]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
15 years agorgw: call url_decode in more appropriate places
Greg Farnum [Fri, 2 Apr 2010 21:21:16 +0000 (14:21 -0700)]
rgw: call url_decode in more appropriate places

15 years agorgw: convert + to space in url_decode
Greg Farnum [Fri, 2 Apr 2010 20:14:12 +0000 (13:14 -0700)]
rgw: convert + to space in url_decode

15 years agorgw: send an error response if reading an ACL off disk fails
Greg Farnum [Wed, 31 Mar 2010 23:06:45 +0000 (16:06 -0700)]
rgw: send an error response if reading an ACL off disk fails

15 years agorgw: remove bad local variable shadowing class member variable
Greg Farnum [Wed, 31 Mar 2010 18:32:40 +0000 (11:32 -0700)]
rgw: remove bad local variable shadowing class member variable

15 years agorgw: remove some invalid FIXMEs
Greg Farnum [Wed, 31 Mar 2010 18:32:08 +0000 (11:32 -0700)]
rgw: remove some invalid FIXMEs

15 years agomds: set laggy flag if dispatch takes too long
Sage Weil [Thu, 1 Apr 2010 23:13:41 +0000 (16:13 -0700)]
mds: set laggy flag if dispatch takes too long

This prevents, e.g., tick() from coming in immediately after and
closing out client sessions after we chew through something particularly
slow.

15 years agomsgr: set OPEN state after accepting connection
Sage Weil [Thu, 1 Apr 2010 20:46:26 +0000 (13:46 -0700)]
msgr: set OPEN state after accepting connection

Not doing so can eventually lead to

msg/SimpleMessenger.cc: In function 'int SimpleMessenger::Pipe::accept()':
msg/SimpleMessenger.cc:765: FAILED assert(existing->state == STATE_CONNECTING)

15 years agomds: fix file recovery
Sage Weil [Thu, 1 Apr 2010 20:06:51 +0000 (13:06 -0700)]
mds: fix file recovery

This was broken by commit d5574993f.

15 years agomds: remove dir from 'new' list on any commit, not just on clean
Sage Weil [Thu, 1 Apr 2010 20:03:04 +0000 (13:03 -0700)]
mds: remove dir from 'new' list on any commit, not just on clean

A dir may be redirtied after the commit, such that it never becomes clean.
It only needs to stay on the 'new' list until it's been written to disk
at least once, though.

15 years agofilejournal: fix journal full handling
Sage Weil [Thu, 1 Apr 2010 18:39:16 +0000 (11:39 -0700)]
filejournal: fix journal full handling

We would block on journal full, but then try to continue where we left off,
which broke thoroughly.  Add return codes, and wait at the proper times:

 - if journal if full on first event, wait. otherwise, write what we have
   so far.
 - wait in write_thread_entry(), not check_for_full().

Also fix up 'room' calculation.

15 years agoqa: make snaptest trigger capsnap dirty(caps)==0 bug
Sage Weil [Thu, 1 Apr 2010 16:44:04 +0000 (09:44 -0700)]
qa: make snaptest trigger capsnap dirty(caps)==0 bug

Prior to ceph-client.git commit 819ccbfa448, this will cause a

VFS: Busy inodes after unmount of ceph. Self-destruct in 5 seconds.  Have a nice day...

on umount.

15 years agomds: remove dentry AND inode when dropping snap metadata; add helper
Sage Weil [Thu, 1 Apr 2010 14:34:24 +0000 (07:34 -0700)]
mds: remove dentry AND inode when dropping snap metadata; add helper

We should only drop obsolete snapped metadata when it is unreferenced, and
at that point we need to drop the dentry AND inode, not just the dentry.
This delays things until caps are released, among other things.

15 years agoosd: improve print of snapc in MOSDOp
Sage Weil [Thu, 1 Apr 2010 14:28:41 +0000 (07:28 -0700)]
osd: improve print of snapc in MOSDOp

15 years agoosd: fix snap trimmer SnapSet adjustment
Sage Weil [Thu, 1 Apr 2010 14:27:59 +0000 (07:27 -0700)]
osd: fix snap trimmer SnapSet adjustment

AFAICS it's totally wrong?

