]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Yehuda Sadeh [Thu, 12 Mar 2009 17:48:24 +0000 (10:48 -0700)]
cconf: restructure, use common infrastructure
Yehuda Sadeh [Thu, 12 Mar 2009 16:33:16 +0000 (09:33 -0700)]
cmon: typo
Sage Weil [Thu, 12 Mar 2009 05:05:57 +0000 (22:05 -0700)]
drop dstart/dstop
Sage Weil [Thu, 12 Mar 2009 05:05:01 +0000 (22:05 -0700)]
initscript: get pid_file so stop/reload/etc work
Sage Weil [Thu, 12 Mar 2009 05:00:08 +0000 (22:00 -0700)]
vstart: fix mkmonfs call
Sage Weil [Thu, 12 Mar 2009 04:57:09 +0000 (21:57 -0700)]
confutils: initialize pointer to null
Yehuda Sadeh [Thu, 12 Mar 2009 00:32:08 +0000 (17:32 -0700)]
conf: parameter substitution work (not fully working)
Sage Weil [Wed, 11 Mar 2009 23:31:00 +0000 (16:31 -0700)]
config: clean up sample conf
Sage Weil [Wed, 11 Mar 2009 23:28:52 +0000 (16:28 -0700)]
makefile: remove sample.cluster.conf
Yehuda Sadeh [Wed, 11 Mar 2009 23:30:15 +0000 (16:30 -0700)]
cconf: added $num to substitution variables
Yehuda Sadeh [Wed, 11 Mar 2009 23:28:54 +0000 (16:28 -0700)]
cconf: can use substitution variables
Sage Weil [Wed, 11 Mar 2009 23:07:16 +0000 (16:07 -0700)]
config: fix sample conf files
Sage Weil [Wed, 11 Mar 2009 23:21:50 +0000 (16:21 -0700)]
initscript: more fixes
Sage Weil [Wed, 11 Mar 2009 23:21:43 +0000 (16:21 -0700)]
mkmonfs: fix argument parsing
Sage Weil [Wed, 11 Mar 2009 23:09:46 +0000 (16:09 -0700)]
config: fix cosd conf files
Sage Weil [Wed, 11 Mar 2009 23:03:05 +0000 (16:03 -0700)]
config: just ceph.conf, no more cluster.conf
Sage Weil [Wed, 11 Mar 2009 22:49:30 +0000 (15:49 -0700)]
config: more config fixes
Sage Weil [Wed, 11 Mar 2009 22:49:17 +0000 (15:49 -0700)]
man: some manpages fixes
Yehuda Sadeh [Wed, 11 Mar 2009 22:52:39 +0000 (15:52 -0700)]
config: can use alternative module name
Yehuda Sadeh [Wed, 11 Mar 2009 22:33:43 +0000 (15:33 -0700)]
config: search in global section by default
Sage Weil [Wed, 11 Mar 2009 22:23:46 +0000 (15:23 -0700)]
initscript: lots of changes
Sage Weil [Wed, 11 Mar 2009 21:58:58 +0000 (14:58 -0700)]
config: fix mds, mon, osd startup
Sage Weil [Wed, 11 Mar 2009 20:42:19 +0000 (13:42 -0700)]
mds: rework mds standby framework
Sage Weil [Wed, 11 Mar 2009 20:08:33 +0000 (13:08 -0700)]
mds: fix up startup options
Sage Weil [Wed, 11 Mar 2009 20:04:03 +0000 (13:04 -0700)]
mds: name mds daemon instances independent of their current logical rank
Yehuda Sadeh [Wed, 11 Mar 2009 22:27:20 +0000 (15:27 -0700)]
cconf: resort to search in global section
Yehuda Sadeh [Wed, 11 Mar 2009 22:24:57 +0000 (15:24 -0700)]
cconf: fix alternate section name
Yehuda Sadeh [Wed, 11 Mar 2009 22:22:41 +0000 (15:22 -0700)]
cconf: search in both ${type}.${id} and ${type}${id}
Yehuda Sadeh [Wed, 11 Mar 2009 22:15:58 +0000 (15:15 -0700)]
cconf: add -i, -t options
Yehuda Sadeh [Wed, 11 Mar 2009 21:39:15 +0000 (14:39 -0700)]
conf: restructure of the conf mechanism
Sage Weil [Wed, 11 Mar 2009 18:36:07 +0000 (11:36 -0700)]
initscript: add 'pre start eval' hook
Sage Weil [Wed, 11 Mar 2009 18:35:53 +0000 (11:35 -0700)]
cosd: keep cores in src dir
Sage Weil [Wed, 11 Mar 2009 18:35:36 +0000 (11:35 -0700)]
mkcephfs: no need for cluster.conf on cosd --mkfs_for_osd
Sage Weil [Wed, 11 Mar 2009 18:35:08 +0000 (11:35 -0700)]
monclient: check for monmap_file _first_
Sage Weil [Wed, 11 Mar 2009 18:06:17 +0000 (11:06 -0700)]
msgr: use daemon() to daemonize
Otherwise we don't clean up all our fds or something and ssh execs hang.
