we need
Sage Weil [Thu, 27 Nov 2008 00:20:48 +0000 (16:20 -0800)]
osd: fix bad sub_op_push assertion; only write data we need

We adjust the pushed buffer so that we only write the portions
of it that we can't clone.

16 years agoobjectcacher: only call flushed callback if there are also no dirty buffers
Sage Weil [Thu, 27 Nov 2008 00:42:01 +0000 (16:42 -0800)]
objectcacher: only call flushed callback if there are also no dirty buffers

Otherwise we call the flushed_callback too soon.

16 years agomds: fix up loner_cap whenever we manually change filelock state
Sage Weil [Thu, 27 Nov 2008 00:32:29 +0000 (16:32 -0800)]
mds: fix up loner_cap whenever we manually change filelock state

This is kind of a pain.

16 years agoosd: clean up touch() calls to use exists bool
Sage Weil [Wed, 26 Nov 2008 22:35:56 +0000 (14:35 -0800)]
osd: clean up touch() calls to use exists bool

We only need to touch of the head object itself doesn't exist.
Do so far all ops.  This works cleanly despite any op munging
above.

16 years agoobjectstore: fix touch()
Sage Weil [Wed, 26 Nov 2008 22:35:02 +0000 (14:35 -0800)]
objectstore: fix touch()

Wrong opcode meant skewed args, and all kinds of badness.

16 years agomon: instruct individual pgs to scrub
Sage Weil [Wed, 26 Nov 2008 21:50:33 +0000 (13:50 -0800)]
mon: instruct individual pgs to scrub

Factor out pg parsing into pg_t.

16 years agoosd: debug write_info a bit, clean up Transaction cruft
Sage Weil [Wed, 26 Nov 2008 21:42:17 +0000 (13:42 -0800)]
osd: debug write_info a bit, clean up Transaction cruft

16 years agoosd: initiate scrub via monitor message
Sage Weil [Wed, 26 Nov 2008 21:41:14 +0000 (13:41 -0800)]
osd: initiate scrub via monitor message

16 years agoobjecter: scan_pgs even on original full map
Sage Weil [Wed, 26 Nov 2008 20:39:36 +0000 (12:39 -0800)]
objecter: scan_pgs even on original full map

We may have queued ops before getting _any_ map; those still need to
be kicked.

16 years agoosd: ensure target osd is still up when sending MPGRemoves
Sage Weil [Wed, 26 Nov 2008 20:18:10 +0000 (12:18 -0800)]
osd: ensure target osd is still up when sending MPGRemoves

16 years agoosd: move stats into PG::Info (disk format change)
Sage Weil [Wed, 26 Nov 2008 19:23:15 +0000 (11:23 -0800)]
osd: move stats into PG::Info (disk format change)

We want the pg stats to propagate along with last_update.  Do so
in merge_log.

Also, stop doing delayed stats update on primary; we always update
the in-core copy of Info, and only delay applying the transaction
to disk.  At least currently.

16 years agoosd: (re)set degraded flag on activate
Sage Weil [Wed, 26 Nov 2008 18:55:54 +0000 (10:55 -0800)]
osd: (re)set degraded flag on activate

This ensures the bit is properly set on newly created PGs..

16 years agotimer: discard unfired events on shutdown
Sage Weil [Wed, 26 Nov 2008 18:55:14 +0000 (10:55 -0800)]
timer: discard unfired events on shutdown

Mostly this just cleans up valgrind leak check output.

16 years agoclient: fix use-after-free
Sage Weil [Wed, 26 Nov 2008 18:44:48 +0000 (10:44 -0800)]
client: fix use-after-free

put_node at the end.

16 years agomsgr: discard queued messages when closing pipe
Sage Weil [Wed, 26 Nov 2008 18:35:23 +0000 (10:35 -0800)]
msgr: discard queued messages when closing pipe

These were leaking whenever we, say, did mark_down().

16 years agoobjectcacher: do callbacks _last_ to avoid a use-after-free
Sage Weil [Wed, 26 Nov 2008 18:19:08 +0000 (10:19 -0800)]
objectcacher: do callbacks _last_ to avoid a use-after-free

The callback may call back into release_set, so do not assume any pointers
will remain valid after the callback.

16 years agomds: initialize CInode::loner_cap when twiddling filelock states after reconnect
Sage Weil [Wed, 26 Nov 2008 17:58:32 +0000 (09:58 -0800)]
mds: initialize CInode::loner_cap when twiddling filelock states after reconnect

loner_cap must either be defined or not defined to match the lock states.

