]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
16 years agoosd: report pg osds, osd peers to pgmonitor; include in pg dump
Sage Weil [Fri, 31 Oct 2008 17:43:48 +0000 (10:43 -0700)]
osd: report pg osds, osd peers to pgmonitor; include in pg dump

16 years agokclient: keep a pointer to the current snap context in the inode
Yehuda Sadeh [Fri, 31 Oct 2008 18:07:23 +0000 (11:07 -0700)]
kclient: keep a pointer to the current snap context in the inode

16 years agokclient: use current snap context if not found
Yehuda Sadeh [Fri, 31 Oct 2008 17:18:00 +0000 (10:18 -0700)]
kclient: use current snap context if not found

16 years agokclient: don't register a new bdi for the same client
Yehuda Sadeh [Thu, 30 Oct 2008 23:35:23 +0000 (16:35 -0700)]
kclient: don't register a new bdi for the same client

16 years agoosd: do all recovery operations in dedicated recovery thread
Sage Weil [Thu, 30 Oct 2008 23:21:33 +0000 (16:21 -0700)]
osd: do all recovery operations in dedicated recovery thread

16 years agodstart.sh: 2x rep only
Sage Weil [Thu, 30 Oct 2008 22:50:46 +0000 (15:50 -0700)]
dstart.sh: 2x rep only

16 years agodstop.sh: stop crun too
Sage Weil [Thu, 30 Oct 2008 22:00:43 +0000 (15:00 -0700)]
dstop.sh: stop crun too

16 years agokclient: override rdcache invalidation time when going down
Yehuda Sadeh [Thu, 30 Oct 2008 22:57:44 +0000 (15:57 -0700)]
kclient: override rdcache invalidation time when going down

16 years agoosd: avoid trimming unless pg is clean; keep recovered_to, complete_to log pointers...
Sage Weil [Thu, 30 Oct 2008 21:23:03 +0000 (14:23 -0700)]
osd: avoid trimming unless pg is clean; keep recovered_to, complete_to log pointers cleared when not in use

16 years agokclient: do LCHOWN, not CHOWN, so that mds does not follow the symlink
Sage Weil [Thu, 30 Oct 2008 21:19:25 +0000 (14:19 -0700)]
kclient: do LCHOWN, not CHOWN, so that mds does not follow the symlink

16 years agokclient: update patch generator
Sage Weil [Thu, 30 Oct 2008 20:01:17 +0000 (13:01 -0700)]
kclient: update patch generator

16 years agomds: fix bad EINVAL error code in unlink
Sage Weil [Thu, 30 Oct 2008 18:48:39 +0000 (11:48 -0700)]
mds: fix bad EINVAL error code in unlink

16 years agokclient: note locked new_dir in ceph_rename
Sage Weil [Thu, 30 Oct 2008 18:48:20 +0000 (11:48 -0700)]
kclient: note locked new_dir in ceph_rename

16 years agomds: fix pre_dirty/mark_dirty order for remote dentries in unlink_local
Sage Weil [Thu, 30 Oct 2008 18:15:55 +0000 (11:15 -0700)]
mds: fix pre_dirty/mark_dirty order for remote dentries in unlink_local

16 years agokclient: use vmalloc for messages larger than one page
Sage Weil [Thu, 30 Oct 2008 18:05:25 +0000 (11:05 -0700)]
kclient: use vmalloc for messages larger than one page

16 years agokclient: make reconnect allocation a bit more generous
Sage Weil [Thu, 30 Oct 2008 18:00:46 +0000 (11:00 -0700)]
kclient: make reconnect allocation a bit more generous

16 years agokclient: remove extra dput on failed rmdir/unlink
Yehuda Sadeh [Wed, 29 Oct 2008 21:38:53 +0000 (14:38 -0700)]
kclient: remove extra dput on failed rmdir/unlink