Sage Weil [Wed, 11 Mar 2009 18:03:57 +0000 (11:03 -0700)]
cosd: use absolute paths
Sage Weil [Wed, 11 Mar 2009 18:03:43 +0000 (11:03 -0700)]
monclient: clean up
Sage Weil [Wed, 11 Mar 2009 18:03:33 +0000 (11:03 -0700)]
initscripts: fix some conf vs clusterconf problem
Sage Weil [Wed, 11 Mar 2009 18:03:09 +0000 (11:03 -0700)]
osd: be more verbose about startup failures
Sage Weil [Wed, 11 Mar 2009 17:05:56 +0000 (10:05 -0700)]
man: fix makefile
Sage Weil [Wed, 11 Mar 2009 16:55:04 +0000 (09:55 -0700)]
config: print warning on daemon startup
Sage Weil [Wed, 11 Mar 2009 04:36:27 +0000 (21:36 -0700)]
debian: don't include dupstore
It's really only useful for ebofs vs filestore debugging. If you're just
using filestore, you can just cp -a.
Sage Weil [Wed, 11 Mar 2009 04:35:44 +0000 (21:35 -0700)]
man: mount.ceph man page
Sage Weil [Wed, 11 Mar 2009 04:20:43 +0000 (21:20 -0700)]
man: mkmonfs, monmaptool, ceph
Sage Weil [Wed, 11 Mar 2009 03:49:30 +0000 (20:49 -0700)]
man: cconf
Sage Weil [Wed, 11 Mar 2009 03:38:47 +0000 (20:38 -0700)]
man: crushtool, osdmaptool
Sage Weil [Tue, 10 Mar 2009 23:49:23 +0000 (16:49 -0700)]
man: csyn, crun man pages
Sage Weil [Tue, 10 Mar 2009 23:30:15 +0000 (16:30 -0700)]
man: add cfuse man page
Sage Weil [Tue, 10 Mar 2009 23:23:52 +0000 (16:23 -0700)]
mkcephfs: fix typo
Sage Weil [Tue, 10 Mar 2009 23:21:15 +0000 (16:21 -0700)]
debian: add man pages so far
Yehuda Sadeh [Tue, 10 Mar 2009 23:26:47 +0000 (16:26 -0700)]
conf: add --show-conf,-S option that dumps conf and exits
Sage Weil [Tue, 10 Mar 2009 23:10:02 +0000 (16:10 -0700)]
man: document -m monaddr
Sage Weil [Tue, 10 Mar 2009 23:08:00 +0000 (16:08 -0700)]
config: -c foo or --conf=foo; -C bar or --cluster-conf=bar
Yehuda Sadeh [Tue, 10 Mar 2009 22:16:15 +0000 (15:16 -0700)]
kclient: add mon enries for sysfs
Sage Weil [Tue, 10 Mar 2009 22:22:52 +0000 (15:22 -0700)]
usage
Sage Weil [Tue, 10 Mar 2009 22:00:46 +0000 (15:00 -0700)]
man: more man pages
Sage Weil [Tue, 10 Mar 2009 21:57:24 +0000 (14:57 -0700)]
cmds: remove unused --standby arg, rename --standby_replay_for to --shadow
Sage Weil [Tue, 10 Mar 2009 21:43:07 +0000 (14:43 -0700)]
config: default to -d for server daemons; -D/--nodaemon for foreground operation
Sage Weil [Tue, 10 Mar 2009 21:40:16 +0000 (14:40 -0700)]
cosd: clean up startup options
Sage Weil [Tue, 10 Mar 2009 21:39:54 +0000 (14:39 -0700)]
man: mkcephfs, cosd man pages
Sage Weil [Tue, 10 Mar 2009 20:22:51 +0000 (13:22 -0700)]