16 years agoobjecter: use full osdmap to get started
Sage Weil [Wed, 26 Nov 2008 17:57:35 +0000 (09:57 -0800)]
objecter: use full osdmap to get started

Previously we requested and then decoded the full history of incrementals,
a big waste of time during startup.

16 years agoosd: ensure 0-byte object created by push
Sage Weil [Wed, 26 Nov 2008 18:07:54 +0000 (10:07 -0800)]
osd: ensure 0-byte object created by push

Specifically, a zero-length HEAD object whose snapset we care
about.

16 years agoosd: mark repops aborteds in on_shutdown
Sage Weil [Wed, 26 Nov 2008 16:28:11 +0000 (08:28 -0800)]
osd: mark repops aborteds in on_shutdown

Otherwise the op_modify_ondisk() completion gets confused.

16 years agodstart: show filestore ops
Sage Weil [Wed, 26 Nov 2008 01:31:34 +0000 (17:31 -0800)]
dstart: show filestore ops

16 years agoosd: clear out workqueue queues on shutdown
Sage Weil [Wed, 26 Nov 2008 01:30:15 +0000 (17:30 -0800)]
osd: clear out workqueue queues on shutdown

16 years agomon: show both user data and actual disk space used
Sage Weil [Wed, 26 Nov 2008 01:28:00 +0000 (17:28 -0800)]
mon: show both user data and actual disk space used

16 years agodebug: rotate old courtesy symlinks to .0, .1, etc.
Sage Weil [Wed, 26 Nov 2008 01:26:56 +0000 (17:26 -0800)]
debug: rotate old courtesy symlinks to .0, .1, etc.

This will make it much easier to find the next-oldest instantiation of
osd foo.

16 years agoosd: remove bad assertion in op_modify_ondisk
Sage Weil [Tue, 25 Nov 2008 22:43:56 +0000 (14:43 -0800)]
osd: remove bad assertion in op_modify_ondisk

We may not be in waitfor_disk if we are marked failed and
on_osd_failure acked and removed us from the set.

16 years agoosd: fix clone push when head is correct old version
Sage Weil [Tue, 25 Nov 2008 22:00:55 +0000 (14:00 -0800)]
osd: fix clone push when head is correct old version

If the replica's head is the clone source version, our push should instruct
the replica to clone its contents.

16 years agomds: remove anchors when destroy stray inodes
Sage Weil [Tue, 25 Nov 2008 21:19:16 +0000 (13:19 -0800)]
mds: remove anchors when destroy stray inodes

purge_stray() is now the sole caller of anchor_destroy, so we can still
avoid any locking considerations.

16 years agomds: clean up request write condition, can_forward logic a bit
Sage Weil [Tue, 25 Nov 2008 20:50:00 +0000 (12:50 -0800)]
mds: clean up request write condition, can_forward logic a bit

16 years agoosd: fix rare memory leak
Sage Weil [Tue, 25 Nov 2008 20:21:20 +0000 (12:21 -0800)]
osd: fix rare memory leak

16 years agocmonctl: print summary every 20 lines when in watch mode
Sage Weil [Tue, 25 Nov 2008 20:14:26 +0000 (12:14 -0800)]
cmonctl: print summary every 20 lines when in watch mode

16 years agoosd: fix up recovery pointers a bit
Sage Weil [Tue, 25 Nov 2008 19:22:14 +0000 (11:22 -0800)]
osd: fix up recovery pointers a bit

16 years agowireshark: wireshark ceph plugin patch
Yehuda Sadeh [Tue, 25 Nov 2008 18:57:37 +0000 (10:57 -0800)]
wireshark: wireshark ceph plugin patch

16 years agoosd: be a bit more verbose in push_to_replica
Sage Weil [Tue, 25 Nov 2008 18:42:36 +0000 (10:42 -0800)]
osd: be a bit more verbose in push_to_replica

16 years agoosd: lots of fixes
Sage Weil [Tue, 25 Nov 2008 18:02:48 +0000 (10:02 -0800)]
osd: lots of fixes

16 years agotodos
Sage Weil [Tue, 25 Nov 2008 00:41:54 +0000 (16:41 -0800)]
todos

16 years agoosd: clean up repop code
Sage Weil [Tue, 25 Nov 2008 14:45:48 +0000 (06:45 -0800)]
osd: clean up repop code

16 years agoosd: fix up repop_ack
Sage Weil [Tue, 25 Nov 2008 04:46:16 +0000 (20:46 -0800)]
osd: fix up repop_ack

16 years agoosd: ack type in osd sub ops
Sage Weil [Tue, 25 Nov 2008 00:52:23 +0000 (16:52 -0800)]
osd: ack type in osd sub ops

16 years agoosd: infrastructure for ack vs nvram vs disk osd_op ack types
Sage Weil [Tue, 25 Nov 2008 00:35:43 +0000 (16:35 -0800)]
osd: infrastructure for ack vs nvram vs disk osd_op ack types

NVRAM ack not generated, yet.  The completion callbacks from the store
need some work first.