15 years agotodo: updates
Sage Weil [Wed, 31 Mar 2010 22:30:29 +0000 (15:30 -0700)]
todo: updates

15 years agoosd: improve snap trimmer debug output
Sage Weil [Wed, 31 Mar 2010 22:30:17 +0000 (15:30 -0700)]
osd: improve snap trimmer debug output

15 years agomds: clear old snap_trace before building new one
Sage Weil [Wed, 31 Mar 2010 22:00:23 +0000 (15:00 -0700)]
mds: clear old snap_trace before building new one

Otherwise we just build up a big stack of historical information and send
it all to the client.

15 years agoqa: make snaptest.sh repeatable in same dir
Sage Weil [Wed, 31 Mar 2010 21:46:20 +0000 (14:46 -0700)]
qa: make snaptest.sh repeatable in same dir

15 years agoMerge branch 'mds' into unstable
Sage Weil [Wed, 31 Mar 2010 16:49:55 +0000 (09:49 -0700)]
Merge branch 'mds' into unstable

15 years agotodo: kclient snap bug
Sage Weil [Wed, 31 Mar 2010 04:27:50 +0000 (21:27 -0700)]
todo: kclient snap bug

15 years agomds: don't adjust subtree map in rename_prepare
Sage Weil [Wed, 31 Mar 2010 04:27:14 +0000 (21:27 -0700)]
mds: don't adjust subtree map in rename_prepare

Not sure what the reasoning behind this was.

This code is from pre git history, and the git->subversion conversion
managed to make pretty git-blame unusable.  I doubt I really documented
what its purpose at that point was anyway.

15 years agomds: subtree map should always reflect current hierarchy
Sage Weil [Wed, 31 Mar 2010 04:25:46 +0000 (21:25 -0700)]
mds: subtree map should always reflect current hierarchy

Not projected!

15 years agoqa: drop numeric prefix from workunits
Sage Weil [Wed, 31 Mar 2010 04:24:45 +0000 (21:24 -0700)]
qa: drop numeric prefix from workunits

15 years agoosd: don't filter out shutdown msg to self
Sage Weil [Tue, 30 Mar 2010 20:40:41 +0000 (13:40 -0700)]
osd: don't filter out shutdown msg to self

15 years agomsgr: init local_pipe peer_{addr,type}
Sage Weil [Tue, 30 Mar 2010 20:19:30 +0000 (13:19 -0700)]
msgr: init local_pipe peer_{addr,type}

15 years agoMerge branch 'mds' into unstable
Sage Weil [Tue, 30 Mar 2010 19:22:37 +0000 (12:22 -0700)]
Merge branch 'mds' into unstable

15 years agomds: todo rejoin bug
Sage Weil [Tue, 30 Mar 2010 19:14:03 +0000 (12:14 -0700)]
mds: todo rejoin bug

15 years agotodo: kclient bug
Sage Weil [Tue, 30 Mar 2010 19:13:33 +0000 (12:13 -0700)]
todo: kclient bug

15 years agorgw: revert accidental commit of get_attr prototype change
Greg Farnum [Tue, 30 Mar 2010 18:48:06 +0000 (11:48 -0700)]
rgw: revert accidental commit of get_attr prototype change

15 years agoMerge branch 'msgr' into unstable
Sage Weil [Tue, 30 Mar 2010 18:45:00 +0000 (11:45 -0700)]
Merge branch 'msgr' into unstable

15 years agomsgr: source, not orig_source, in dbg output
Sage Weil [Fri, 26 Mar 2010 19:31:39 +0000 (12:31 -0700)]
msgr: source, not orig_source, in dbg output

15 years agomds: don't use get_orig_source
Sage Weil [Thu, 25 Mar 2010 23:09:42 +0000 (16:09 -0700)]
mds: don't use get_orig_source

MDS doesn't do any request forwarding, so there is no need.

15 years agomsgr: NOSRCADDR protocol feature drops src/orig_src in msg header
Sage Weil [Thu, 25 Mar 2010 23:04:18 +0000 (16:04 -0700)]
msgr: NOSRCADDR protocol feature drops src/orig_src in msg header

This lets us drop the src, orig_src ceph_entity_addr's from the message
header, saving about 160 bytes per message.

The feature is optional.  We can still talk to peers who use the old
protocol.

