]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Fri, 7 Oct 2011 19:54:59 +0000 (12:54 -0700)]
crush: try _all_ bucket items when doing exhaustive search
N-1 isn't exhaustive.
Fixes: #1594
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 7 Oct 2011 18:10:46 +0000 (11:10 -0700)]
ceph.spec.in: build on suse
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 7 Oct 2011 16:53:45 +0000 (09:53 -0700)]
mon: tolerate 50ms (instead of 10ms) of clock drift
NTP on sepia can't manage 10ms, it seems. Sigh.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Fri, 7 Oct 2011 16:32:34 +0000 (09:32 -0700)]
rgw: fix printf format warning
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Fri, 7 Oct 2011 00:39:40 +0000 (17:39 -0700)]
rgw: make log object naming configurable
Use date(1) codes for object name, plus %i and %n for bucket id/name, and
make UTC vs localtime configurable.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 23:10:10 +0000 (16:10 -0700)]
mon: make other send_incremental variant handle map discontinuity
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 23:04:12 +0000 (16:04 -0700)]
auth: fix authorizer leak fix
We were never reusing the authorizer, so make it local. This fixes a crash
(double-free) on reset().
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 23:00:34 +0000 (16:00 -0700)]
mon: fix osdmap trimming unsigned overflow
This was trimming way to aggressively for early epochs due to the unsigned
version overflowing.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 22:54:48 +0000 (15:54 -0700)]
objecter: cope with missing osdmaps in history
Signed-off-by: Sage Weil <sage@newdream.net>
Josh Durgin [Thu, 6 Oct 2011 00:07:07 +0000 (17:07 -0700)]
osd, pg: ignore responses to obsolete queries
This adds a query_epoch to notify and log messages, which are
sent in response to queries from the primary during peering. To
guarantee we don't try to process old logs and notifies after
restarting peering, query_epoch is set to the epoch at which the
query was sent. If query_epoch is less than last_peering_reset,
the primary discards the message.
This caused a "bad state machine event" crash in the following
scenario:
1. Primary tells a stray to generate a backlog at epoch 199.
2. The up set changes because a stray goes up.
3. Primary restarts peering at epoch 200.
4. Stray gets new map for epoch 200, sees that acting set did not
change, and sends log to primary.
5. Primary crashes.
Related to #1403, #1449
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Josh Durgin [Thu, 6 Oct 2011 18:52:42 +0000 (11:52 -0700)]
pg: rename warm_restart and last_warm_restart
These names make more sense, since last_warm_restart was updated
outside of the warm_restart function.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
Samuel Just [Thu, 6 Oct 2011 22:59:47 +0000 (15:59 -0700)]
.gitignore: add multi_stress_watch
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Thu, 6 Oct 2011 21:45:09 +0000 (14:45 -0700)]
ReplicatedPG: assert *_FLUSHING unreachable for AccessMode
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Samuel Just [Thu, 6 Oct 2011 21:38:24 +0000 (14:38 -0700)]
test/multi_stress_watch.cc: add watch notify tester for teuthology
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Sage Weil [Thu, 6 Oct 2011 20:56:58 +0000 (13:56 -0700)]
qa: rados-api: try harder to make these pool names unique
Still getting -17 sometimes on sepia... :/
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 6 Oct 2011 20:39:33 +0000 (13:39 -0700)]
xlist: more assertive
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Thu, 6 Oct 2011 20:12:16 +0000 (13:12 -0700)]
Revert "osd: simplify the session ref-counting branches"
This reverts commit
f002144ef7d314e80d3d19a5cd2c4cef38851056 .
Sage Weil [Thu, 6 Oct 2011 19:58:23 +0000 (12:58 -0700)]
osd: boot with map of oldest-1
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 18:41:53 +0000 (11:41 -0700)]
radosgw-admin: UTC for time in 'log show', local time in separate field
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 18:41:27 +0000 (11:41 -0700)]
utime: add gmtime() output function
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 18:01:37 +0000 (11:01 -0700)]
radosgw-admin: for date filtering for 'log list'
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Thu, 6 Oct 2011 17:30:45 +0000 (10:30 -0700)]
Merge branch 'wip-leaks'
Greg Farnum [Thu, 6 Oct 2011 16:58:48 +0000 (09:58 -0700)]
cephx: don't leak Authorizers on each request
It's not clear to me why this is a class member -- it's only
written to or read from in this function, which allocates a fresh
one each time it's called.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 19 Sep 2011 22:51:15 +0000 (15:51 -0700)]
osd: put session ref in ms_handle_reset
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 19 Sep 2011 22:47:14 +0000 (15:47 -0700)]
osd: simplify the session ref-counting branches
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Thu, 6 Oct 2011 16:40:09 +0000 (09:40 -0700)]
rgw: fix chdir result warning
warning: rgw/rgw_main.cc:265: ignoring return value of ‘int chdir(const char*)’, declared with attribute warn_unused_result
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 15:52:35 +0000 (08:52 -0700)]
rgw: fix rgw_switch_url_previx printf arg
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 15:50:50 +0000 (08:50 -0700)]
osd: same_acting_since -> same_interval_since
Interval is (currently) defined as same acting and up sets.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Thu, 6 Oct 2011 04:41:15 +0000 (21:41 -0700)]
osd: fix same_acting_since update
same_acting_since actually means "same acting AND up set since". This is
what the other same_acting_since update does (see
a22511d ).