16 years agoosd: initialize curclone
Sage Weil [Wed, 29 Oct 2008 21:59:35 +0000 (14:59 -0700)]
osd: initialize curclone

16 years agofilestore: fix getattrs return value
Sage Weil [Wed, 29 Oct 2008 21:59:21 +0000 (14:59 -0700)]
filestore: fix getattrs return value

16 years agokclient: fix mds request reference after free bug
Sage Weil [Wed, 29 Oct 2008 21:47:22 +0000 (14:47 -0700)]
kclient: fix mds request reference after free bug

16 years agoRevert "mds: CInode::mark_dirty should then dirty the projected parent inode, if...
Sage Weil [Wed, 29 Oct 2008 21:22:02 +0000 (14:22 -0700)]
Revert "mds: CInode::mark_dirty should then dirty the projected parent inode, if present"

We _shouldn't_ use the projected dentry, only the current dentry, since updates
to this inode may be pipelined with other updates and/or a parent change.  If
we are really doing a rename, fix the parent, _then_ dirty the inode.

This reverts commit 89c635d3ee6038daee352e238d00b94094927571.

16 years agomon: lossless communication between monitors
Sage Weil [Wed, 29 Oct 2008 21:03:05 +0000 (14:03 -0700)]
mon: lossless communication between monitors

16 years agokclient: set r_got_reply _after_ we ensure we have a session
Sage Weil [Wed, 29 Oct 2008 20:58:42 +0000 (13:58 -0700)]
kclient: set r_got_reply _after_ we ensure we have a session

16 years agomds: ClientCaps typo
Sage Weil [Wed, 29 Oct 2008 20:50:27 +0000 (13:50 -0700)]
mds: ClientCaps typo

16 years agomds: eval FileLock after last wrlock releases
Sage Weil [Wed, 29 Oct 2008 20:50:15 +0000 (13:50 -0700)]
mds: eval FileLock after last wrlock releases

16 years agokclient: no cap delay if mdsc->stopping (unmounting)
Sage Weil [Wed, 29 Oct 2008 20:48:57 +0000 (13:48 -0700)]
kclient: no cap delay if mdsc->stopping (unmounting)

16 years agoosd: clean up debug output a bit
Sage Weil [Wed, 29 Oct 2008 19:14:27 +0000 (12:14 -0700)]
osd: clean up debug output a bit

Less chatter about pg stat updates to monitor

16 years agofilestore: don't clobber file on clone_range
Sage Weil [Wed, 29 Oct 2008 19:06:37 +0000 (12:06 -0700)]
filestore: don't clobber file on clone_range

16 years agoosd: initialize clone_overlap on clone
Sage Weil [Wed, 29 Oct 2008 19:05:51 +0000 (12:05 -0700)]
osd: initialize clone_overlap on clone

Initially, the clone overlaps completely with the newer item (head).

16 years agoosd: skip pg log object when generating backlog
Sage Weil [Wed, 29 Oct 2008 18:15:50 +0000 (11:15 -0700)]
osd: skip pg log object when generating backlog

16 years agofilestore: use strtoull (unsigned) instead of strtoll when parsing object names,...
Sage Weil [Wed, 29 Oct 2008 18:15:17 +0000 (11:15 -0700)]
filestore: use strtoull (unsigned) instead of strtoll when parsing object names, etc.

Otherwise we fall over on CEPH_NOSNAP (-2), notably.

16 years agobuffer: print 'no raw' for 'null' bufferptrs
Sage Weil [Wed, 29 Oct 2008 18:13:44 +0000 (11:13 -0700)]
buffer: print 'no raw' for 'null' bufferptrs

Aids debugging...

16 years agoobjectstore: do not include 0-length bufferptr in bufferlist if attr not found
Sage Weil [Wed, 29 Oct 2008 18:13:18 +0000 (11:13 -0700)]
objectstore: do not include 0-length bufferptr in bufferlist if attr not found

Having a "null" bufferptr causes problems in a number of places down the
line.  We want an empty bufferlist instead.