16 years agoos: separate onjournal, ondisk callbacks for apply_transaction
Sage Weil [Mon, 24 Nov 2008 22:56:46 +0000 (14:56 -0800)]
os: separate onjournal, ondisk callbacks for apply_transaction

16 years agocontext: all C_Gather to OR instead of AND subs.
Sage Weil [Mon, 24 Nov 2008 22:06:49 +0000 (14:06 -0800)]
context: all C_Gather to OR instead of AND subs.

That is, call completion when first sub completes.

16 years agoosd: remove snap collection after it is trimmed
Sage Weil [Mon, 24 Nov 2008 21:49:17 +0000 (13:49 -0800)]
osd: remove snap collection after it is trimmed

We can still end up with empty collections for existing snaps but no
local objects.  However, they'll eventually go away when the snap is
deleted, so who cares.

16 years agoos: add collection_empty method
Sage Weil [Mon, 24 Nov 2008 21:47:54 +0000 (13:47 -0800)]
os: add collection_empty method

16 years agotodos
Sage Weil [Mon, 24 Nov 2008 20:06:18 +0000 (12:06 -0800)]
todos

16 years agoosd: reply with EBLACKLISTED if sender is blacklisted
Sage Weil [Mon, 24 Nov 2008 20:06:08 +0000 (12:06 -0800)]
osd: reply with EBLACKLISTED if sender is blacklisted

Move reply_op_error helper into OSD.cc.

16 years agoosd: include a blacklist in the OSDMap
Sage Weil [Mon, 24 Nov 2008 19:49:10 +0000 (11:49 -0800)]
osd: include a blacklist in the OSDMap

16 years agoosd: comment
Sage Weil [Mon, 24 Nov 2008 19:39:51 +0000 (11:39 -0800)]
osd: comment

16 years agorev wire, disk formats
Sage Weil [Mon, 24 Nov 2008 19:39:43 +0000 (11:39 -0800)]
rev wire, disk formats

osdmap encoding changed.

16 years agoosd: ignore intervals prior to last_epoch_started in build_prior
Sage Weil [Mon, 24 Nov 2008 19:34:02 +0000 (11:34 -0800)]
osd: ignore intervals prior to last_epoch_started in build_prior

We may have raised last_epoch_started without trimming past_intervals.

16 years agoosd: simplify osdmap tracking of osd up/down epochs; fix pg build_prior logic
Sage Weil [Mon, 24 Nov 2008 19:30:17 +0000 (11:30 -0800)]
osd: simplify osdmap tracking of osd up/down epochs; fix pg build_prior logic

Use a single struct to track all of our osd up/down info.  Include
down_at, the epoch we last marked the osd down.

Fix PG::build_prior to require that the osd was clean through the _entire_
interval in question.

In monitor, adjust new clean interval foward to down_at-1 if the up_from
matches the interval we mounted.  That is, if the OSD shut down cleanly,
it obviously remained clean at least until we marked it down in the map.

16 years agomds: move filelock to lock state if we can't wrlock but lock is stable
Sage Weil [Mon, 24 Nov 2008 21:26:11 +0000 (13:26 -0800)]
mds: move filelock to lock state if we can't wrlock but lock is stable

For example, lock may be sync c=1 when we're trying to reset
max_size to 0.  We need to make sure the lock will change state
before we wait on WAIT_STABLE.

16 years agomds: rejournal using EUpdate instead of EOpen if no caps in check_inode_max_size
Sage Weil [Mon, 24 Nov 2008 21:23:53 +0000 (13:23 -0800)]
mds: rejournal using EUpdate instead of EOpen if no caps in check_inode_max_size

Using EOpen on non-open files is imprecise.  More importantly, if
it is a snapped inode, the EOpen replay code won't be able to
look up the ino and will throw an assertion.

So, use EUpdate instead to record the new info when necessary.

16 years agomds: use vector for EOpen
Sage Weil [Mon, 24 Nov 2008 21:21:15 +0000 (13:21 -0800)]
mds: use vector for EOpen

16 years agomake assertion output look more like gcc
Sage Weil [Mon, 24 Nov 2008 21:20:52 +0000 (13:20 -0800)]
make assertion output look more like gcc

16 years agomds: remove bad assertion
Sage Weil [Mon, 24 Nov 2008 20:08:49 +0000 (12:08 -0800)]
mds: remove bad assertion

remove_inode should do the asserting, here, as it may do some
cleanup.

