Commit graph

95 commits

Author SHA1 Message Date
Linus Torvalds
b12ece7d85 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  ceph: avoid picking MDS that is not active
  ceph: avoid immediate cap check after import
  ceph: fix flushing of caps vs cap import
  ceph: fix erroneous cap flush to non-auth mds
  ceph: fix cap_wanted_delay_{min,max} mount option initialization
  ceph: fix xattr rbtree search
  ceph: fix getattr on directory when using norbytes
2011-01-28 12:12:58 +10:00
Sage Weil
d66bbd441c ceph: avoid picking MDS that is not active
Ignore replication or auth frag data if it indicates an MDS that is not
active.  This can happen if the MDS shuts down and the client has stale
data about the namespace distribution across the MDS cluster.  If that's
the case, fall back to directing the request based on the auth cap (which
should always be accurate).

Signed-off-by: Sage Weil <sage@newdream.net>
2011-01-25 08:16:37 -08:00
Linus Torvalds
a170315420 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  rbd: fix cleanup when trying to mount inexistent image
  net/ceph: make ceph_msgr_wq non-reentrant
  ceph: fsc->*_wq's aren't used in memory reclaim path
  ceph: Always free allocated memory in osdmap_decode()
  ceph: Makefile: Remove unnessary code
  ceph: associate requests with opening sessions
  ceph: drop redundant r_mds field
  ceph: implement DIRLAYOUTHASH feature to get dir layout from MDS
  ceph: add dir_layout to inode
2011-01-13 10:25:24 -08:00
Sage Weil
dc69e2e9fc ceph: associate requests with opening sessions
Associate request with sessions that aren't yep open.  This makes the
debugfs mdsc request list more informative.

Signed-off-by: Sage Weil <sage@newdream.net>
2011-01-12 15:15:13 -08:00
Sage Weil
4af25fdda6 ceph: drop redundant r_mds field
The r_mds field is redundant, since we can find the same information at
r_session->s_mds, and when r_session is NULL then r_mds is meaningless.

Signed-off-by: Sage Weil <sage@newdream.net>
2011-01-12 15:15:13 -08:00
Sage Weil
14303d20f3 ceph: implement DIRLAYOUTHASH feature to get dir layout from MDS
This implements the DIRLAYOUTHASH protocol feature, which passes the dir
layout over the wire from the MDS.  This gives the client knowledge
of the correct hash function to use for mapping dentries among dir
fragments.

Note that if this feature is _not_ present on the client but is on the
MDS, the client may misdirect requests.  This will result in a forward
and degrade performance.  It may also result in inaccurate NFS filehandle
generation, which will prevent fh resolution when the inode is not present
in the client cache and the parent directories have been fragmented.

Signed-off-by: Sage Weil <sage@newdream.net>
2011-01-12 15:15:13 -08:00
Nick Piggin
b7ab39f631 fs: dcache scale dentry refcount
Make d_count non-atomic and protect it with d_lock. This allows us to ensure a
0 refcount dentry remains 0 without dcache_lock. It is also fairly natural when
we start protecting many other dentry members with d_lock.

Signed-off-by: Nick Piggin <npiggin@kernel.dk>
2011-01-07 17:50:21 +11:00
Herb Shiu
25933abdd8 ceph: Handle file locks in replies from the MDS.
Previously the kernel client incorrectly assumed everything was a directory.

Signed-off-by: Herb Shiu <herb_shiu@tcloudcomputing.com>
Acked-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-12-01 14:22:27 -08:00
Linus Torvalds
76db8ac45f Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  ceph: fix readdir EOVERFLOW on 32-bit archs
  ceph: fix frag offset for non-leftmost frags
  ceph: fix dangling pointer
  ceph: explicitly specify page alignment in network messages
  ceph: make page alignment explicit in osd interface
  ceph: fix comment, remove extraneous args
  ceph: fix update of ctime from MDS
  ceph: fix version check on racing inode updates
  ceph: fix uid/gid on resent mds requests
  ceph: fix rdcache_gen usage and invalidate
  ceph: re-request max_size if cap auth changes
  ceph: only let auth caps update max_size
  ceph: fix open for write on clustered mds
  ceph: fix bad pointer dereference in ceph_fill_trace
  ceph: fix small seq message skipping
  Revert "ceph: update issue_seq on cap grant"
2010-11-19 15:32:22 -08:00
Arnd Bergmann
451a3c24b0 BKL: remove extraneous #include <smp_lock.h>
The big kernel lock has been removed from all these files at some point,
leaving only the #include.

