]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
16 years agocrun: replace $* with "$@"
Yehuda Sadeh [Fri, 27 Feb 2009 20:55:41 +0000 (12:55 -0800)]
crun: replace $* with "$@"

16 years agoceph-daemons: replace $* with "$@"
Yehuda Sadeh [Fri, 27 Feb 2009 20:51:51 +0000 (12:51 -0800)]
ceph-daemons: replace $* with "$@"

16 years agocrun: simplify to just restart if coredump
Sage Weil [Fri, 27 Feb 2009 20:43:10 +0000 (12:43 -0800)]
crun: simplify to just restart if coredump

16 years agocephd: added "osd journal" option
Yehuda Sadeh [Fri, 27 Feb 2009 20:36:38 +0000 (12:36 -0800)]
cephd: added "osd journal" option

16 years agoceph-daemons: add --allhosts option
Sage Weil [Fri, 27 Feb 2009 19:38:22 +0000 (11:38 -0800)]
ceph-daemons: add --allhosts option

By default only start hosts on the current host if 'host' is defined.  If
--allhosts is specified, ssh into the remote host to start/stop as needed.

16 years agosample startup.conf
Sage Weil [Fri, 27 Feb 2009 19:18:45 +0000 (11:18 -0800)]
sample startup.conf

16 years agoceph-daemons: restructured (but incomplete) init.d script
Sage Weil [Fri, 27 Feb 2009 19:03:23 +0000 (11:03 -0800)]
ceph-daemons: restructured (but incomplete) init.d script

Only start/stop items explicitly defined in startup.conf.

Start specific osds or mdss (e.g. 'start osd12')

No NUM_OSD type iteration.. that only makes sense in a testing/dev
environment.

Extract options for each daemon in both start AND stop path, so that we can
stop the right daemon... e.g. if it is on a remote host or some such thing.

Try to be smart about looking for things in . or /usr/bin, /etc/ceph, etc.

16 years agoconfig: optionally maintain a pid file
Sage Weil [Fri, 27 Feb 2009 18:57:24 +0000 (10:57 -0800)]
config: optionally maintain a pid file

Remove it when we shut down _only_ if it matches getpid().

16 years agocconf: ability to list section names
Sage Weil [Fri, 27 Feb 2009 18:23:21 +0000 (10:23 -0800)]
cconf: ability to list section names

'./cconf -c ceph.conf -l foo' will list all section names that begin with
'foo'.

16 years agoceph_init: assume CEPH_BIN is `dirname ceph_init`
Sage Weil [Fri, 27 Feb 2009 17:20:37 +0000 (09:20 -0800)]
ceph_init: assume CEPH_BIN is `dirname ceph_init`

16 years agoceph_init: default to /etc/ceph/startup.conf
Sage Weil [Fri, 27 Feb 2009 17:05:19 +0000 (09:05 -0800)]
ceph_init: default to /etc/ceph/startup.conf

16 years agodebian: Build-depends libedit-dev
Sage Weil [Fri, 27 Feb 2009 16:46:41 +0000 (08:46 -0800)]
debian: Build-depends libedit-dev

16 years agologger: validate log type; remove unused mds.logger2
Sage Weil [Fri, 27 Feb 2009 04:47:11 +0000 (20:47 -0800)]
logger: validate log type; remove unused mds.logger2

Catch any undefined columns early.  Only real non-startup runtime danger
is sending events to the wrong logger, now.

16 years agodebug: ensure a SIGHUP reopens log reasonably soon
Sage Weil [Fri, 27 Feb 2009 04:27:56 +0000 (20:27 -0800)]
debug: ensure a SIGHUP reopens log reasonably soon

Since nothing new may be written to the log for a while, check if it needs
to be reopened in the mds/osd/mon tick() method.

16 years agologger: fix again
Sage Weil [Fri, 27 Feb 2009 02:34:42 +0000 (18:34 -0800)]
logger: fix again

bah

16 years agoconfig: only build an abspath if it's not already absolute
Sage Weil [Fri, 27 Feb 2009 02:33:40 +0000 (18:33 -0800)]
config: only build an abspath if it's not already absolute

16 years agoconfig: default to etc/ceph/ceph.conf
Sage Weil [Fri, 27 Feb 2009 01:36:25 +0000 (17:36 -0800)]
config: default to etc/ceph/ceph.conf

16 years agodebian: include var/log/ceph in package
Sage Weil [Fri, 27 Feb 2009 01:23:11 +0000 (17:23 -0800)]
debian: include var/log/ceph in package

16 years agologger: cleanup; make feeble attempt to create log dir
Sage Weil [Fri, 27 Feb 2009 01:23:03 +0000 (17:23 -0800)]
logger: cleanup; make feeble attempt to create log dir

16 years agoconfig: prefix default paths with install prefix
Sage Weil [Fri, 27 Feb 2009 01:13:43 +0000 (17:13 -0800)]
config: prefix default paths with install prefix

16 years agoconfig: verify argument exists before dereferencing args[++i]
Sage Weil [Thu, 26 Feb 2009 23:05:08 +0000 (15:05 -0800)]
config: verify argument exists before dereferencing args[++i]

Avoid segfault on malformed command line args.

