Sage Weil [Mon, 25 Jan 2010 19:33:08 +0000 (11:33 -0800)]
ceph: properly handle aborted mds requests
Previously, if the MDS request was interrupted, we would unregister the
request and ignore any reply. This could cause the caps or other cache
state to become out of sync. (For instance, aborting dbench and doing
rm -r on clients would complain about a non-empty directory because the
client didn't realize it's aborted file create request completed.)
Even we don't unregister, we still can't process the reply normally because
we are no longer holding the caller's locks (like the dir i_mutex).
So, mark aborted operations with r_aborted, and in the reply handler, be
sure to process all the caps. Do not process the namespace changes,
though, since we no longer will hold the dir i_mutex. The dentry lease
state can also be ignored as it's more forgiving.
Yehuda Sadeh [Thu, 7 Jan 2010 23:36:32 +0000 (15:36 -0800)]
ceph: change dentry offset and position after splice_dentry
This fixes a bug, where we had the parent list have dentries with
offsets that are not monotonically increasing, which caused the ceph
dcache_readdir to skip entries.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 23 Dec 2009 20:21:51 +0000 (12:21 -0800)]
ceph: use ceph_pagelist for mds reconnect message; change encoding (protocol change)
Use the ceph_pagelist to encode the MDS reconnect message. We change the
message encoding (protocol change!) at the same time to make our life
easier (we don't know how many snaprealms we have when we start encoding).
An empty message implies the session is closed/does not exist.
Sage Weil [Wed, 23 Dec 2009 20:12:31 +0000 (12:12 -0800)]
ceph: support ceph_pagelist for message payload
The ceph_pagelist is a simple list of whole pages, strung together via
their lru list_head. It facilitates encoding to a "buffer" of unknown
size. Allow its use in place of the ceph_msg page vector.
This will be used to fix the huge buffer preallocation woes of MDS
reconnection.
Sage Weil [Wed, 23 Dec 2009 17:30:21 +0000 (09:30 -0800)]
ceph: add feature bits to connection handshake (protocol change)
Define supported and required feature set. Fail connection if the server
requires features we do not support (TAG_FEATURES), or if the server does
not support features we require.
Sage Weil [Tue, 22 Dec 2009 19:24:33 +0000 (11:24 -0800)]
ceph: include transaction id in ceph_msg_header (protocol change)
Many (most?) message types include a transaction id. By including it in
the fixed size header, we always have it available even when we are unable
to allocate memory for the (larger, variable sized) message body. This
will allow us to error out the appropriate request instead of (silently)
dropping the reply.
Sage Weil [Tue, 22 Dec 2009 18:45:45 +0000 (10:45 -0800)]
ceph: control access to page vector for incoming data
When we issue an OSD read, we specify a vector of pages that the data is to
be read into. The request may be sent multiple times, to multiple OSDs, if
the osdmap changes, which means we can get more than one reply.
Only read data into the page vector if the reply is coming from the
OSD we last sent the request to. Keep track of which connection is using
the vector by taking a reference. If another connection was already
using the vector before and a new reply comes in on the right connection,
revoke the pages from the other connection.
Sage Weil [Tue, 22 Dec 2009 18:43:42 +0000 (10:43 -0800)]
ceph: use connection mutex to protect read and write stages
Use a single mutex (previously out_mutex) to protect both read and write
activity from concurrent ceph_con_* calls. Drop the mutex when doing
callbacks to avoid nested locking (the callback may need to call something
like ceph_con_close).
Sage Weil [Tue, 22 Dec 2009 18:29:39 +0000 (10:29 -0800)]
ceph: unregister canceled/timed out osd requests
Canceled or timed out osd requests were getting left in the request list
and never deallocated (until umount). Unregister if they are canceled
(control-c) or time out.
This works around a bug in vfs_rename_dir() that rehashes the target
dentry. Ensure such dentries always fail revalidation by timing out the
dentry lease and kicking it out of the current directory lease gen.