Remove this too as a cleanup.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-11-17 08:59:32 -08:00
Sage Weil
cb4276cca4 ceph: fix uid/gid on resent mds requests
MDS requests can be rebuilt and resent in non-process context, but were
filling in uid/gid from current_fsuid/gid.  Put that information in the
request struct on request setup.

This fixes incorrect (and root) uid/gid getting set for requests that
are forwarded between MDSs, usually due to metadata migrations.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-11-08 07:29:05 -08:00
Sage Weil
496e59553c ceph: switch from BKL to lock_flocks()
Switch from using the BKL explicitly to the new lock_flocks() interface.
Eventually this will turn into a spinlock.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-10-20 15:38:18 -07:00
Greg Farnum
fca4451acf ceph: preallocate flock state without locks held
When the lock_kernel() turns into lock_flocks() and a spinlock, we won't
be able to do allocations with the lock held.  Preallocate space without
the lock, and retry if the lock state changes out from underneath us.

Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-10-20 15:38:17 -07:00
Yehuda Sadeh
3d14c5d2b6 ceph: factor out libceph from Ceph file system
This factors out protocol and low-level storage parts of ceph into a
separate libceph module living in net/ceph and include/linux/ceph.  This
is mostly a matter of moving files around.  However, a few key pieces
of the interface change as well:

 - ceph_client becomes ceph_fs_client and ceph_client, where the latter
   captures the mon and osd clients, and the fs_client gets the mds client
   and file system specific pieces.
 - Mount option parsing and debugfs setup is correspondingly broken into
   two pieces.
 - The mon client gets a generic handler callback for otherwise unknown
   messages (mds map, in this case).
 - The basic supported/required feature bits can be expanded (and are by
   ceph_fs_client).

No functional change, aside from some subtle error handling cases that got
cleaned up in the refactoring process.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-10-20 15:37:28 -07:00
Sage Weil
3612abbd5d ceph: fix reconnect encoding for old servers
Fix the reconnect encoding to encode the cap record when the MDS does not
have the FLOCK capability (i.e., pre v0.22).

Signed-off-by: Sage Weil <sage@newdream.net>
2010-09-11 10:52:47 -07:00
Sage Weil
e072f8aa35 ceph: don't BUG on ENOMEM during mds reconnect
We are in a position to return an error; do that instead.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-26 09:26:37 -07:00
Sage Weil
eb6bb1c5bd ceph: direct requests in snapped namespace based on nonsnap parent
When making a request in the virtual snapdir or a snapped portion of the
namespace, we should choose the MDS based on the first nonsnap parent (and
its caps).  If that is not the best place, we will get forward hints to
find the right MDS in the cluster.  This fixes ESTALE errors when using
the .snap directory and namespace with multiple MDSs.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22 15:16:48 -07:00
Sage Weil
f3c60c5918 ceph: fix multiple mds session shutdown
The use of a completion when waiting for session shutdown during umount is
inappropriate, given the complexity of the condition.  For multiple MDS's,
this resulted in the umount thread spinning, often preventing the session
close message from being processed in some cases.

Switch to a waitqueue and defined a condition helper.  This cleans things
up nicely.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22 15:04:43 -07:00
Sage Weil
213c99ee0c ceph: whitespace cleanup
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-03 10:25:11 -07:00
Greg Farnum
40819f6fb2 ceph: add flock/fcntl lock support
Implement flock inode operation to support advisory file locking.  All
lock/unlock operations are synchronous with the MDS.  Lock state is
sent when reconnecting to a recovering MDS to restore the shared lock
state.

Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02 16:10:53 -07:00
Sage Weil
20cb34ae9e ceph: support v2 reconnect encoding
Encode either old or v2 encoding of client_reconnect message, depending on
whether the peer has the FLOCK feature bit.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02 15:48:50 -07:00
Greg Farnum
e55b71f802 ceph: handle ESTALE properly; on receipt send to authority if it wasn't
Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:41 -07:00
Sage Weil
154f42c2c3 ceph: connect to export targets on cap export
When we get a cap EXPORT message, make sure we are connected to all export
targets to ensure we can handle the matching IMPORT.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:41 -07:00
Sage Weil
cb170a2215 ceph: connect to export targets if mds is laggy
If an MDS we are talking to may have failed, we need to open sessions to
its potential export targets to ensure that any in-progress migration that
may have involved some of our caps is properly handled.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:40 -07:00
Sage Weil
ed0552a1a2 ceph: introduce helper to connect to mds export targets
There are a few cases where we need to open sessions with a given mds's
potential export targets.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:40 -07:00
Yehuda Sadeh
37151668ba ceph: do caps accounting per mds_client
Caps related accounting is now being done per mds client instead
of just being global. This prepares ground work for a later revision
of the caps preallocated reservation list.

Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:40 -07:00
Sage Weil
0deb01c999 ceph: track laggy state of mds from mdsmap
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:40 -07:00
Sage Weil
38e8883ee3 ceph: simplify add_cap_releases
No functional change, aside from more useful debug output.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:39 -07:00
Sage Weil
ee6b272b9c ceph: drop unused argument
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01 20:11:39 -07:00
Yehuda Sadeh
03066f2345 ceph: use complete_all and wake_up_all
This fixes an issue triggered by running concurrent syncs. One of the syncs
would go through while the other would just hang indefinitely. In any case, we
never actually want to wake a single waiter, so the *_all functions should
be used.

Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-07-27 13:11:17 -07:00
Sage Weil
e979cf5039 ceph: do not include cap/dentry releases in replayed messages
Strip the cap and dentry releases from replayed messages.  They can
cause the shared state to get out of sync because they were generated
(with the request message) earlier, and no longer reflect the current
client state.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-07-16 10:30:18 -07:00
Sage Weil
01a92f174f ceph: reuse request message when replaying against recovering mds
Replayed rename operations (after an mds failure/recovery) were broken
because the request paths were regenerated from the dentry names, which
get mangled when d_move() is called.

Instead, resend the previous request message when replaying completed
operations.  Just make sure the REPLAY flag is set and the target ino is
filled in.

This fixes problems with workloads doing renames when the MDS restarts,
where the rename operation appears to succeed, but on mds restart then
fails (leading to client confusion, app breakage, etc.).

Signed-off-by: Sage Weil <sage@newdream.net>
2010-07-16 10:30:17 -07:00
Sage Weil
17c688c3df ceph: delay umount until all mds requests drop inode+dentry refs
This fixes a race between handle_reply finishing an mds request, signalling
completion, and then dropping the request structing and its dentry+inode
refs, and pre_umount function waiting for requests to finish before
letting the vfs tear down the dcache.  If umount was delayed waiting for
mds requests, we could race and BUG in shrink_dcache_for_umount_subtree
because of a slow dput.

This delays umount until the msgr queue flushes, which means handle_reply
will exit and will have dropped the ceph_mds_request struct.  I'm assuming
the VFS has already ensured that its calls have all completed and those
request refs have thus been dropped as well (I haven't seen that race, at
least).

Signed-off-by: Sage Weil <sage@newdream.net>
2010-06-21 16:11:50 -07:00
Sage Weil
2b2300d62e ceph: try to send partial cap release on cap message on missing inode
If we have enough memory to allocate a new cap release message, do so, so
that we can send a partial release message immediately.  This keeps us from
making the MDS wait when the cap release it needs is in a partially full
release message.

If we fail because of ENOMEM, oh well, they'll just have to wait a bit
longer.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-06-10 13:30:25 -07:00
Sage Weil
3d7ded4d81 ceph: release cap on import if we don't have the inode
If we get an IMPORT that give us a cap, but we don't have the inode, queue
a release (and try to send it immediately) so that the MDS doesn't get
stuck waiting for us.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-06-10 13:30:07 -07:00
Sage Weil
1e5ea23df1 ceph: fix lease revocation when seq doesn't match
If the client revokes a lease with a higher seq than what we have, keep
the mds's seq, so that it honors our release.  Otherwise, we can hang
indefinitely.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-06-04 10:05:40 -07:00
Sage Weil
2a8e5e3637 ceph: clean up on forwarded aborted mds request
If an mds request is aborted (timeout, SIGKILL), it is left registered to
keep our state in sync with the mds.  If we get a forward notification,
though, we know the request didn't succeed and we can unregister it
safely.  We were trying to resend it, but then bailing out (and not
unregistering) in __do_request.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-29 09:42:05 -07:00
Sage Weil
dd1c905736 ceph: make lease code DN specific
The lease code includes a mask in the CEPH_LOCK_* namespace, but that
namespace is changing, and only one mask (formerly _DN == 1) is used, so
hard code for that value for now.