16 years agoosd: use last_clean_interval in build_prior logic
Sage Weil [Mon, 24 Nov 2008 18:00:21 +0000 (10:00 -0800)]
osd: use last_clean_interval in build_prior logic

We now mark a PG crashed if any of the OSDs during a given interval
is not either still alive or cleanly shut down during the interval.  If
those two conditions are not yet, it may have crashed.

It isn't a perfect set of criteria, since last_clean_interval is only
clean shutdowns of an OSD.  We could, for example, track another interval
generated via old osd_up_thru, but the conditions for that are different,
since that requires survival past the end of the interval, not a clean
shutdown during the interval.  This should capture the common case, though,
of a clean unmount.
to

16 years agoosd: track last_clean_interval in osdmap; simplify encoding/decoding a bit
Sage Weil [Mon, 24 Nov 2008 17:57:11 +0000 (09:57 -0800)]
osd: track last_clean_interval in osdmap; simplify encoding/decoding a bit

Break osdmap into "base" and "extended" portions, so that clients can
ignore the extended portions completely.

Track last_clean_interval in the osdmap so we know when the osd last
cleanly shut down.

Disk and wire format changes.

16 years agoosd: track last_clean_interval in superblock
Sage Weil [Mon, 24 Nov 2008 17:55:23 +0000 (09:55 -0800)]
osd: track last_clean_interval in superblock

16 years agoosd: add a few getattr assertions
Sage Weil [Mon, 24 Nov 2008 18:27:22 +0000 (10:27 -0800)]
osd: add a few getattr assertions

Ensure we got the snapset attr before we decode it.

16 years agoosd: log a delete only if the head object is deleted
Sage Weil [Mon, 24 Nov 2008 18:25:11 +0000 (10:25 -0800)]
osd: log a delete only if the head object is deleted

Whether the head logically exists is not relevant to our logging.
Do not look at snapset.head_exists.

16 years agoosd: remove bad assertion in pick_read_snap
Sage Weil [Mon, 24 Nov 2008 18:17:58 +0000 (10:17 -0800)]
osd: remove bad assertion in pick_read_snap

The clone oid.snap does not necessary correspond to the newest
snap, since snaps may be deleted, or because the snap the snap is
named based on the snap context seq and not the oldest snap it
contains.

16 years agomds: remove capless inodes from logsegment open_file lists after reconnect
Sage Weil [Sun, 23 Nov 2008 18:57:28 +0000 (10:57 -0800)]
mds: remove capless inodes from logsegment open_file lists after reconnect

Inodes get added during replay of EOpen events.  We remove capless inodes
after reconnect restores from clients.

We only want inodes with caps on those lists.  Add an assertion to
enforce constraint.

Also, remove ourselves explicitly in remove_inode(), since that may happen
during replay when an inode is destroyed.

16 years agomds: fix up completed_request handling during journal replay
Sage Weil [Sat, 22 Nov 2008 16:57:19 +0000 (08:57 -0800)]
mds: fix up completed_request handling during journal replay

The completed_requests is handled separately from the session table
itself, in that we may add completed requests to the table even when
we may have loaded newer info.  But the handling was a bit wrong.
We make sure we only add completed requests if the session is already
open... and remove the unnecessary trim (if the sessionmap is newer, the
session is already closed, and thus we have no request info).

16 years agoosd: clean up info setattrs, such that write_info is called once per map update
Sage Weil [Sat, 22 Nov 2008 05:49:11 +0000 (21:49 -0800)]
osd: clean up info setattrs, such that write_info is called once per map update

Use pg dirty flags, and do the final write_info and/or write_log right
before we apply the advance/activate_map transaction.

PG.cc still calls the methods directly in methods that are not in the
advance/activate paths.

16 years agofilestore: clean up debug output
Sage Weil [Sat, 22 Nov 2008 04:46:10 +0000 (20:46 -0800)]
filestore: clean up debug output

10 ops
15 commits
20 waits

16 years agoosd: minor append_log cleanup
Sage Weil [Sat, 22 Nov 2008 04:36:38 +0000 (20:36 -0800)]
osd: minor append_log cleanup

16 years agofilestore: only commit if changes are pending
Sage Weil [Sat, 22 Nov 2008 04:31:53 +0000 (20:31 -0800)]
filestore: only commit if changes are pending

Avoid needlessly scribbling all over the superblocks.