We should rename it same_interval_since or something similar.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 22:28:37 +0000 (15:28 -0700)]
rgw: use conf for rgw_swift_url[_prefix] instead of fcgi env
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 22:21:32 +0000 (15:21 -0700)]
rgw: use conf for rgw_print_continue instead of fcgi environment
Also remove unused expect100cont bool.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 22:18:35 +0000 (15:18 -0700)]
rgw: use conf for rgw_dns_name instead of fcgi environment
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 22:18:08 +0000 (15:18 -0700)]
rgw: add conf rgw_remote_addr_param instead of using fcgi environment
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 21:13:22 +0000 (14:13 -0700)]
osd: fix pg stats encoding for unfound
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 18:15:01 +0000 (11:15 -0700)]
vstart: use normal name for .asok files
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 20:48:46 +0000 (13:48 -0700)]
rgw: fix initscript default user
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 20:37:10 +0000 (13:37 -0700)]
rgw: daemonize by default
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 20:07:59 +0000 (13:07 -0700)]
rgw: set locator key before calling get_obj_state()
get_obj_state assumes the io_ctx is ready to go. Set the locator key
earlier rather than later.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Wed, 5 Oct 2011 20:00:55 +0000 (13:00 -0700)]
objecter: normalize oloc.key
If oloc.key == oid, clear it. Users can set it to the oid or leave it
blank; they are equivalent.
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Wed, 5 Oct 2011 19:02:10 +0000 (12:02 -0700)]
cli: fix test contents again.
I don't know why this didn't come across properly from my wip branch...
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 18:13:13 +0000 (11:13 -0700)]
osd: fix map epoch check in handle_osd_map
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 18:12:51 +0000 (11:12 -0700)]
paxos: update first_committed
Otherwise this stays at 0, which isn't accurate.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Greg Farnum [Wed, 5 Oct 2011 17:45:23 +0000 (10:45 -0700)]
rgw: fix mis-merge issues.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Wed, 5 Oct 2011 16:54:59 +0000 (09:54 -0700)]
Merge branch 'wip-rgw-pools'
Conflicts:
src/rgw/rgw_bucket.cc
src/rgw/rgw_main.cc
Greg Farnum [Wed, 5 Oct 2011 16:50:51 +0000 (09:50 -0700)]
monclient: add an initialized bool to guard shutdown.
The addition of a Finisher in
9c56070bc20878e87fcb4715b0a3559dd1aaf9ff
broke shutdown in the case where MonClient::init() was never called, so
add a guard variable to keep track.
I'm not sure this is actually the best solution (Timer guard itself,
for instance; maybe Finisher should too?), but I don't want to change
the Finisher interface without looking at it more carefully than I'm
going to right now.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 21:56:42 +0000 (14:56 -0700)]
rgw: remove the rgw_buckets files
We don't need them any more. Hurrah!
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 20:42:19 +0000 (13:42 -0700)]
rgw: remove the rgw_get_bucket_info functions
Push them down into the RGWAccess layers.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 18:23:43 +0000 (11:23 -0700)]
rgw: move rgw_store_bucket_info into RGWRados.
create_bucket is the only user, so just make it a simple helper.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 18:03:19 +0000 (11:03 -0700)]
rgw: remove select_bucket_placement from RGWAccess interface.
RGWRados::create_bucket is the only user now, so make it private
and make the interface a little tighter.
(We are going to need to handle placement at some point in the future,
but the interface needs to be designed a lot more carefully than this
one [wasn't].)
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 17:52:22 +0000 (10:52 -0700)]
rgw: remove rgw_create_bucket.
Push all its extra functionality down into RGWRados::create_bucket. Convert
callers to the different interface (there's no reason to pass in the
bucket name apart from the bucket, and all callers know if they're
using a system bucket or not).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Tue, 4 Oct 2011 16:25:43 +0000 (09:25 -0700)]
rgw: new command to add pools to the list of available places for storage.