16 years agoconfig: use sane default paths for logs; use .conf for *start.sh
Sage Weil [Thu, 26 Feb 2009 22:50:38 +0000 (14:50 -0800)]
config: use sane default paths for logs; use .conf for *start.sh

Specify sane deafult paths for logs (/var/log/ceph/*).  Use workingdir.conf
when starting up from dstart.sh or vstart.sh to use directories relative
to the current (presumably, build) directory.

16 years agologger: fixes
Sage Weil [Thu, 26 Feb 2009 22:48:40 +0000 (14:48 -0800)]
logger: fixes

16 years agologger: identify fields with int constants instead of strings
Sage Weil [Thu, 26 Feb 2009 19:25:37 +0000 (11:25 -0800)]
logger: identify fields with int constants instead of strings

This avoids the crufty hashes and so forth.  The only downside is that
_all_ fields must be predeclared.  That's probably a good thing anyway.

Logger will assert if the key is invalid or the name hasn't been
registered.  Hopefully I haven't missed any.

16 years agoconfig: remove ms_hosts
Sage Weil [Thu, 26 Feb 2009 17:52:20 +0000 (09:52 -0800)]
config: remove ms_hosts

16 years agologger: configurable output dir
Sage Weil [Thu, 26 Feb 2009 17:52:00 +0000 (09:52 -0800)]
logger: configurable output dir

16 years agoosd: strip out hard-coded osd device paths
Sage Weil [Thu, 26 Feb 2009 17:47:42 +0000 (09:47 -0800)]
osd: strip out hard-coded osd device paths

The correct device path should be specified on the command line.

16 years agokclient: zero_user_segments doesn't appear until 2.6.25
Sage Weil [Thu, 26 Feb 2009 17:23:34 +0000 (09:23 -0800)]
kclient: zero_user_segments doesn't appear until 2.6.25

16 years agokclient: more sysfs cleanup, bug fixes
Sage Weil [Wed, 25 Feb 2009 23:47:41 +0000 (15:47 -0800)]
kclient: more sysfs cleanup, bug fixes

16 years agokclient: fix kbuild patch, import script
Sage Weil [Wed, 25 Feb 2009 23:07:38 +0000 (15:07 -0800)]
kclient: fix kbuild patch, import script

16 years agokclient: move parens
Sage Weil [Wed, 25 Feb 2009 23:07:24 +0000 (15:07 -0800)]
kclient: move parens

16 years agokclient: changes for 2.6.29
Sage Weil [Wed, 25 Feb 2009 22:40:55 +0000 (14:40 -0800)]
kclient: changes for 2.6.29

16 years agokclient: avoid simple_prepare_write()
Sage Weil [Wed, 25 Feb 2009 22:39:08 +0000 (14:39 -0800)]
kclient: avoid simple_prepare_write()

No longer exported.

16 years agokclient: separate Kconfig
Sage Weil [Wed, 25 Feb 2009 22:34:55 +0000 (14:34 -0800)]
kclient: separate Kconfig

16 years agotodos
Sage Weil [Wed, 25 Feb 2009 21:07:23 +0000 (13:07 -0800)]
todos

16 years agokclient: add module author
Sage Weil [Wed, 25 Feb 2009 20:49:48 +0000 (12:49 -0800)]
kclient: add module author

16 years agokclient: dup kobj_sysfs_ops from kobject.c
Sage Weil [Wed, 25 Feb 2009 20:49:18 +0000 (12:49 -0800)]
kclient: dup kobj_sysfs_ops from kobject.c

They're not exported, bleh.

16 years agojournal: initiate full commit when journal hits 50% full
Sage Weil [Wed, 25 Feb 2009 20:38:32 +0000 (12:38 -0800)]
journal: initiate full commit when journal hits 50% full

We want to avoid filling up the journal, particularly since
_start_sync is a no-op with a journal enabled.

