Sage Weil
b780d54597
monc: pick monitor before sending initial mount
2009-09-17 14:56:34 -07:00
Sage Weil
7e17e27788
uclient: implement basic I_COMPLETE support
2009-09-17 14:54:36 -07:00
Sage Weil
b53f9a4efa
uclient: factor out command cap issue checks
...
rdcache_gen, shared_gen
2009-09-17 14:25:55 -07:00
Sage Weil
9e3f6fd7fa
mds: minor comments
2009-09-17 14:16:05 -07:00
Sage Weil
310af17d42
kclient: move some inode update code inside i_lock
2009-09-17 14:15:53 -07:00
Sage Weil
ee0f1588b9
kclient: use shared_gen, not cache_gen, for I_COMPLETE tracking
2009-09-17 14:15:39 -07:00
Sage Weil
0f892e34be
todo
2009-09-17 13:47:35 -07:00
Sage Weil
e9f480d62a
monclient: small cleanup
2009-09-17 13:18:11 -07:00
Sage Weil
c7b904e213
todo
2009-09-17 13:18:08 -07:00
Sage Weil
81bb175003
kclient: fix console msg
2009-09-17 13:15:29 -07:00
Sage Weil
131fa21c10
kclient: move caps stats into client debugfs dir; fix error checking
2009-09-17 12:04:35 -07:00
Sage Weil
2cc5826946
kclient: put fsid in debugfs dir name
2009-09-17 12:04:35 -07:00
Sage Weil
f11918b1db
mds: make parallel_fetch work with hard/remote links
...
We need to open remote ino's, and possibly add to gather (like
a fetch). Required some refactoring. Killed/renamed
otherwise unused traverse_to_dir.
2009-09-17 11:58:06 -07:00
Sage Weil
69f13a2e3f
cosd.conf: fix date stamp
2009-09-17 10:50:14 -07:00
Sage Weil
50f3ee431f
msgr: fix null deref setting message connection ref
2009-09-17 10:41:07 -07:00
Sage Weil
71c1f4f41c
configure: don't link expat, fcgi to everything
...
The default AC_CHECK_LIB action is to add -lfoo to the linker args,
which we don't want!
2009-09-16 10:34:04 -07:00
Sage Weil
3d63b7c922
Makefile: fix mode on installed headers
2009-09-15 21:27:41 -07:00
Sage Weil
decde38057
msgr: change dispatcher interfaces; some monclient cleanups
2009-09-15 15:02:17 -07:00
Sage Weil
cca45f3c7b
uclient: do not do locking for 'atomic' writes
...
They aren't tested and aren't safe (client may fail; there's no cleanup).
2009-09-15 13:14:01 -07:00
Sage Weil
147f25a454
mds: flush log in caller(s), not scatter_writebehind
...
Otherwise scatter_tick(), in particular, will call scatter_writebehind
a zillion times and generate lots of small log flushes. Instead, flush
at the end of the whole tick.
2009-09-14 15:34:47 -07:00
Sage Weil
ac49cccc57
check_mds_request_latency: fix parsing for current debug output
2009-09-14 15:07:30 -07:00
Greg Farnum
af1c653443
Hadoop: Final submission prep.
...
IO classes should be public, updated Javadoc, and added license grants.
2009-09-11 14:20:52 -07:00
Greg Farnum
746bb5ecfa
Hadoop: Remove old reference CephFileSystem material.
2009-09-11 14:20:52 -07:00
Greg Farnum
c76574b8f6
Hadoop: Change to work with trunk rather than .20; update documentation.
2009-09-11 14:20:52 -07:00
Greg Farnum
784bcc25b1
Hadoop: Remove unused CephException class.
2009-09-11 14:20:52 -07:00
Greg Farnum
244692bc3c
Hadoop: Changed debug printing so it doesn't construct useless strings.
...
Removed some commented-out code.
2009-09-11 14:20:52 -07:00
Sage Weil
c2b966453a
uclient: fix warning
2009-09-10 17:01:01 -07:00
Sage Weil
e5b1d5ba43
kclient: set/verify seq in renew caps messages
2009-09-10 16:36:55 -07:00
Sage Weil
70ca9c95b6
uclient: fix cap renewal
...
Renew caps on session open.
Set/verify seq # in renew caps reply.
2009-09-10 16:33:20 -07:00
Sage Weil
61d52a53fb
mds: include original seq in renew caps reply
2009-09-10 16:32:25 -07:00
Sage Weil
83c5d1063e
synclient: fix bad leading / on play ~ prefix
2009-09-10 16:24:40 -07:00
Sage Weil
4fddaad08d
debian: fix radosgw package
2009-09-10 15:10:29 -07:00
Sage Weil
10096bc545
msgr: be slightly smarter about corking on partial msgs, acks
...
Actually, if MSG_MORE will cork indefinitely, that may not be
what we want for write_ack()... :/
2009-09-10 14:57:13 -07:00
Sage Weil
b40b64d5d1
msgr: make ack le64 to match seq
2009-09-10 14:44:52 -07:00
Sage Weil
d20510956e
kclient: fix format strings for entity num
2009-09-10 14:32:05 -07:00
Sage Weil
933320f75f
logger: fix avg calculation
2009-09-10 14:22:25 -07:00
Sage Weil
0888b5f1cb
mds: log latency of _first_ (unsafe) reply, not safe
...
Client rarely blocks on safe reply.
2009-09-10 14:16:19 -07:00
Sage Weil
40f83e8f79
uclient: and reply cond guard; make forward _always_ resend request
2009-09-10 14:00:53 -07:00
Sage Weil
335ea1d703
debian: create radosgw package
2009-09-10 13:48:47 -07:00
Sage Weil
3b0a6c9a37
configure: fix radosgw conditional compilation
2009-09-10 13:48:12 -07:00
Sage Weil
6b9d6dea6e
kclient: sparse endian fixes, fsid cleanup
2009-09-10 11:59:10 -07:00
Sage Weil
122c612156
kclient: fix checkpatch label warnings
2009-09-10 11:23:33 -07:00
Sage Weil
c7987d758d
kclient: avoid short reconnect loop on connection close
2009-09-10 11:19:52 -07:00
Sage Weil
235914c7ff
msgr: kill per-msg protocol version values
2009-09-10 11:18:52 -07:00
Sage Weil
1917c2edb4
msgr: check protocol version during connection
2009-09-10 11:15:36 -07:00
Sage Weil
5f3b046982
kclient: allow sent msgs to be revoked from a connection
...
The simplest solution seemed to be to replace the spinlock with a
mutex and carry it for the duration of try_write() (including the
NOWAIT calls to kernel_sendmsg, etc). If there is contention the
overhead will be a bit higher, but the code is much simpler.
2009-09-09 16:46:42 -07:00
Sage Weil
78b005b317
crush: fix long line
2009-09-09 12:48:39 -07:00
Sage Weil
86addcf660
kclient: whitespace
2009-09-09 12:48:35 -07:00
Sage Weil
df985c474a
mds: avoid client lookup by using per-connection msgr state
2009-09-09 12:27:46 -07:00
Sage Weil
7026e35d89
msgr: per connection state
2009-09-09 12:27:19 -07:00