16 years agovstartnew.sh: clean up out/* even if owned by root
Sage Weil [Tue, 28 Oct 2008 23:50:38 +0000 (16:50 -0700)]
vstartnew.sh: clean up out/* even if owned by root

16 years agoosd: sync on shutdown
Sage Weil [Tue, 28 Oct 2008 23:50:17 +0000 (16:50 -0700)]
osd: sync on shutdown

There may be write completion callbacks in flight that depend on what
we're about to tear down.

16 years agoosd: batch pg remove messages, do them all at once
Sage Weil [Tue, 28 Oct 2008 23:49:46 +0000 (16:49 -0700)]
osd: batch pg remove messages, do them all at once

16 years agofilestore: encode transaction only if journaling it
Sage Weil [Tue, 28 Oct 2008 23:48:59 +0000 (16:48 -0700)]
filestore: encode transaction only if journaling it

And do it later.  No need to encode it before applying, now.

16 years agosynclient: mksnapfile to generate a simple multi-snapshot object
Sage Weil [Tue, 28 Oct 2008 23:48:21 +0000 (16:48 -0700)]
synclient: mksnapfile to generate a simple multi-snapshot object

writes 0~4M, takes snapshot, then overwrites 1~2M.

16 years agomds: generate valid seq value on rmsnap
Sage Weil [Tue, 28 Oct 2008 23:31:36 +0000 (16:31 -0700)]
mds: generate valid seq value on rmsnap

16 years agoosd: fix typo on setting complete_to
Sage Weil [Tue, 28 Oct 2008 21:34:58 +0000 (14:34 -0700)]
osd: fix typo on setting complete_to

16 years agoosd: include fsid in MPGStats
Sage Weil [Tue, 28 Oct 2008 20:49:32 +0000 (13:49 -0700)]
osd: include fsid in MPGStats

16 years agobuffer: skip crc calc on zero length bufferptrs
Sage Weil [Tue, 28 Oct 2008 20:47:16 +0000 (13:47 -0700)]
buffer: skip crc calc on zero length bufferptrs

16 years agomds: CInode::mark_dirty should then dirty the projected parent inode, if present
Sage Weil [Tue, 28 Oct 2008 20:59:32 +0000 (13:59 -0700)]
mds: CInode::mark_dirty should then dirty the projected parent inode, if present

16 years agomon: ensure snap doesn't already exist in pending set
Sage Weil [Tue, 28 Oct 2008 20:45:30 +0000 (13:45 -0700)]
mon: ensure snap doesn't already exist in pending set

16 years agokclient: comment on paths for mds ops
Sage Weil [Tue, 28 Oct 2008 20:26:22 +0000 (13:26 -0700)]
kclient: comment on paths for mds ops

16 years agoosd: reduce max_rep to 3 by default
Sage Weil [Mon, 27 Oct 2008 23:07:34 +0000 (16:07 -0700)]
osd: reduce max_rep to 3 by default

16 years agomds: increase max_size based on loner AND other caps
Sage Weil [Mon, 27 Oct 2008 22:22:54 +0000 (15:22 -0700)]
mds: increase max_size based on loner AND other caps

16 years agoMerge branch 'recovery' into unstable
Sage Weil [Mon, 27 Oct 2008 21:02:46 +0000 (14:02 -0700)]
Merge branch 'recovery' into unstable

16 years agoosd: recovery more efficiently keeps N ops in flight
Sage Weil [Mon, 27 Oct 2008 21:01:56 +0000 (14:01 -0700)]
osd: recovery more efficiently keeps N ops in flight

16 years agoobjectcacher: fix commit callback accounting
Sage Weil [Mon, 27 Oct 2008 20:28:43 +0000 (13:28 -0700)]
objectcacher: fix commit callback accounting