16 years agofilestore: don't initiate sync on _start_sync if there's a journal
Sage Weil [Wed, 25 Feb 2009 20:37:28 +0000 (12:37 -0800)]
filestore: don't initiate sync on _start_sync if there's a journal

Journal will commit immediately anyway.. don't slow the rest
down with a full commit to disk.

16 years agoosd: don't clobber log output stream
Sage Weil [Wed, 25 Feb 2009 20:20:43 +0000 (12:20 -0800)]
osd: don't clobber log output stream

Don't print a null char pointer.

16 years agomon: only do big sync if we're flushing lots of states
Sage Weil [Wed, 25 Feb 2009 20:19:54 +0000 (12:19 -0800)]
mon: only do big sync if we're flushing lots of states

The rest of the time, fsync them individually.

16 years agokclient: include sysfs (and not proc) in kernel client patchset
Sage Weil [Wed, 25 Feb 2009 19:10:43 +0000 (11:10 -0800)]
kclient: include sysfs (and not proc) in kernel client patchset

16 years agoosd: take journal on command line
Sage Weil [Wed, 25 Feb 2009 19:05:55 +0000 (11:05 -0800)]
osd: take journal on command line

./cosd dev -j journaldev ...

16 years agokclient: remove /proc/fs/ceph
Sage Weil [Wed, 25 Feb 2009 18:09:19 +0000 (10:09 -0800)]
kclient: remove /proc/fs/ceph

/sys/fs/ceph now does everything it used to.

16 years agokclient: add other maps to sysfs
Sage Weil [Wed, 25 Feb 2009 18:06:03 +0000 (10:06 -0800)]
kclient: add other maps to sysfs

16 years agokclient: add debug levels, mask to /sys/fs/ceph
Sage Weil [Wed, 25 Feb 2009 17:29:56 +0000 (09:29 -0800)]
kclient: add debug levels, mask to /sys/fs/ceph

16 years agokclient: some sysfs cleanup
Sage Weil [Wed, 25 Feb 2009 05:14:45 +0000 (21:14 -0800)]
kclient: some sysfs cleanup

16 years agodstop: remove cosd hostname prefix
Sage Weil [Wed, 25 Feb 2009 00:58:29 +0000 (16:58 -0800)]
dstop: remove cosd hostname prefix

16 years agotodos
Sage Weil [Wed, 25 Feb 2009 00:51:09 +0000 (16:51 -0800)]
todos

16 years agodstart: remove cosd hostname prefix
Sage Weil [Wed, 25 Feb 2009 00:51:04 +0000 (16:51 -0800)]
dstart: remove cosd hostname prefix

16 years agokclient: consider issued AND implemented AND dirty when updating inode
Sage Weil [Wed, 25 Feb 2009 00:37:55 +0000 (16:37 -0800)]
kclient: consider issued AND implemented AND dirty when updating inode

We do not want to update the local inode field(s) if we
 * still hold EXCL
 * are having EXCL revoked (not issued, but still implemented)
 * are writing back EXCL (not issued or implemented, but dirty)

16 years agomds: flush cap update if affected lock is unstable
Sage Weil [Wed, 25 Feb 2009 00:33:46 +0000 (16:33 -0800)]
mds: flush cap update if affected lock is unstable

16 years agodebian: include new binaries, scripts in ceph-common
Sage Weil [Tue, 24 Feb 2009 22:30:12 +0000 (14:30 -0800)]
debian: include new binaries, scripts in ceph-common

16 years agoMakefile: mount.ceph goes in usr/sbin
Sage Weil [Tue, 24 Feb 2009 22:30:00 +0000 (14:30 -0800)]
Makefile: mount.ceph goes in usr/sbin

16 years agodebian: pull version from configure.ac
Sage Weil [Tue, 24 Feb 2009 22:21:25 +0000 (14:21 -0800)]
debian: pull version from configure.ac

16 years agodebian: specify arch on command line
Sage Weil [Tue, 24 Feb 2009 22:13:37 +0000 (14:13 -0800)]
debian: specify arch on command line

16 years agomakefile: add missing files for kernel module
Sage Weil [Tue, 24 Feb 2009 22:08:19 +0000 (14:08 -0800)]
makefile: add missing files for kernel module

16 years agodebian: script to build and upload debs to repo
Sage Weil [Tue, 24 Feb 2009 21:57:16 +0000 (13:57 -0800)]
debian: script to build and upload debs to repo

16 years agomake unstable v0.7
Sage Weil [Tue, 24 Feb 2009 21:15:53 +0000 (13:15 -0800)]
make unstable v0.7