Sage Weil [Fri, 20 Nov 2009 23:14:15 +0000 (15:14 -0800)]
ceph: reset msgr backoff during open, not after successful handshake
Reset the backoff delay when we reopen the connection, so that the delays
for any initial connection problems are reasonable. We were resetting only
after a successful handshake, which was of limited utility.
Sage Weil [Fri, 20 Nov 2009 21:43:45 +0000 (13:43 -0800)]
ceph: reset requested max_size after mds reconnect
The max_size increase request to the MDS can get lost during an MDS
restart and reconnect. Reset our requested value after the MDS recovers,
so that any blocked writes will re-request a larger max_size upon waking.
Also, explicit wake session caps after the reconnect. Normally the cap
renewal catches this, but not in the cases where the caps didn't go stale
in the first place, which would leave writers waiting on max_size asleep.
Sage Weil [Thu, 19 Nov 2009 00:50:41 +0000 (16:50 -0800)]
ceph: fix debugfs entry, simplify fsid checks
We may first learn our fsid from any of the mon, osd, or mds maps
(whichever the monitor sends first). Consolidate checks in a single
helper. Initialize the client debugfs entry then, since we need the
fsid (and global_id) for the directory name.
When we open a monitor session, we send an initial AUTH message listing
the auth protocols we support, our entity name, and (possibly) a previously
assigned global_id. The monitor chooses a protocol and responds with an
initial message.
Initially implement AUTH_NONE, a dummy protocol that provides no security,
but works within the new framework. It generates 'authorizers' that are
used when connecting to (mds, osd) services that simply state our entity
name and global_id.
Sage Weil [Wed, 18 Nov 2009 19:22:36 +0000 (11:22 -0800)]
ceph: remove bad calls to ceph_con_shutdown
We want to ceph_con_close when we're done with the connection, before
the ref count reaches 0. Once it does, do not call ceph_con_shutdown,
as that takes the con mutex and may sleep, and besides that is
unnecessary.
Sage Weil [Thu, 12 Nov 2009 23:07:24 +0000 (15:07 -0800)]
ceph: fix page invalidation deadlock
We occasionally want to make a best-effort attempt to invalidate cache
pages without fear of blocking. If this fails, we fall back to an async
invalidate in another thread.
Use invalidate_mapping_pages instead of invalidate_inode_page2, as that
will skip locked pages, and not deadlock.
Sage Weil [Wed, 11 Nov 2009 00:02:23 +0000 (16:02 -0800)]
ceph: remove recon_gen logic
We don't get an explicit affirmative confirmation that our caps reconnect,
nor do we necessarily want to pay that cost. So, take all this code out
for now.
Sage Weil [Tue, 10 Nov 2009 22:34:36 +0000 (14:34 -0800)]
ceph: separate banner and connect during handshake into distinct stages
We need to make sure we only swab the address during the banner once. So
break process_banner out of process_connect, and clean up the surrounding
code so that these are distinct phases of the handshake.
Sage Weil [Mon, 9 Nov 2009 20:05:48 +0000 (12:05 -0800)]
ceph: do not confuse stale and dead (unreconnected) caps
We were using the cap_gen to track both stale caps (caps that timed out
due to temporarily losing touch with the mds) and dead caps that did not
reconnect after an MDS failure. Introduce a recon_gen counter to track
reconnections to restarted MDSs and kill dead caps based on that instead.
Rename gen to cap_gen while we're at it to make it more clear which is
which.
Sage Weil [Sun, 8 Nov 2009 04:18:22 +0000 (20:18 -0800)]
ceph: make CRUSH hash function a bucket property
Make the integer hash function a property of the bucket it is used on. This
allows us to gracefully add support for new hash functions without starting
from scatch.
Sage Weil [Sat, 7 Nov 2009 05:55:25 +0000 (21:55 -0800)]
ceph: make object hash a pg_pool property
The object will be hashed to a placement seed (ps) based on the pg_pool's
hash function. This allows new hashes to be introduced into an existing
object store, or selection of a hash appropriate to the objects that
will be stored in a particular pool.