16 years agokclient: clean out when generating patches
Sage Weil [Mon, 27 Oct 2008 18:31:38 +0000 (11:31 -0700)]
kclient: clean out when generating patches

16 years agoxlist: implement push_front()
Sage Weil [Mon, 27 Oct 2008 18:53:32 +0000 (11:53 -0700)]
xlist: implement push_front()

16 years agokclient: remove warning due to missing format
Yehuda Sadeh [Mon, 27 Oct 2008 18:49:44 +0000 (11:49 -0700)]
kclient: remove warning due to missing format

16 years agomds: make loner unitialized warning go away
Sage Weil [Mon, 27 Oct 2008 17:32:52 +0000 (10:32 -0700)]
mds: make loner unitialized warning go away

16 years agokclient: bleh
Sage Weil [Mon, 27 Oct 2008 18:23:06 +0000 (11:23 -0700)]
kclient: bleh

16 years agokclient: include documentation in patch queue
Sage Weil [Mon, 27 Oct 2008 18:20:10 +0000 (11:20 -0700)]
kclient: include documentation in patch queue

16 years agofilestore: be less noisy
Sage Weil [Mon, 27 Oct 2008 17:28:18 +0000 (10:28 -0700)]
filestore: be less noisy

16 years agoMerge commit 'origin/unstable' into recovery
Sage Weil [Mon, 27 Oct 2008 17:03:30 +0000 (10:03 -0700)]
Merge commit 'origin/unstable' into recovery

16 years agokclient: typo
Sage Weil [Mon, 27 Oct 2008 03:09:42 +0000 (20:09 -0700)]
kclient: typo

16 years agomds: fix filelock loner modes
Sage Weil [Mon, 27 Oct 2008 03:07:02 +0000 (20:07 -0700)]
mds: fix filelock loner modes

Rewrote Filelock handling of "loner" mode, which gives a single client access
to do cached/bufferd read/write on a file.  To make this sane, CInode::loner_cap
identifies the distinguished client (if any), and the FileLock caps methods
provides issued/allowed/wanted values for loner and other clients.  A new
lock state is added to track revocation of RDCACHE from lock->loner on the
non-loner clients.

Overall, the new code is much cleaner and should be less susceptible to
odd corner cases...

16 years agomds: fix rfile/rsubdirs accounting vs linkunlink
Sage Weil [Fri, 24 Oct 2008 23:47:19 +0000 (16:47 -0700)]
mds: fix rfile/rsubdirs accounting vs linkunlink

16 years agokclient: xattr rsubdirs fix
Yehuda Sadeh [Fri, 24 Oct 2008 22:53:51 +0000 (15:53 -0700)]
kclient: xattr rsubdirs fix

16 years agokclient: take i_lock in flush_snaps
Sage Weil [Fri, 24 Oct 2008 22:29:52 +0000 (15:29 -0700)]
kclient: take i_lock in flush_snaps

16 years agokclient: get code up to date with latest linux kernel
Yehuda Sadeh [Fri, 24 Oct 2008 21:07:29 +0000 (14:07 -0700)]
kclient: get code up to date with latest linux kernel

16 years agokclient: flush/release all caps back to mds before closing a session
Sage Weil [Fri, 24 Oct 2008 20:56:45 +0000 (13:56 -0700)]
kclient: flush/release all caps back to mds before closing a session

This allows us to handle races that can occur if an inode moves between
MDSs.  It also ensures that _all_ dirty cap data is safely flushed before
we throw it out.

To make this work, we drop the PIN reference on the subdirectory that
we mounted.  Since VFS doesn't tell us when each vfsmount goes away,
we can't remove them as we go.  And we're more or less screwed anyway
if the directory gets renamed (currently, at least), so there's not
much point in having this reference in the first place.