If we ever extend this code to handle leases over different data types we
can extend it accordingly.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-29 09:12:42 -07:00
Sage Weil
aa91647c89 ceph: make mds requests killable, not interruptible
The underlying problem is that many mds requests can't be restarted.  For
example, a restarted create() would return -EEXIST if the original request
succeeds.  However, we do not want a hung MDS to hang the client too.  So,
use the _killable wait_for_completion variants to abort on SIGKILL but
nothing else.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-29 09:12:35 -07:00
Tobias Klauser
9e32789f63 ceph: Storage class should be before const qualifier
The C99 specification states in section 6.11.5:

The placement of a storage-class specifier other than at the beginning
of the declaration specifiers in a declaration is an obsolescent
feature.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-21 15:01:21 -07:00
Yehuda Sadeh
34d23762d9 ceph: all allocation functions should get gfp_mask
This is essential, as for the rados block device we'll need
to run in different contexts that would need flags that
are other than GFP_NOFS.

Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:42 -07:00
Sage Weil
167c9e352d ceph: use common helper for aborted dir request invalidation
We invalidate I_COMPLETE and dentry leases in two places: on aborted mds
request and on request replay.  Use common helper to avoid duplicate code.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:40 -07:00
Sage Weil
85792d0dd6 ceph: cope with out of order (unsafe after safe) mds reply
Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:39 -07:00
Sage Weil
6c99f2545d ceph: throw out dirty caps metadata, data on session teardown
The remove_session_caps() helper is called when an MDS closes out our
session (either normally, or as a result of a failed reconnect), and when
we tear down state for umount.  If we remove the last cap, and there are
no cap migrations in progress, then there is little hope of us flushing
out that data to the mds (without heroic efforts to reconnect and flush).

So, to avoid leaving inodes pinned (due to dirty state) and crashing after
umount, throw out dirty caps state and unpin the inodes.  Print a warning
to the console so we know something was lost.

NOTE: Although we drop wrbuffer refs, we don't actually mark pages clean;
maybe a truncate should be queued?

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:37 -07:00
Sage Weil
7e70f0ed9f ceph: attempt mds reconnect if mds closes our session
Currently, if our session is closed (due to a timeout, or explicit close,
or whatever), we just sit there doing nothing unless/until the MDS
restarts, at which point we try to reconnect.

Change client to attempt an immediate reconnect if our session is closed.

Note that currently the MDS doesn't support this, and our attempt will
fail.  We'll get a session CLOSE, our caps and dirty cap state will be
dropped, and the client will be free to attempt to reconnect.  That's
clearly not as nice as a successful reconnect, but it at least allows us
to try to carry on, and in the future the MDS will support a reconnect
and we will fare better.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:36 -07:00
Sage Weil
34b6c855fa ceph: clean up send_mds_reconnect interface
Pass a ceph_mds_session, since the caller has it.

Remove the dead code for sending empty reconnects.  It used to be used
when the MDS contacted _us_ to solicit a reconnect, and we could reply
saying "go away, I have no session."  Now we only send reconnects based
on the mds map, and only when we do in fact have an open session.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:35 -07:00
Sage Weil
29790f26ab ceph: wait for mds OPEN reply to indicate reconnect success
We used to infer reconnect success by watching the MDS state, essentially
assuming that hearing nothing meant things were ok.  That wasn't
particularly reliable.  Instead, the MDS replies with an explicit OPEN
message to indicate success.

Strictly speaking, this is a protocol change, but it is a backwards
compatible one that does not break new clients + old servers or old
clients + new servers.  At least not yet.

Drop unused @all argument from kick_requests while we're at it.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:35 -07:00
Sage Weil
aab53dd9e8 ceph: only send cap releases when mds is OPEN|HUNG
On OPENING we shouldn't have any caps (or releases).
On CLOSING, we should wait until we succeed (and throw it all out), or
don't (and are OPEN again).
On RECONNECTING we can wait until we are OPEN.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:34 -07:00
Sage Weil
e01a594646 ceph: dicard cap releases on mds restart
If the MDS restarts, the expire caps state is no longer shared, and can be
thrown out.  Caps state will be rebuilt on the MDS during the reconnect
process that follows.  Zero out any release messages and adjust the
release counter accordingly.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:33 -07:00
Sage Weil
0f8605f2bd ceph: clean up cap release loop vs spinlock
Signed-off-by: Sage Weil <sage@newdream.net>
2010-05-17 15:25:31 -07:00