debian: fix lintian errors (except missing man pages)
Sage Weil [Tue, 10 Mar 2009 20:22:37 +0000 (13:22 -0700)]
initscript: LSB block; add reload and force-reload
Sage Weil [Tue, 10 Mar 2009 19:41:15 +0000 (12:41 -0700)]
initscript: pre/post start/stop/forcestop hooks
Sage Weil [Tue, 10 Mar 2009 19:34:23 +0000 (12:34 -0700)]
config: allow chdir to arbitrary dir on daemonize (not just /)
This lets you configure where your core files go.
Sage Weil [Tue, 10 Mar 2009 19:11:30 +0000 (12:11 -0700)]
mkcephfs: take compiled 'crush map' or 'crush map src' from cluster.conf
Sage Weil [Tue, 10 Mar 2009 18:18:32 +0000 (11:18 -0700)]
crushtool: add --build command
Sage Weil [Tue, 10 Mar 2009 16:49:06 +0000 (09:49 -0700)]
remove user error induced asserts
Yehuda Sadeh [Tue, 10 Mar 2009 00:01:45 +0000 (17:01 -0700)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Tue, 10 Mar 2009 00:01:31 +0000 (17:01 -0700)]
kclient: add osdc data to sysfs
Sage Weil [Mon, 9 Mar 2009 22:39:16 +0000 (15:39 -0700)]
bump version to v0.7.1 for unstable branch
Sage Weil [Mon, 9 Mar 2009 22:28:26 +0000 (15:28 -0700)]
mkcephfs: btrfsctl -a between mkfs and mount
Sage Weil [Mon, 9 Mar 2009 22:25:46 +0000 (15:25 -0700)]
kclient: patch series updates
Sage Weil [Mon, 9 Mar 2009 21:58:20 +0000 (14:58 -0700)]
kclient: fix /sys/fs/ceph/mdsmap
Sage Weil [Mon, 9 Mar 2009 21:56:40 +0000 (14:56 -0700)]
kclient: sysfs whitespace cleanup
Wrap at 80 columns...
Sage Weil [Mon, 9 Mar 2009 21:51:27 +0000 (14:51 -0700)]
kclient: fix patch series
Sage Weil [Mon, 9 Mar 2009 21:51:15 +0000 (14:51 -0700)]
osdmap: use generic crush_build_bucket
Easier to switch bucket types. Avoids type cast.
Sage Weil [Mon, 9 Mar 2009 21:50:37 +0000 (14:50 -0700)]
kclient: fix osdmap decoding of straw buckets
Surprised this worked until now.
Yehuda Sadeh [Mon, 9 Mar 2009 21:55:54 +0000 (14:55 -0700)]
kclient: sysfs, free path after build_path
Yehuda Sadeh [Mon, 9 Mar 2009 21:50:17 +0000 (14:50 -0700)]
kclient: sysfs cleanup
Yehuda Sadeh [Mon, 9 Mar 2009 21:32:17 +0000 (14:32 -0700)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Mon, 9 Mar 2009 21:31:42 +0000 (14:31 -0700)]
kclient: sysfs, add mds name and op of current requests
Sage Weil [Mon, 9 Mar 2009 18:34:34 +0000 (11:34 -0700)]
kclient: simplify osd timeouts
Call handle_timeout N seconds after oldest request. In
handle_timeout, reschedule N seconds in future. Reschedule
with new oldest request when oldest requests completes.