Do this by adding an add_bucket_placement function to RGWAccess. Hurray
leaky abstractions! :(
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 03:36:04 +0000 (20:36 -0700)]
Merge branch 'stable'
Sage Weil [Wed, 5 Oct 2011 03:34:27 +0000 (20:34 -0700)]
osd: fix PG::up calculation on startup
We weren't filling in pg->up, which meant we booked an incorrect past
interval in PG::warm_restart(), leading to unknown strangeness down the
line.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Wed, 5 Oct 2011 03:30:15 +0000 (20:30 -0700)]
mon: fix maps sent after osd boot
This was a bad hunk that crept in while I was rebasing wip-oldmaps. :(
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Tue, 4 Oct 2011 23:08:15 +0000 (16:08 -0700)]
Merge branch 'wip-rgw-log'
Sage Weil [Tue, 4 Oct 2011 22:48:24 +0000 (15:48 -0700)]
osd: remove some dead code
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 23:05:34 +0000 (16:05 -0700)]
rgw: write debug acl output to the debug output (not cout)
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 22:56:35 +0000 (15:56 -0700)]
rgw: more sane defaults
Notably, this gives us log_to_stderr = 1.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 22:59:08 +0000 (15:59 -0700)]
common: CINIT_FLAG_NO_BANNER
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 22:40:10 +0000 (15:40 -0700)]
rgw: rgw_log -> debug_rgw
Normalize with the rest of the code base.
Also, 'rgw log' might be confused with 'log file'.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 22:36:15 +0000 (15:36 -0700)]
rgw: remove useless RGWConf::log_level
Not used for anything.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 22:36:01 +0000 (15:36 -0700)]
rgw: set dout condvar to rgw_log
We would normally set DOUT_SUBSYS and get debug_$foo, but the setting in
this case is rgw_log.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 21:59:44 +0000 (14:59 -0700)]
rgw: RGW_LOG -> dout
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 21:08:16 +0000 (14:08 -0700)]
rados-api/test: don't srand(time())
This will give duplicate results if called more than once per second, which
leads to things like -EEXIST (#1558).
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Tue, 4 Oct 2011 21:07:30 +0000 (14:07 -0700)]
mon: make MPoolOp*::print() readable
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Tue, 4 Oct 2011 20:46:52 +0000 (13:46 -0700)]
Merge branch 'wip-oldmaps'
Sage Weil [Mon, 3 Oct 2011 22:13:31 +0000 (15:13 -0700)]
radosgw: run as 'user' field in ceph.conf
Signed-off-by: Sage Weil <sage@newdream.net>
Henry C Chang [Mon, 3 Oct 2011 20:23:31 +0000 (13:23 -0700)]
rgw: set xmlns on ListBucketResult element
Certain S3 clients (e.g., s3fs) require the xmlns to parse the
results correctly.
Signed-off-by: Henry C Chang <henry.cy.chang@gmail.com>
Sage Weil [Tue, 4 Oct 2011 20:33:17 +0000 (13:33 -0700)]
osd: get latest osdmaps before booting
- get the latest osdmaps before adding/marking ourselves up
- behave if there is a discontinuity in the osdmap history
This lets us behave sanely if an osd has been down for a very long time,
or if we replace (wipe) an osd, or otherwise take a fresh new osd and add
it to an aged cluster (with lots of old maps or, more likely, an oldest
map that has a large epoch).
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:20:55 +0000 (13:20 -0700)]
osdmap: include addrs in json dump
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:19:50 +0000 (13:19 -0700)]
osdmap: show weight for out osds
This info is reduncant (weight is 0 for out osds), but makes the output
cleaner.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:18:39 +0000 (13:18 -0700)]
osdmap: show old addrs for down osds
This is useful for admins to find which ip/node to connect to.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:43:29 +0000 (13:43 -0700)]
mon: limit maps send on onetime osdmap subscribe
This throttles the load put on the monitor by making the client request
osdmaps in discrete chunks. Use a feature bit to control this, since the
old kernel clients will only send a single onetime request if they think
they are behind.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:43:25 +0000 (13:43 -0700)]
osdmap: include oldest, newest known epoch in message
If non-zero, these are filled in with the oldest and newest map epochs
available. This helps the client decide whether to ask for newer or older
maps.