16 years agocommon: uninit var
Sage Weil [Tue, 24 Feb 2009 20:33:21 +0000 (12:33 -0800)]
common: uninit var

16 years agodebian: cleanup, changelog fix
Sage Weil [Tue, 24 Feb 2009 20:29:28 +0000 (12:29 -0800)]
debian: cleanup, changelog fix

16 years agokclient: typo
Sage Weil [Tue, 24 Feb 2009 20:22:16 +0000 (12:22 -0800)]
kclient: typo

16 years agomds: xlock filelock during truncate
Sage Weil [Tue, 24 Feb 2009 20:21:45 +0000 (12:21 -0800)]
mds: xlock filelock during truncate

This allows client to see projected truncation before it commits.

16 years agomds: look at active (not projected) inode when issuing truncation
Sage Weil [Tue, 24 Feb 2009 19:55:58 +0000 (11:55 -0800)]
mds: look at active (not projected) inode when issuing truncation

We are interested in applying the truncation that just committed, not one
that is projected.

16 years agomds: only do truncation process when file size goes down
Sage Weil [Tue, 24 Feb 2009 19:55:23 +0000 (11:55 -0800)]
mds: only do truncation process when file size goes down

For file size increase, no special action is required...

16 years agomds: limit propagation of full auth_pin count up hierarchy
Sage Weil [Tue, 24 Feb 2009 19:52:56 +0000 (11:52 -0800)]
mds: limit propagation of full auth_pin count up hierarchy

We only care if there are _some_ auth pins, not the exact count.  So only
propagation whether there are any pings from dir -> inode, to avoid walking
all the way up the hierarchy for each pin/unpin event.

16 years agokclient: don't remove from clients list if we're not on it
Sage Weil [Mon, 23 Feb 2009 21:46:46 +0000 (13:46 -0800)]
kclient: don't remove from clients list if we're not on it

16 years agoMakefile: link libcommon, lockdep last
Sage Weil [Mon, 23 Feb 2009 18:18:46 +0000 (10:18 -0800)]
Makefile: link libcommon, lockdep last

Link libcommon.a last for everything.

Include lockdep last in libcommon so to initialize first, shutdown
last, so that g_lockdep is initialized to zero for all Mutexes.

Put config.cc next to last so that the atomic_t thread counter is
initialized before all threads are created, shutdown after all
threads are destroyed.

16 years agoobjecter: shouldn't swap bufferlist buffers
Yehuda Sadeh [Fri, 20 Feb 2009 22:49:44 +0000 (14:49 -0800)]
objecter: shouldn't swap bufferlist buffers

16 years agocommon: remove warning
Yehuda Sadeh [Thu, 19 Feb 2009 23:42:56 +0000 (15:42 -0800)]
common: remove warning

16 years agocommon: virtual destructor for class Timer
Yehuda Sadeh [Thu, 19 Feb 2009 23:41:36 +0000 (15:41 -0800)]
common: virtual destructor for class Timer

16 years agotodo
Sage Weil [Thu, 19 Feb 2009 23:25:21 +0000 (15:25 -0800)]
todo

16 years agoSafeTimer: getting rid of global timer
Yehuda Sadeh [Fri, 13 Feb 2009 21:28:24 +0000 (13:28 -0800)]
SafeTimer: getting rid of global timer

16 years agokclient: only unregister after both acks
Sage Weil [Thu, 19 Feb 2009 23:22:12 +0000 (15:22 -0800)]
kclient: only unregister after both acks

Partially reverts behavior of b732c0ddb8a990a97b98606a6d9a387e5a21e3e5

We need the request to stay in the request_tree so that when the
safe arrives we can look it up (duh).  Also, we need to keep a
reference to the request during that period...

16 years agomds: reverse rename argument order
Sage Weil [Thu, 19 Feb 2009 22:47:03 +0000 (14:47 -0800)]
mds: reverse rename argument order

Order change introduced by 3dc43aa8e6606be15695d565715b6a7a1b2d38ea

16 years agoclient: reverse rename argument order
Sage Weil [Thu, 19 Feb 2009 22:45:37 +0000 (14:45 -0800)]
client: reverse rename argument order

16 years agokclient: add missing r_wait list_head init
Sage Weil [Thu, 19 Feb 2009 22:38:33 +0000 (14:38 -0800)]
kclient: add missing r_wait list_head init

16 years agokclient: initialize request list_heads
Sage Weil [Thu, 19 Feb 2009 22:30:16 +0000 (14:30 -0800)]
kclient: initialize request list_heads

16 years agokclient: cast path
Sage Weil [Thu, 19 Feb 2009 22:23:38 +0000 (14:23 -0800)]
kclient: cast path

path[12] will either point to freshly allocated paths (that we
later kfree) or a provided const path.