Sage Weil [Sat, 7 Nov 2009 00:44:05 +0000 (16:44 -0800)]
ceph: use strong hash function for mapping objects to pgs
We were using the (weak) dcache hash function, but it was leaving lower
bits consecutive for consecutive (inode) objects. We really want to make
the object to pg mapping random and uniform, so use a proper hash function
here.
This is Robert Jenkin's public domain hash function (with some minor
cleanup):
http://burtleburtle.net/bob/hash/evahash.html
Sage Weil [Wed, 4 Nov 2009 19:39:12 +0000 (11:39 -0800)]
ceph: fix endian conversions for ceph_pg
The endian conversions don't quite work with the old union ceph_pg. Just
make it a regular struct, and make each field __le. This is simpler and it
has the added bonus of actually working.
Sage Weil [Tue, 3 Nov 2009 23:17:56 +0000 (15:17 -0800)]
ceph: use fixed endian encoding for ceph_entity_addr
We exchange struct ceph_entity_addr over the wire and store it on disk.
The sockaddr_storage.ss_family field, however, is host endianness. So,
fix ss_family endianness to big endian when sending/receiving over the
wire.
Sage Weil [Mon, 2 Nov 2009 01:53:24 +0000 (17:53 -0800)]
crush: always return a value from crush_bucket_choose
Even when we encounter a corrupt bucket. We still BUG(). This fixes
the warning
fs/ceph/crush/mapper.c: In function 'crush_choose':
fs/ceph/crush/mapper.c:352: warning: control may reach end of non-void function
'crush_bucket_choose' being inlined
Noah Watkins [Fri, 30 Oct 2009 19:57:30 +0000 (12:57 -0700)]
ceph: fix intra strip unit length calculation
Commit 645a102581b3639836b17d147c35d574fd6e8267 fixes calculation of object
offset for layouts with multiple stripes per object. This updates the
calculation of the length written to take into account multiple stripes per
object.
Signed-off-by: Noah Watkins <noah@noahdesu.com> Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 28 Oct 2009 22:15:05 +0000 (15:15 -0700)]
ceph: fix object striping calculation for non-default striping schemes
We were incorrectly calculationing of object offset. If we have multiple
stripe units per object, we need to shift to the start of the current
su in addition to the offset within the su.
Also rename bno to ono (object number) to avoid some variable naming
confusion.
Sage Weil [Tue, 27 Oct 2009 18:50:50 +0000 (11:50 -0700)]
ceph: allocate and parse mount args before client instance
This simplifies much of the error handling during mount. It also means
that we have the mount args before client creation, and we can initialize
based on those options.
Sage Weil [Thu, 22 Oct 2009 17:53:02 +0000 (10:53 -0700)]
ceph: remove small mon addr limit; use CEPH_MAX_MON where appropriate
Get rid of separate max mon limit; use the system limit instead. This
allows mounts when there are lots of mon addrs provided by mount.ceph (as
with a host with lots of A/AAAA records).
Sage Weil [Wed, 21 Oct 2009 18:21:49 +0000 (11:21 -0700)]
ceph: move directory size logic to ceph_getattr
We can't fill i_size with rbytes at the fill_file_size stage without
adding additional checks for directories. Notably, we want st_blocks
to remain 0 on directories so that 'du' still works.
Fill in i_blocks, i_size specially in ceph_getattr instead.
Sage Weil [Mon, 19 Oct 2009 18:41:51 +0000 (11:41 -0700)]
ceph: include preferred osd in placement seed
Mix the preferred osd (if any) into the placement seed that is fed into
the CRUSH object placement calculation. This prevents all the placement
pgs from peering with the same osds.
Sage Weil [Thu, 15 Oct 2009 00:36:07 +0000 (17:36 -0700)]
ceph: warn on allocation from msgpool with larger front_len
Pass the front_len we need when pulling a message off a msgpool,
and WARN if it is greater than the pool's size. Then try to
allocate a new message (to continue without failing).