Sage Weil
b732c0ddb8
kclient: unregister mds requests on unsafe reply
...
Fresh vs unsafe mds request replay is handled separately. So,
unregister requests on first reply. This avoids a stray dentry
reference if we unmount before getting a final safe (although
eventually those should be explicitly flushed as well).
2009-02-19 13:59:15 -08:00
Sage Weil
d7cb96bd12
kclient: clean up some mdsc locking
...
Do not drop mds mutex in __do_request().
2009-02-19 13:11:07 -08:00
Sage Weil
919b21429b
kclient: fix path memory leak
2009-02-19 13:11:07 -08:00
Sage Weil
c3ac056a41
kclient: async mds requestion completion callbacks
2009-02-19 13:11:07 -08:00
Sage Weil
0990d83f83
kclient: fix fh_to_dentry; move build_path to mds_client.c
2009-02-19 11:58:25 -08:00
Sage Weil
3dc43aa8e6
kclient: mds requests in terms of dentries, not paths
...
Generate mds request paths when request is sent, so that it reflects
the current state of mds capabilities. Eventually we can then make
the relative paths based on the current capabilities for that mds..
2009-02-19 11:58:25 -08:00
Sage Weil
376ac88a3b
kclient: async mds requests
...
Restructure the mds client to handle mds request asynchronously, so
that we can handle requests that are not managed by a blocking
thread.
2009-02-19 11:58:25 -08:00
Yehuda Sadeh
77a5f24f21
kclient: remove extra log output
2009-02-19 11:53:11 -08:00
Yehuda Sadeh
0c46645be3
kclient: sync writeout when not pdflush
2009-02-19 11:52:10 -08:00
Sage Weil
2d17b3a0d2
mds: only flush journal when we need to
...
Force a flush every tick just in case.
2009-02-19 11:01:51 -08:00
Sage Weil
0f05de942d
kclient: set 'sync' flag for fsync
...
This will ensure that fsynced data is committed immediately by the
OSDs.
2009-02-18 14:20:05 -08:00
Sage Weil
94c5031e6d
script: fix check_osd_request_latency
2009-02-18 11:53:51 -08:00
Sage Weil
cec662cc5f
script: check_osd_request_latency
2009-02-18 11:47:23 -08:00
Yehuda Sadeh
ffc756ea68
conf: can turn on debug logs in startup.conf
2009-02-18 11:07:19 -08:00
Yehuda Sadeh
6a71ef7525
conf: add debug_ms option
2009-02-17 13:26:24 -08:00
Sage Weil
dbfa9c3e63
mds: don't rdlock stat fields that client has EXCL cap for
2009-02-17 12:15:39 -08:00
yehudasa
4e88a943ed
add ceph_init to distribution
2009-02-17 10:57:48 -08:00
Yehuda Sadeh
d8ac8e5e5d
kclient: backport changes to kernel 2.6.25
2009-02-17 10:45:34 -08:00
Yehuda Sadeh
f62e4db415
kclient: fix mdsc->snap_empty cleanup
2009-02-13 17:09:12 -08:00
Yehuda Sadeh
f0a2967f05
kclient: remove client from client list before destorying
2009-02-13 17:06:22 -08:00
Yehuda Sadeh
c5ed89a98f
kclient: fix proc cleanup
2009-02-13 15:02:51 -08:00
Yehuda Sadeh
c98bb11a39
assert: some more assert_warn
2009-02-12 16:47:56 -08:00
Yehuda Sadeh
af812a0621
cosd: change some assert to assert_warn and clean exit
2009-02-12 15:05:12 -08:00
Yehuda Sadeh
5dc7beaf16
kclient: show some meaningful info in client proc entry
2009-02-12 12:43:17 -08:00
Yehuda Sadeh
172afed512
kclient: per client proc data file
2009-02-12 11:05:45 -08:00
Yehuda Sadeh
2b82ea40b2
kclient: added a proc function to read client data
2009-02-10 16:21:20 -08:00
Yehuda Sadeh
9ddbf9dbb8
kclient: proc subdir for each ceph client
2009-02-10 16:08:02 -08:00
Yehuda Sadeh
3b2cffdd61
kclient: keep global list of ceph clients
2009-02-10 15:42:51 -08:00
Yehuda Sadeh
b1203c227e
assert: warn on assertion if requested for current thread
2009-02-09 16:29:43 -08:00
Yehuda Sadeh
c511d68dea
added a common initialization function
2009-02-09 15:46:02 -08:00
Yehuda Sadeh
3ce47d046a
tls: addeed a thread local storage infrastructure
...
(cherry picked from commit 1b6b128f4e2aaab016575f48f2ec50d4abe002c9)
2009-02-09 15:01:21 -08:00
Yehuda Sadeh
62ee4089c8
osd: fix log level
2009-01-30 16:24:41 -08:00
Sage Weil
7e10dfac34
mds: make sure we eval_stray when inode was replicated
...
Also, drop dn to end of lru if unlinked and last cap.
2009-01-30 16:03:13 -08:00
Sage Weil
db46dadae9
mds: remove traces of CDir::committed_version_equivalent
...
Don't think it's coming back
2009-01-30 16:03:13 -08:00
Sage Weil
306e5560a7
kclient: release CAP_LINK_RDCACHE before unlink
2009-01-30 16:03:13 -08:00
Sage Weil
83c19c81e1
vstart: fix call to stop.sh
2009-01-30 16:03:13 -08:00
Sage Weil
d37775de55
todos
2009-01-30 16:03:11 -08:00
Sage Weil
95f63ea3c7
osd: print ops args correctly
2009-01-30 16:03:11 -08:00
Sage Weil
0b75ee62d4
mds: use generic helper when removing a cap
...
This way we make sure to do all the right checks on the locks,
mds_caps, and to call eval_stray if the inode is unlinked.
2009-01-30 16:03:11 -08:00
Sage Weil
179f7a81e2
mds: fix try_remove_unlinked_dn
...
Clear dentry NEW flag when we commit it.. pretty sure that was the
old bug. And we can't cheat the directory commit because the fnode
also gets updated.
2009-01-30 16:03:11 -08:00
Sage Weil
569c5d2e41
mds: fix purge of 0-length inode
2009-01-30 16:03:11 -08:00
Sage Weil
7b118b3d54
mds: remove unused inode purge fields
2009-01-30 16:03:10 -08:00
Yehuda Sadeh
756a4c1371
ceph_init: fix initalization
2009-01-30 15:56:36 -08:00
Yehuda Sadeh
bfb6cda4f3
ceph_init: added missing file
2009-01-30 15:48:56 -08:00
Yehuda Sadeh
c6ff2e6aaf
ceph_init: initial version
2009-01-30 15:47:59 -08:00
Yehuda Sadeh
7ed99eabc3
crun: restart only processes that should have dumped core
2009-01-30 12:46:17 -08:00
Sage Weil
fe1a365365
todo
2009-01-30 11:44:34 -08:00
Sage Weil
81bf6efdea
osd: put full original reqid in osd_op_reply
2009-01-30 11:44:34 -08:00
Sage Weil
4968553c22
mds: typos
2009-01-30 11:44:34 -08:00
Sage Weil
1df22a51a0
kclient: don't retain caps if inode is closed and nlink == 0
2009-01-30 11:44:33 -08:00