16 years agokclient: unregister mds requests on unsafe reply
Sage Weil [Thu, 19 Feb 2009 21:59:15 +0000 (13:59 -0800)]
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).

16 years agokclient: clean up some mdsc locking
Sage Weil [Thu, 19 Feb 2009 20:10:40 +0000 (12:10 -0800)]
kclient: clean up some mdsc locking

Do not drop mds mutex in __do_request().

16 years agokclient: fix path memory leak
Sage Weil [Thu, 19 Feb 2009 21:08:27 +0000 (13:08 -0800)]
kclient: fix path memory leak

16 years agokclient: async mds requestion completion callbacks
Sage Weil [Thu, 19 Feb 2009 21:10:40 +0000 (13:10 -0800)]
kclient: async mds requestion completion callbacks

16 years agokclient: fix fh_to_dentry; move build_path to mds_client.c
Sage Weil [Thu, 19 Feb 2009 00:35:45 +0000 (16:35 -0800)]
kclient: fix fh_to_dentry; move build_path to mds_client.c

16 years agokclient: mds requests in terms of dentries, not paths
Sage Weil [Thu, 19 Feb 2009 00:19:39 +0000 (16:19 -0800)]
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..

16 years agokclient: async mds requests
Sage Weil [Wed, 18 Feb 2009 22:13:49 +0000 (14:13 -0800)]
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.

16 years agokclient: remove extra log output
Yehuda Sadeh [Thu, 19 Feb 2009 19:53:11 +0000 (11:53 -0800)]
kclient: remove extra log output

16 years agokclient: sync writeout when not pdflush
Yehuda Sadeh [Thu, 19 Feb 2009 17:43:28 +0000 (09:43 -0800)]
kclient: sync writeout when not pdflush

16 years agomds: only flush journal when we need to
Sage Weil [Thu, 19 Feb 2009 19:01:51 +0000 (11:01 -0800)]
mds: only flush journal when we need to

Force a flush every tick just in case.

16 years agokclient: set 'sync' flag for fsync
Sage Weil [Wed, 18 Feb 2009 22:20:05 +0000 (14:20 -0800)]
kclient: set 'sync' flag for fsync

This will ensure that fsynced data is committed immediately by the
OSDs.

16 years agoscript: fix check_osd_request_latency
Sage Weil [Wed, 18 Feb 2009 19:53:51 +0000 (11:53 -0800)]
script: fix check_osd_request_latency

16 years agoscript: check_osd_request_latency
Sage Weil [Wed, 18 Feb 2009 19:47:23 +0000 (11:47 -0800)]
script: check_osd_request_latency

16 years agoconf: can turn on debug logs in startup.conf
Yehuda Sadeh [Wed, 18 Feb 2009 19:07:19 +0000 (11:07 -0800)]
conf: can turn on debug logs in startup.conf

16 years agoconf: add debug_ms option
Yehuda Sadeh [Tue, 17 Feb 2009 21:26:24 +0000 (13:26 -0800)]
conf: add debug_ms option

16 years agomds: don't rdlock stat fields that client has EXCL cap for
Sage Weil [Tue, 17 Feb 2009 20:15:39 +0000 (12:15 -0800)]
mds: don't rdlock stat fields that client has EXCL cap for

16 years agoadd ceph_init to distribution
yehudasa [Tue, 17 Feb 2009 18:57:48 +0000 (10:57 -0800)]
add ceph_init to distribution

16 years agokclient: backport changes to kernel 2.6.25
Yehuda Sadeh [Tue, 17 Feb 2009 18:45:34 +0000 (10:45 -0800)]
kclient: backport changes to kernel 2.6.25

16 years agokclient: fix mdsc->snap_empty cleanup
Yehuda Sadeh [Sat, 14 Feb 2009 01:09:12 +0000 (17:09 -0800)]
kclient: fix mdsc->snap_empty cleanup

16 years agokclient: remove client from client list before destorying
Yehuda Sadeh [Sat, 14 Feb 2009 01:06:22 +0000 (17:06 -0800)]
kclient: remove client from client list before destorying

16 years agokclient: fix proc cleanup
Yehuda Sadeh [Fri, 13 Feb 2009 23:02:51 +0000 (15:02 -0800)]
kclient: fix proc cleanup