Commit Graph

10043 Commits

Author SHA1 Message Date
Sage Weil
07e1a2f6d3 throttle: add fairness
This assumes that SignalOne() will wake people up in the right order.  Not
sure if that's true...
2010-04-28 10:36:30 -07:00
Sage Weil
f688ffabe6 mds: fix snaprealm split on capless inode 2010-04-26 14:08:27 -07:00
Sage Weil
b3bb8c608c buffer: clean up namespace pollution 2010-04-26 14:08:19 -07:00
Yehuda Sadeh
6ec8870d08 poolop: rename and move definitions to ceph_fs.h
This was done so that we can use it in the kclient
2010-04-24 10:10:00 -07:00
Sage Weil
0decb33d4a Makefile: define __CEPH__ in ceph builds. use it.
Avoid breakage when headers are used outside of ceph build (librados-dev)
2010-04-23 15:50:29 -07:00
Sage Weil
a6ee3aef93 mds: fix req use-after-free to get connection to direct reply 2010-04-23 15:22:34 -07:00
Greg Farnum
282b242828 atomic: reverse if test so we don't break things on separately-compiled librados 2010-04-23 15:13:35 -07:00
Greg Farnum
07dea4486a mds: remove unused variable from Locker 2010-04-23 15:12:03 -07:00
Sage Weil
3ed1a55d51 debian: build-depends libatomicops-dev 2010-04-23 14:50:26 -07:00
Sage Weil
e5a2080db8 Merge branch 'snaps' into unstable 2010-04-23 14:47:39 -07:00
Sage Weil
8f5d5ea7c7 mon: clean up mon snap type checks 2010-04-23 14:46:57 -07:00
Sage Weil
cf8ce29d3c osd: no error checking in data structure (caller should do it) 2010-04-23 14:42:49 -07:00
Sage Weil
ca464284ab mon: don't create/delete unmanaged snaps if pool snaps exist 2010-04-23 14:42:49 -07:00
Sage Weil
4a53f1c65c librados: clean up headres 2010-04-23 14:42:49 -07:00
Sage Weil
ec7ecc57a0 objecter: do pool op result parsing in Objecter completion wrapper 2010-04-23 14:42:48 -07:00
Greg Farnum
832c91921e librados: Adjust snapid allocation ufnctions and implement new objecter spec 2010-04-23 14:42:48 -07:00
Greg Farnum
076517fb85 objecter: Implement bufferlist-based return code for pool ops 2010-04-23 14:42:48 -07:00
Greg Farnum
f612ab86b1 mon: OSDMonitor fills in MPoolOpReply's bufferlist on unmanaged snap creation 2010-04-23 14:42:48 -07:00
Greg Farnum
57064765b7 msg: Give MPoolOpReply a bufferlist for reply data 2010-04-23 14:42:48 -07:00
Greg Farnum
503c41e6b1 mon: change the unmanaged snap interface for better distinction of reply codes
and snapid numbers.

Of course, this now means that the snapid isn't actually passed
back to the client, but that is about to change.
2010-04-23 14:42:48 -07:00
Greg Farnum
0fb4a6e85a librados: Set selfmanaged snap functions to use int instead of int64_t, too 2010-04-23 14:42:47 -07:00
Greg Farnum
39817b3df6 librados: add small bit of documentation, fix headers 2010-04-23 14:42:46 -07:00
Greg Farnum
4b0021fbe2 mon: don't crash if a client tries to self-manage managed snaps, or vice versa 2010-04-23 14:42:46 -07:00
Greg Farnum
d1d235e7b2 librados: add functions for client-managed snapshots in all the interfaces 2010-04-23 14:42:46 -07:00
Greg Farnum
5fb4e8fab3 objecter: New functions for client-managed snapshots. 2010-04-23 14:42:45 -07:00
Greg Farnum
eba044ce44 mon: OSDMonitor handles client-managed snapshot requests. 2010-04-23 14:42:45 -07:00
Greg Farnum
ac129d7895 osd: pg_pool_t gets new functions for unmanaged (ie, client-managed) snaps 2010-04-23 14:42:45 -07:00
Greg Farnum
5116700fdc msg: Add space for snapid in MPoolOp. Reply now uses an int64_t replyCode.
This allows space for a full 64-bit snapid to be returned, which is about
to be necessary.
2010-04-23 14:42:45 -07:00
Sage Weil
4fda5f922b msgr: fix atomic_t test
Missed this in merge 50285ecf5c
2010-04-23 14:30:23 -07:00
Sage Weil
5b6cfff36d Merge branch 'atomic' into unstable 2010-04-23 14:22:41 -07:00
Sage Weil
50285ecf5c Merge branch 'msgr' into unstable
Conflicts:
	src/TODO
	src/mds/Server.cc
	src/msg/SimpleMessenger.cc
	src/msg/SimpleMessenger.h
	src/osd/OSD.cc
2010-04-23 14:22:04 -07:00
Sage Weil
756b251bac v0.20 2010-04-23 12:10:24 -07:00
Sage Weil
609eaa14ca mds: skip client snap notification on unlink
This is cheating a bit, but should be harmless.  Basically, we spit off the
snaprealm when we unlink to keep the hierarchy vs snaprealm invariants
intact.  But we don't really care if the client does so, so we skip the
client_snap notifications.