16 years agocrush: return 1.0 from get_offload if osd is not in the map
Sage Weil [Fri, 24 Oct 2008 18:44:11 +0000 (11:44 -0700)]
crush: return 1.0 from get_offload if osd is not in the map

16 years agokclient: remove realm from parent's child list on removal; avoid letting ref count...
Sage Weil [Fri, 24 Oct 2008 18:30:06 +0000 (11:30 -0700)]
kclient: remove realm from parent's child list on removal; avoid letting ref count hit 0 during split

16 years agokclient: remove some sparse warnings
Yehuda Sadeh [Fri, 24 Oct 2008 18:09:12 +0000 (11:09 -0700)]
kclient: remove some sparse warnings

16 years agokclient: fix cap flush, and drop ref when acked by mds
Sage Weil [Fri, 24 Oct 2008 18:16:56 +0000 (11:16 -0700)]
kclient: fix cap flush, and drop ref when acked by mds

16 years agoMerge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Fri, 24 Oct 2008 17:47:09 +0000 (10:47 -0700)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable

16 years agokclient: finish_cap_snap fixes
Sage Weil [Fri, 24 Oct 2008 17:57:39 +0000 (10:57 -0700)]
kclient: finish_cap_snap fixes

16 years agokclient: queue finished cap_snaps and flush them after snap_rwsem is dropped
Sage Weil [Fri, 24 Oct 2008 17:47:52 +0000 (10:47 -0700)]
kclient: queue finished cap_snaps and flush them after snap_rwsem is dropped

This avoids some lock ordering problems.  It's also just smarter.. we don't
want to send all those messages while holding snap_rwsem.

16 years agokclient: cleanup some sparse messages
Yehuda Sadeh [Fri, 24 Oct 2008 17:47:07 +0000 (10:47 -0700)]
kclient: cleanup some sparse messages

16 years agokclient: free snap_rwsem at ceph_handle_snap
Yehuda Sadeh [Fri, 24 Oct 2008 17:33:24 +0000 (10:33 -0700)]
kclient: free snap_rwsem at ceph_handle_snap

16 years agokclient: remove the use of snap_rwsem at ceph_mdsc_close_sessions()
Yehuda Sadeh [Fri, 24 Oct 2008 17:31:08 +0000 (10:31 -0700)]
kclient: remove the use of snap_rwsem at ceph_mdsc_close_sessions()

16 years agokclient: fix __ceph_flush_snaps use of 'follows'
Yehuda Sadeh [Fri, 24 Oct 2008 17:30:21 +0000 (10:30 -0700)]
kclient: fix __ceph_flush_snaps use of 'follows'

16 years agokclient: drop reply mutex, add a flag instead
Yehuda Sadeh [Thu, 23 Oct 2008 19:48:42 +0000 (12:48 -0700)]
kclient: drop reply mutex, add a flag instead

16 years agokclient: use a new reply mutex, fixes locks dependency problem
Yehuda Sadeh [Thu, 23 Oct 2008 18:49:27 +0000 (11:49 -0700)]
kclient: use a new reply mutex, fixes locks dependency problem

16 years agokclient: fix allocation flags for osd client radix tree
Sage Weil [Thu, 23 Oct 2008 03:36:52 +0000 (20:36 -0700)]
kclient: fix allocation flags for osd client radix tree

16 years agokclient: try to release RDCACHE if _no_ caps are wanted
Sage Weil [Thu, 23 Oct 2008 02:17:21 +0000 (19:17 -0700)]
kclient: try to release RDCACHE if _no_ caps are wanted

As Yehuda pointed out, we aren't really concerned about invalidating pages and
releasing RDCACHE if we are still open for write.  So only bother releasing
RDCACHE when we have _no_ more interest in the file.  If there is a conflict,
the MDS will just revoke.