16 years agofilestore: fix truncate argument, subsequent Transaction fuggering
Sage Weil [Sat, 22 Nov 2008 04:30:00 +0000 (20:30 -0800)]
filestore: fix truncate argument, subsequent Transaction fuggering

16 years agodstart: modprobe btrfs
Sage Weil [Sat, 22 Nov 2008 00:35:16 +0000 (16:35 -0800)]
dstart: modprobe btrfs

16 years agomds: assert ref count is 0 on inode deletion
Sage Weil [Fri, 21 Nov 2008 23:01:22 +0000 (15:01 -0800)]
mds: assert ref count is 0 on inode deletion

16 years agomds: mark dirty inode clean before dropping
Sage Weil [Fri, 21 Nov 2008 23:00:55 +0000 (15:00 -0800)]
mds: mark dirty inode clean before dropping

This has the effect of removing the inode from the dirty xlist.

16 years agomds: exclude BADREMOTEINO dentriess in readdir
Sage Weil [Fri, 21 Nov 2008 22:58:14 +0000 (14:58 -0800)]
mds: exclude BADREMOTEINO dentriess in readdir

16 years agomds: make open_remote_ino terminate if the anchortrace refers to a non-existent ino
Sage Weil [Fri, 21 Nov 2008 22:58:00 +0000 (14:58 -0800)]
mds: make open_remote_ino terminate if the anchortrace refers to a non-existent ino

Since anchor lookup is racy, we may have to do multiple lookups
(in the case of a concurrent anchor table update).  If we don't
find the ino in the anchor, remember the anchor version when we
try again.  If we fail again at the same point, and the anchor
has not changed, fail.

Create an open_remote_dentry helper that does this.  If we fail,
set the CDentry::STATE_BADREMOTEINO state bit.

16 years agomds: fix cdentry states
Sage Weil [Fri, 21 Nov 2008 22:55:59 +0000 (14:55 -0800)]
mds: fix cdentry states

Also add BADREMOTEINO, which we'll use shortly.

16 years agomds: add version to anchor; avoid looping in open_remote_ino
Sage Weil [Fri, 21 Nov 2008 22:34:40 +0000 (14:34 -0800)]
mds: add version to anchor; avoid looping in open_remote_ino

If we do not find the item referenced by teh anchor trace, we
try again, but keep track of the anchor version we ended on.  If
we hit a dead end at the same point next time and the anchor
hasn't changed, we give up.

16 years agomon: send osdmap to _original_ PGStats source
Sage Weil [Fri, 21 Nov 2008 21:30:27 +0000 (13:30 -0800)]
mon: send osdmap to _original_ PGStats source

..not the most recent sender, which may be another monitor.

16 years agomsgr: fix replace_connection bucket list manipulation
Sage Weil [Fri, 21 Nov 2008 21:24:40 +0000 (13:24 -0800)]
msgr: fix replace_connection bucket list manipulation

16 years agokclient: increase max data and front sizes to 16MB
Yehuda Sadeh [Fri, 21 Nov 2008 21:15:35 +0000 (13:15 -0800)]
kclient: increase max data and front sizes to 16MB

16 years agokclient: preserve peer_name across connection replacement
Sage Weil [Fri, 21 Nov 2008 21:11:46 +0000 (13:11 -0800)]
kclient: preserve peer_name across connection replacement

16 years agoosd: factor out add_log_entry into a helper that adjusts pg info accordingly
Sage Weil [Fri, 21 Nov 2008 20:35:20 +0000 (12:35 -0800)]
osd: factor out add_log_entry into a helper that adjusts pg info accordingly

We want to keep log.top, info.last_update etc. in sync.

16 years agoosd: use old acting set when noting past intervals
Sage Weil [Fri, 21 Nov 2008 20:02:56 +0000 (12:02 -0800)]
osd: use old acting set when noting past intervals

16 years agodstart: only clean up old output on mkfs
Sage Weil [Fri, 21 Nov 2008 19:56:59 +0000 (11:56 -0800)]
dstart: only clean up old output on mkfs

16 years agoclient: ignore dropped messages
Sage Weil [Fri, 21 Nov 2008 19:39:37 +0000 (11:39 -0800)]
client: ignore dropped messages

16 years agopg: fix build_prior debug output
Sage Weil [Fri, 21 Nov 2008 19:21:59 +0000 (11:21 -0800)]
pg: fix build_prior debug output

16 years agoclient: fix ms_handle_failure
Sage Weil [Fri, 21 Nov 2008 19:21:46 +0000 (11:21 -0800)]
client: fix ms_handle_failure

16 years agoosd: only trim past intervals that _fully_ preceed last epoch started
Sage Weil [Fri, 21 Nov 2008 19:08:23 +0000 (11:08 -0800)]
osd: only trim past intervals that _fully_ preceed last epoch started