15 years agomsgr: put features in connection_state
Sage Weil [Thu, 25 Mar 2010 23:01:34 +0000 (16:01 -0700)]
msgr: put features in connection_state

15 years agomsgr: use connection for src addr
Sage Weil [Thu, 25 Mar 2010 21:49:56 +0000 (14:49 -0700)]
msgr: use connection for src addr

15 years agomsgr: make Message::get_orig_* differ only when explicitly directed to
Sage Weil [Thu, 25 Mar 2010 20:46:16 +0000 (13:46 -0700)]
msgr: make Message::get_orig_* differ only when explicitly directed to

This paves the way for removal of the orig_src field from the message
header.

15 years agomds: fix MDSTableClient ack double journaling
Sage Weil [Tue, 30 Mar 2010 17:30:02 +0000 (10:30 -0700)]
mds: fix MDSTableClient ack double journaling

Do not journal ack unless the tid is registered in the LogSegment.  Once
we journal it, we remove it from the LogSegment list, and once it's
journaled, we remove the pending_commit[tid] entry.

This fixes a bug where the mds got two acks, journaled both of them, and
crashed in the completion for the second because pending_commit[tid] was
gone.  The second ack should have been ignored.

15 years agomds: start file recovery after sending rejoin ack
Sage Weil [Mon, 29 Mar 2010 23:26:54 +0000 (16:26 -0700)]
mds: start file recovery after sending rejoin ack

The rejoin ack intializes replica lock states correctly; we can't send any
lock messages before that.  This fixes both the check max size call (which
sends lock messages taking the wrlock) and the file_recover() call
(which does the same).

Instead, we make two lists, files to recover and those to fix up.  The lock
states for both are set to PRE_SCAN (LOCK on replica).  After the rejoin
acks go out, we either check_inode_max_size or file_recover.

If file_recover someday grows another caller, this may need something a bit
more sophisticated.

15 years agomds: delay check_inode_max_size if frozen
Sage Weil [Mon, 29 Mar 2010 22:52:51 +0000 (15:52 -0700)]
mds: delay check_inode_max_size if frozen

15 years agomds: less noisy about snaprealm siblings
Sage Weil [Mon, 29 Mar 2010 21:34:52 +0000 (14:34 -0700)]
mds: less noisy about snaprealm siblings

This can get _very_ large for the stray dir.

15 years agorgw: add optional auid parameter to create_bucket.
Greg Farnum [Mon, 29 Mar 2010 18:43:45 +0000 (11:43 -0700)]
rgw: add optional auid parameter to create_bucket.

If set this will set the RADOS pool's auid to the given one, granting
partial data coherency.

15 years agorgw: remove dead code. rebuild_policy exists elsewhere; the acl needs no main()
Greg Farnum [Mon, 29 Mar 2010 17:03:29 +0000 (10:03 -0700)]
rgw: remove dead code. rebuild_policy exists elsewhere; the acl needs no main()

15 years agorgw: radosgw_admin will quit gracefully if it can't open the store
Greg Farnum [Mon, 29 Mar 2010 14:40:51 +0000 (07:40 -0700)]
rgw: radosgw_admin will quit gracefully if it can't open the store

15 years agorgw: set auid if specified at creation
Greg Farnum [Mon, 29 Mar 2010 14:40:28 +0000 (07:40 -0700)]
rgw: set auid if specified at creation

15 years agorgw: add auid field to RGWUserInfo
Greg Farnum [Fri, 26 Mar 2010 23:29:11 +0000 (16:29 -0700)]
rgw: add auid field to RGWUserInfo

15 years agoauth: Keyring no longer segfaults if HOME env variable is unset.
Greg Farnum [Fri, 26 Mar 2010 22:57:14 +0000 (15:57 -0700)]
auth: Keyring no longer segfaults if HOME env variable is unset.

Just skip that path, instead. We can't recover what it meant.

15 years agorgw: if you can't init_storage_provider(), quit instead
Greg Farnum [Fri, 26 Mar 2010 22:56:37 +0000 (15:56 -0700)]
rgw: if you can't init_storage_provider(), quit instead
of segfaulting later.