Some minor refactoring in the monitor.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 19:42:32 +0000 (12:42 -0700)]
osd: use pointers for handle_advance_map
So that lastmap can be optional.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 22:37:23 +0000 (15:37 -0700)]
objecter: fix locking on is_latest_map checks
We need to take the lock protecting these structures in the completion
context.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 22:36:56 +0000 (15:36 -0700)]
monclient: use an async finisher
Otherwise we can deadlock if any completions call back into MonClient.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 22:29:16 +0000 (15:29 -0700)]
monclient: clean up old version_requests on session reset
This throws out the old contexts. We assume that the caller will not wait
for them.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 22:26:39 +0000 (15:26 -0700)]
monclient: implement get_version() to get newest _and_ oldest map versions
Reimplement is_latest_map() in terms of more general helper. Add field
to reply message.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 20:34:34 +0000 (13:34 -0700)]
osd: is_active() instead of !is_booting() for map sharing checks
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:36:05 +0000 (13:36 -0700)]
osd: do not request osdmap when loner if not booted
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Tue, 4 Oct 2011 20:34:59 +0000 (13:34 -0700)]
mon: trim old osdmaps
Trakc the oldest clean pg, and trim osdmaps that are older than that.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 26 Sep 2011 22:19:18 +0000 (15:19 -0700)]
osd: add last_epoch_clean to pg_stat_t
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 26 Sep 2011 22:00:22 +0000 (15:00 -0700)]
mon: implicitly mark lost_at when a fresh osd rejoins the cluster
If an osd is wiped an rejoins the cluster, record that by implicitly
setting lost_at.
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 22:13:31 +0000 (15:13 -0700)]
radosgw: run as 'user' field in ceph.conf
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 23:26:43 +0000 (16:26 -0700)]
radosgw-admin: make 'log list' behave when .log pool dne
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Sage Weil [Tue, 4 Oct 2011 03:03:25 +0000 (20:03 -0700)]
debian: don't recommend gceph
Not until it's worth installing.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
Greg Farnum [Mon, 3 Oct 2011 22:40:35 +0000 (15:40 -0700)]
rgw: don't specify create_pool and set_marker in create_bucket.
It's wildly inappropriate for that kind of implementation detail to
leak out of the interface. For the moment, leave in a "system_bucket"
parameter, which provides the necessary information. But we should
probably be able to push all of that information down below the
interface layer eventually, and then get rid of that param too.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Mon, 3 Oct 2011 22:23:28 +0000 (15:23 -0700)]
Merge branch 'stable'
Sage Weil [Mon, 3 Oct 2011 22:23:15 +0000 (15:23 -0700)]
radosgw: make stop succeed when not running
This fixes apt-get errors like
No /usr/bin/radosgw found running; none killed.
invoke-rc.d: initscript radosgw, action "stop" failed.
dpkg: warning: subprocess old pre-removal script returned error exit status 1
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Mon, 3 Oct 2011 22:14:02 +0000 (15:14 -0700)]
rgw: move rgw_bucket_select_host_pool behind RGWAccess as select_bucket_placement
This doesn't really belong in front of the interface. Maybe later we
can hide it completely, but for now we can put it behind the RGWAccess
interface in a way that at least pretends to be backing-store-agnostic.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Sage Weil [Mon, 3 Oct 2011 22:02:24 +0000 (15:02 -0700)]
radosgw: fix init-radosgw provides
Signed-off-by: Sage Weil <sage@newdream.net>
Greg Farnum [Mon, 3 Oct 2011 19:02:19 +0000 (12:02 -0700)]
rgw: remove withdraw_pool function.
I think its use was broken anyway?
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 3 Oct 2011 19:01:56 +0000 (12:01 -0700)]
rgw: remove preallocation of pools
Rename rgw_bucket_allocate_pool to rgw_bucket_select_host_pool, since
that better describes functionality we might actually use someday (and
what it's actually doing now).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Greg Farnum [Mon, 3 Oct 2011 17:39:25 +0000 (10:39 -0700)]
rgw: remove preallocating pools maintenance tick
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
Henry C Chang [Mon, 3 Oct 2011 20:23:31 +0000 (13:23 -0700)]
rgw: set xmlns on ListBucketResult element
Certain S3 clients (e.g., s3fs) require the xmlns to parse the
results correctly.
Signed-off-by: Henry C Chang <henry.cy.chang@gmail.com>
Sage Weil [Mon, 3 Oct 2011 20:18:42 +0000 (13:18 -0700)]
debian: add non-stripped exception for libcls_rgw rados class
Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Mon, 3 Oct 2011 18:49:43 +0000 (11:49 -0700)]
Makefile: more radosgw_admin -> radosgw-admin
Signed-off-by: Sage Weil <sage@newdream.net>
Brandon Seibel [Mon, 3 Oct 2011 18:45:17 +0000 (11:45 -0700)]
mds: fix possible deadlock in multi-mds setup
This should fix the file_excl case on a file_max update when there
is more than one mds.
If we dont issue caps here its possible when we eval_gather on
the returning lockack that we don't push the lock forward which
will leave the inode auth pinned. If an export occurs that tries
to freeze a tree this inode is in, we'll have a freezing tree
waiting for an auth_pin that can no longer be removed.
Signed-off-by: Brandon Seibel <brandon@seibelnet.ca>
Josh Durgin [Mon, 3 Oct 2011 19:14:59 +0000 (12:14 -0700)]
workunit: clean up temporary files
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>