16 years agokclient: invalidate pages if RDCACHE used but not wanted
Sage Weil [Thu, 23 Oct 2008 00:16:06 +0000 (17:16 -0700)]
kclient: invalidate pages if RDCACHE used but not wanted

16 years agokclient: changing around the rdcache invalidation stuff
Sage Weil [Wed, 22 Oct 2008 23:40:44 +0000 (16:40 -0700)]
kclient: changing around the rdcache invalidation stuff

16 years agokclient: fix messenger tmp var name
Sage Weil [Wed, 22 Oct 2008 22:58:36 +0000 (15:58 -0700)]
kclient: fix messenger tmp var name

16 years agokclient: cleanup
Yehuda Sadeh [Wed, 22 Oct 2008 22:32:42 +0000 (15:32 -0700)]
kclient: cleanup

16 years agokclient: revocation of rdcache handling
Yehuda Sadeh [Wed, 22 Oct 2008 22:12:12 +0000 (15:12 -0700)]
kclient: revocation of rdcache handling

16 years agokclient: more patch queue futzing
Sage Weil [Wed, 22 Oct 2008 22:26:50 +0000 (15:26 -0700)]
kclient: more patch queue futzing

16 years agocrush: comments, cleanup
Sage Weil [Wed, 22 Oct 2008 22:26:04 +0000 (15:26 -0700)]
crush: comments, cleanup

16 years agokclient: flush cap_snaps on session close
Sage Weil [Wed, 22 Oct 2008 19:35:43 +0000 (12:35 -0700)]
kclient: flush cap_snaps on session close

16 years agokclient: build patches too in import script
Sage Weil [Wed, 22 Oct 2008 17:42:15 +0000 (10:42 -0700)]
kclient: build patches too in import script

16 years agokclient: remove remaining #include/#ifdef cruft from ceph_fs.h
Sage Weil [Wed, 22 Oct 2008 17:54:44 +0000 (10:54 -0700)]
kclient: remove remaining #include/#ifdef cruft from ceph_fs.h

16 years agokclient: patch queue commit comments
Sage Weil [Wed, 22 Oct 2008 05:14:06 +0000 (22:14 -0700)]
kclient: patch queue commit comments

16 years agoclean up ceph_fs.h a bit
Sage Weil [Wed, 22 Oct 2008 04:42:48 +0000 (21:42 -0700)]
clean up ceph_fs.h a bit

16 years agokclient: more patch queue fun
Sage Weil [Tue, 21 Oct 2008 23:57:27 +0000 (16:57 -0700)]
kclient: more patch queue fun

16 years agokclient: adjust import queue
Sage Weil [Wed, 22 Oct 2008 00:04:45 +0000 (17:04 -0700)]
kclient: adjust import queue

16 years agokclient: allow monitor port to be specified with the ip(s) during mount
Sage Weil [Tue, 21 Oct 2008 23:27:48 +0000 (16:27 -0700)]
kclient: allow monitor port to be specified with the ip(s) during mount

16 years agokclient: include Documentation/filesystems/ceph.txt in patch series
Sage Weil [Tue, 21 Oct 2008 22:08:46 +0000 (15:08 -0700)]
kclient: include Documentation/filesystems/ceph.txt in patch series

16 years agokclient: some changes for 2.6.28
Yehuda Sadeh [Tue, 21 Oct 2008 22:16:15 +0000 (15:16 -0700)]
kclient: some changes for 2.6.28

16 years agokclient: added crush.h for the git import tool
Yehuda Sadeh [Tue, 21 Oct 2008 00:00:12 +0000 (17:00 -0700)]
kclient: added crush.h for the git import tool

16 years agoMerge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Mon, 20 Oct 2008 23:45:30 +0000 (16:45 -0700)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable

16 years agokclient: add dput to getattr for do_lookup splice case
Sage Weil [Mon, 20 Oct 2008 23:55:49 +0000 (16:55 -0700)]
kclient: add dput to getattr for do_lookup splice case

If do_lookup returns a dentry, it has a new ref on it.  Make sure we
dput() that, since we don't actually need it.  (In contrast, if there is no
splice, do_lookup does not dget, as the caller has a ref on the dentry it
provided in the first place.)