15 years agorgw: say why the store failed to open in admin
Greg Farnum [Fri, 26 Mar 2010 22:56:14 +0000 (15:56 -0700)]
rgw: say why the store failed to open in admin

15 years agorgw: add minor data to schema examination
Greg Farnum [Fri, 26 Mar 2010 22:55:47 +0000 (15:55 -0700)]
rgw: add minor data to schema examination

15 years agomkcephfs: osd=allow * caps for osd, mds
Sage Weil [Mon, 29 Mar 2010 16:48:28 +0000 (09:48 -0700)]
mkcephfs: osd=allow * caps for osd, mds

15 years agomds: migrate frag/nest scatterlock info on bounding frags during export
Sage Weil [Fri, 26 Mar 2010 23:01:35 +0000 (16:01 -0700)]
mds: migrate frag/nest scatterlock info on bounding frags during export

This ensures that the auth inode continues to maintain accurate scatterlock
info about open frags.  We include info on export if it is a bounding frag.
On import, we only take it if we are !auth.  This mirrors the scatterlock
scatter/gather logic in CInode::{encode,decode}_lock_state.

15 years agoauth: standardize on -n/--name [type.]name
Sage Weil [Fri, 26 Mar 2010 23:04:26 +0000 (16:04 -0700)]
auth: standardize on -n/--name [type.]name

15 years agoauth: initialized uninitialized variable, don't free pts we don't own
Sage Weil [Fri, 26 Mar 2010 22:46:52 +0000 (15:46 -0700)]
auth: initialized uninitialized variable, don't free pts we don't own

15 years agomds: make rdlock waiter wait on STABLE | RD, not just RD, when lock is unstable
Sage Weil [Fri, 26 Mar 2010 21:12:42 +0000 (14:12 -0700)]
mds: make rdlock waiter wait on STABLE | RD, not just RD, when lock is unstable

15 years agomds: use 32 bits for MDSCacheObject::ref count
Sage Weil [Fri, 26 Mar 2010 20:35:42 +0000 (13:35 -0700)]
mds: use 32 bits for MDSCacheObject::ref count

16 is not enough!  :)

15 years agoMerge branch 'unstable' into mds
Sage Weil [Fri, 26 Mar 2010 20:22:04 +0000 (13:22 -0700)]
Merge branch 'unstable' into mds

Conflicts:
src/TODO

15 years agomkcephf: behave when user is not defined
Sage Weil [Fri, 26 Mar 2010 19:09:18 +0000 (12:09 -0700)]
mkcephf: behave when user is not defined

15 years agoauth: fix a few memory leaks, uninit vars
Sage Weil [Thu, 25 Mar 2010 19:51:05 +0000 (12:51 -0700)]
auth: fix a few memory leaks, uninit vars

Courtesy of cppcheck

15 years agomds: skip purge_stray when stray pins are present
Sage Weil [Fri, 26 Mar 2010 18:56:34 +0000 (11:56 -0700)]
mds: skip purge_stray when stray pins are present

Otherwise we can get into a loop:

 eval_stray
 purge_stray
 _purge_stray_purged
  journal truncate
 eval_stray
 ...

This doesn't explain what the stray pin was that I saw, but it at least
avoids the looping.

15 years agoMakefile: add missing MDentryLink.h
Sage Weil [Thu, 25 Mar 2010 19:00:34 +0000 (12:00 -0700)]
Makefile: add missing MDentryLink.h

15 years agomds: drop unused CInode::PIN_PURGING
Sage Weil [Fri, 26 Mar 2010 18:54:30 +0000 (11:54 -0700)]
mds: drop unused CInode::PIN_PURGING

15 years agosrc/README is required
Sage Weil [Thu, 25 Mar 2010 18:48:25 +0000 (11:48 -0700)]
src/README is required

15 years agofilestore: less noisy about OpSequencer queue
Sage Weil [Thu, 25 Mar 2010 18:31:05 +0000 (11:31 -0700)]
filestore: less noisy about OpSequencer queue

15 years agomds: drop obsolete hack for base inodes
Sage Weil [Thu, 25 Mar 2010 17:58:21 +0000 (10:58 -0700)]
mds: drop obsolete hack for base inodes

We used to skip base inodes for scatter_writebehind.  But we can
journal these just like anything else, and it potentially breaks
try_to_expire if a base inode's lock is dirty, because the
completion queued on WAIT_STABLE by scatter_nudge never gets
completed.