Maintain a r_timeout_stamp value for each request.
Sage Weil [Mon, 9 Mar 2009 20:57:22 +0000 (13:57 -0700)]
cosd: fewer osds, with big btrfs devices
Sage Weil [Mon, 9 Mar 2009 20:18:56 +0000 (13:18 -0700)]
initscripts: allow vars with spaces
Sage Weil [Mon, 9 Mar 2009 20:18:05 +0000 (13:18 -0700)]
osdmap: build default crush maps out of mostly straw buckets
Sage Weil [Mon, 9 Mar 2009 19:51:04 +0000 (12:51 -0700)]
confutils: make max line very large
Yehuda Sadeh [Mon, 9 Mar 2009 17:08:51 +0000 (10:08 -0700)]
Merge branch 'unstable' of ssh://ceph.newdream.net/git/ceph into unstable
Yehuda Sadeh [Mon, 9 Mar 2009 17:08:29 +0000 (10:08 -0700)]
kclient: add sysfs directory per mds request
Sage Weil [Sat, 7 Mar 2009 00:45:55 +0000 (16:45 -0800)]
filepath: always prepend ino if it's nonzero
That means the only time we don't have an ino perfix is for fully
relative paths.
Sage Weil [Sat, 7 Mar 2009 00:44:37 +0000 (16:44 -0800)]
client: fix request paths
Use filepath, not bare string, whenever we can.
Sage Weil [Sat, 7 Mar 2009 00:26:05 +0000 (16:26 -0800)]
kclient: fix up replay after mds recovery
Do not kick unsafe requests in kick_requests (and in particular, do not
drop the session refs).
Set retry_attempt to MDS_REQUEST_REPLAY to indicate which requests are
replays.
Sage Weil [Sat, 7 Mar 2009 00:25:04 +0000 (16:25 -0800)]
mds: incorporate reconnect cap info after requests replay if ino originally missing
If the client reasserts caps from replayed requests, the inodes don't yet
exists during the reconnect (or even rejoin) stage. So, if we don't find
the inode, keep the reconnect info around. When processing a replayed
request in CInode::encode_inodestat, set wanted/issued appropriately.
This is incomplete. We really need something to ensure we deal with
replayed requests before new requests are handled.. and on a cluster-wide
basis, since requests may involve slave requests to other mds's.
We also should clean up reconnects unclaimed after all replays are
complete. And somehow inform the client when the cap is officially
nonexistent and EBADF.
Sage Weil [Fri, 6 Mar 2009 22:06:19 +0000 (14:06 -0800)]
mds: include path base ino in client reconnect paths
Sage Weil [Fri, 6 Mar 2009 22:04:19 +0000 (14:04 -0800)]
filepath: no more leading / weirdness
Get rid of weird leading '/' iff ino==1 semantic. Clean up interface a bit.
Sage Weil [Fri, 6 Mar 2009 19:51:39 +0000 (11:51 -0800)]
kclient: drop unused redirty code from set_page_dirty
Sage Weil [Fri, 6 Mar 2009 19:31:12 +0000 (11:31 -0800)]
filestore: lower min_sync_interval
We mainly need something non-zero here to ensure we batch some
requests up when we get a stream of requests with 'startsync'.
Yehuda Sadeh [Fri, 6 Mar 2009 18:29:21 +0000 (10:29 -0800)]
kclient: some bookkeeper fixes
Sage Weil [Fri, 6 Mar 2009 18:16:12 +0000 (10:16 -0800)]
kclient: fix bookkeeper ifdef
Sage Weil [Fri, 6 Mar 2009 05:13:22 +0000 (21:13 -0800)]
kclient: patch series descriptions, updates
Sage Weil [Fri, 6 Mar 2009 04:58:01 +0000 (20:58 -0800)]
kclient: docs