That means the client will leave unlinked inodes in the realm they were
in at the time of unlink.  I'm pretty sure that won't cause problems
later.
2010-04-23 11:07:16 -07:00
Sage Weil
7a8daedf00 osd: more useful debugging on repop version, out of order completion 2010-04-22 15:12:58 -07:00
Sage Weil
f502547f45 osd: print a bit more debug info about compat problems 2010-04-22 15:12:58 -07:00
Sage Weil
feca69da56 mds: verify mdsmap compatset on startup 2010-04-22 15:12:58 -07:00
Sage Weil
4d7b2ceb08 mds: add CompatSet to mdsmap 2010-04-22 15:12:58 -07:00
Yehuda Sadeh
bfd962959d rbd: change rbd header disk format, update protocol version 2010-04-22 14:40:09 -07:00
Sage Weil
e3ef14840a msgr: never (re)open pipe when sending message directly to Connection* 2010-04-22 13:10:40 -07:00
Sage Weil
763203736a mdsmap: add separate struct version for extended portion of mdsmap 2010-04-22 12:04:01 -07:00
Sage Weil
d03bddc65f cauthtool: add --gen-print-key
Just generate a random cryptokey and print it out.
2010-04-22 11:43:37 -07:00
Sage Weil
5d2dcce770 auth: take key or file containing key on command line
e.g.

ceph --key AQBggdBLCJwLABAAuPVK/qJ3/ySbRvEnJHNJ8w== -s

ceph --keyfile file_with_key -s
ceph -K file_with_key -s

ceph -k keyring.bin -s
2010-04-22 11:39:40 -07:00
Sage Weil
2acb843bd6 cauthtool: always print entire encoded CryptoKey
Make cauthtool -l output consistent with what you need to feed to
mount.ceph(8).
2010-04-22 11:27:00 -07:00
Sage Weil
0fc47e267b msgr: increment connect_seq when STANDBY->CONNECTING
Otherwise we get this on the peer:

msg/SimpleMessenger.cc: In function 'int SimpleMessenger::Pipe::accept()':
msg/SimpleMessenger.cc:767: FAILED assert(existing->state == STATE_CONNECTING)
 1: (SimpleMessenger::Pipe::accept()+0x14b2) [0x654888]
 2: (SimpleMessenger::Pipe::reader()+0x32) [0x65538c]
 3: (SimpleMessenger::Pipe::Reader::entry()+0x19) [0x649493]
 4: (Thread::_entry_func(void*)+0x20) [0x659b06]
 5: /lib/libpthread.so.0 [0x7fcdc782f73a]
 6: (clone()+0x6d) [0x7fcdc6a5969d]
 NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
2010-04-22 07:37:13 -07:00
Sage Weil
629853516e Merge remote branch 'origin/mdssnaps' into unstable 2010-04-21 15:28:48 -07:00
Sage Weil
44208739bc filestore: default to writeahead journal, and no btrfs snaps
At least until btrfs snap deletion doesn't require a full commit (i.e. each
commit cycle doesn't do a commit for the snap creation AND another for the
old snap deletion).
2010-04-21 15:23:46 -07:00
Sage Weil
2caf513594 rbdtool: add, use prettybyte_t type 2010-04-21 14:56:35 -07:00
Sage Weil
0cf079ac77 rbdtool: add --info, --resize commands 2010-04-21 14:52:29 -07:00
Sage Weil
f4ef6a938a auth: make 'admin' the default user
We don't create a 'guest' user anyway.
2010-04-21 14:52:15 -07:00
Sage Weil
99ac12b10c librados: adjust C++ aio completion creation api 2010-04-21 14:24:26 -07:00