15 years agomds: fix lost completion in scatter_nudge (broke log trimming)
Sage Weil [Thu, 25 Mar 2010 17:54:07 +0000 (10:54 -0700)]
mds: fix lost completion in scatter_nudge (broke log trimming)

15 years agorados_bencher: Don't re-initialize passed-in pool, it breaks stuff
Greg Farnum [Thu, 25 Mar 2010 22:39:58 +0000 (15:39 -0700)]
rados_bencher: Don't re-initialize passed-in pool, it breaks stuff

15 years agoosd: Fix up OSDCaps::get_pool_cap to work/make more sense
Greg Farnum [Thu, 25 Mar 2010 22:38:52 +0000 (15:38 -0700)]
osd: Fix up OSDCaps::get_pool_cap to work/make more sense

15 years agorados: add interface to change pool owners
Greg Farnum [Tue, 23 Mar 2010 22:08:13 +0000 (15:08 -0700)]
rados: add interface to change pool owners

15 years agolibrados: add change_pool_auid functions
Greg Farnum [Tue, 23 Mar 2010 20:41:01 +0000 (13:41 -0700)]
librados: add change_pool_auid functions

15 years agoobjecter: add change_pool_auid function.
Greg Farnum [Tue, 23 Mar 2010 20:38:02 +0000 (13:38 -0700)]
objecter: add change_pool_auid function.

I'm reluctant to stick this in the objecter since it doesn't quite fit, but
it's a pool management function and putting it here makes it easy to use
elsewhere while maintaining librados' standard function flow.

15 years agomds: allow rdlock on replica to request sync from auth
Sage Weil [Thu, 25 Mar 2010 20:17:06 +0000 (13:17 -0700)]
mds: allow rdlock on replica to request sync from auth

This fixes a ping-pong readdir between inode and dirfrag auth.

15 years agomon: Set MForward::client_caps explicitly; this fixes a bad connection deref.
Greg Farnum [Thu, 25 Mar 2010 19:58:14 +0000 (12:58 -0700)]
mon: Set MForward::client_caps explicitly; this fixes a bad connection deref.

Previously the decoded PaxosServiceMessage would attempt to follow an
uninitialized Connection* and segfault the system.

15 years agomon: set mon_caps with authenticated monitor check instead of claimed one.
Greg Farnum [Wed, 24 Mar 2010 23:07:51 +0000 (16:07 -0700)]
mon: set mon_caps with authenticated monitor check instead of claimed one.

15 years agomon: Fix all uses of Session and get_caps.
Greg Farnum [Thu, 25 Mar 2010 17:43:04 +0000 (10:43 -0700)]
mon: Fix all uses of Session and get_caps.
They now use MonSession and get_service()->caps, respectively.

15 years agomon: Change name of Session to MonSession.
Greg Farnum [Thu, 25 Mar 2010 17:41:31 +0000 (10:41 -0700)]
mon: Change name of Session to MonSession.

15 years agomsg: Add get_session function to PaxosServiceMessage
Greg Farnum [Thu, 25 Mar 2010 17:11:20 +0000 (10:11 -0700)]
msg: Add get_session function to PaxosServiceMessage

15 years agoRevert "msg: PaxosServiceMessage includes a MonCap*"
Greg Farnum [Thu, 25 Mar 2010 16:54:58 +0000 (09:54 -0700)]
Revert "msg: PaxosServiceMessage includes a MonCap*"

This reverts commit 167b28a0d4cb2977f237be1972e40e04aed24207.

15 years agomon: MonCaps default initializes more variables
Greg Farnum [Wed, 24 Mar 2010 20:54:20 +0000 (13:54 -0700)]
mon: MonCaps default initializes more variables

15 years agoosd: drop useless ENOMEM check
Sage Weil [Thu, 25 Mar 2010 17:19:43 +0000 (10:19 -0700)]
osd: drop useless ENOMEM check

'new' will throw an exception if it fails, not return NULL.

15 years agovstart.sh: include mon secret during mkfs
Sage Weil [Thu, 25 Mar 2010 17:19:25 +0000 (10:19 -0700)]
vstart.sh: include mon secret during mkfs

15 years agomon: fix up authorizer generation/validation
Sage Weil [Thu, 25 Mar 2010 17:19:02 +0000 (10:19 -0700)]
mon: fix up authorizer generation/validation

Only generate authorizer for other mons. Only require authorizer for other
mons when CEPHX is enabled.  Check it properly (before we were letting
mons connect without any authorizer!).

15 years agomds: fix bug in handle_discover on mdsdir
Sage Weil [Tue, 23 Mar 2010 18:47:01 +0000 (11:47 -0700)]
mds: fix bug in handle_discover on mdsdir

15 years agomds: check myin, not stray, in rejoin_send_acks
Sage Weil [Tue, 23 Mar 2010 18:46:49 +0000 (11:46 -0700)]
mds: check myin, not stray, in rejoin_send_acks

15 years agomon: change forwarding behavior somewhat
Sage Weil [Wed, 24 Mar 2010 22:50:31 +0000 (15:50 -0700)]
mon: change forwarding behavior somewhat

15 years agomon: initialize routed_request_tid
Sage Weil [Wed, 24 Mar 2010 22:24:07 +0000 (15:24 -0700)]
mon: initialize routed_request_tid

15 years agomsgr: set m->connection for messages sent to self
Sage Weil [Wed, 24 Mar 2010 21:04:49 +0000 (14:04 -0700)]
msgr: set m->connection for messages sent to self

15 years agoosd: detect TMAPUP no-op and do no work
Sage Weil [Wed, 24 Mar 2010 18:43:57 +0000 (11:43 -0700)]
osd: detect TMAPUP no-op and do no work

15 years agodebug: include timestamp + thread in _all_ debug output
Sage Weil [Wed, 24 Mar 2010 18:54:44 +0000 (11:54 -0700)]
debug: include timestamp + thread in _all_ debug output

15 years agoqa: fix loopall.sh mode
Sage Weil [Wed, 24 Mar 2010 18:42:56 +0000 (11:42 -0700)]
qa: fix loopall.sh mode

15 years agotodo
Sage Weil [Wed, 24 Mar 2010 18:50:17 +0000 (11:50 -0700)]
todo

15 years agotodo mds bug
Sage Weil [Tue, 23 Mar 2010 23:47:59 +0000 (16:47 -0700)]
todo mds bug

15 years agomon: Hack in setting monitor caps
Greg Farnum [Tue, 23 Mar 2010 22:25:58 +0000 (15:25 -0700)]
mon: Hack in setting monitor caps

15 years agofix up READMEs
Sage Weil [Tue, 23 Mar 2010 19:24:34 +0000 (12:24 -0700)]
fix up READMEs

15 years agomds: fix up client session importing
Sage Weil [Tue, 23 Mar 2010 21:54:13 +0000 (14:54 -0700)]
mds: fix up client session importing

Keep import counter for each session, for overlapping imports.

Prevent a session close or kill during an import.

15 years agomon: Fix initialization-crossing goto jump
Greg Farnum [Tue, 23 Mar 2010 21:03:40 +0000 (14:03 -0700)]
mon: Fix initialization-crossing goto jump

15 years agomds: importing, ugh
Sage Weil [Tue, 23 Mar 2010 20:39:38 +0000 (13:39 -0700)]
mds: importing, ugh

15 years agomds: clean up kill_session()
Sage Weil [Tue, 23 Mar 2010 20:32:58 +0000 (13:32 -0700)]
mds: clean up kill_session()

15 years agomds: possible session 'importing' state
Sage Weil [Tue, 23 Mar 2010 20:20:12 +0000 (13:20 -0700)]
mds: possible session 'importing' state

15 years agomds: drop purge step when closing client sessions
Sage Weil [Tue, 23 Mar 2010 20:15:58 +0000 (13:15 -0700)]
mds: drop purge step when closing client sessions

The purge step when closing client sessions is pointless.  If the
MDS is alive, we know which inos were used and which weren't.  It
is only the _journal_ that doesn't necessarily know which have
been recently allocated: any cleanup needs to happen on journal
replay during mds recovery.

Also clean up session states some more.

15 years agomon: Removing dead code; been commented out for 5 months
Greg Farnum [Tue, 23 Mar 2010 19:40:55 +0000 (12:40 -0700)]
mon: Removing dead code; been commented out for 5 months