]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Florian Haas [Mon, 2 Jul 2012 14:54:08 +0000 (16:54 +0200)]
doc: drop mention of MDS capabilities from radosgw man page
Signed-off-by: Florian Haas <florian@hastexo.com>
Sage Weil [Fri, 29 Jun 2012 05:25:14 +0000 (22:25 -0700)]
Merge pull request #16 from fghaas/doc-osd
doc: document osd tell bench
Florian Haas [Thu, 28 Jun 2012 19:07:22 +0000 (21:07 +0200)]
doc: document osd tell bench
Signed-off-by: Florian Haas <florian@hastexo.com>
John Wilkins [Thu, 28 Jun 2012 18:41:08 +0000 (11:41 -0700)]
doc: minor edits.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Samuel Just [Thu, 28 Jun 2012 18:27:07 +0000 (11:27 -0700)]
buffer.h: fix operator> and operator>= increment error
Signed-off-by: Samuel Just <sam.just@inktank.com>
Mike Ryan [Wed, 27 Jun 2012 21:14:30 +0000 (14:14 -0700)]
pg: report scrub status
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Mike Ryan [Wed, 27 Jun 2012 20:30:45 +0000 (13:30 -0700)]
pg: track who we are waiting for maps from
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Mike Ryan [Tue, 26 Jun 2012 23:25:27 +0000 (16:25 -0700)]
pg: reduce scrub write lock window
Wait for all replicas to construct the base scrub map before finalizing
the scrub and locking out writes.
Signed-off-by: Mike Ryan <mike.ryan@inktank.com>
Yehuda Sadeh [Wed, 27 Jun 2012 00:16:11 +0000 (17:16 -0700)]
rest-bench: mark request as complete later
We marked a request as complete in the callback, however
it might be that we're still inside S3_runall_request_context()
which means that request is not really complete yet.
Possibly fixes bug #2652.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Wed, 27 Jun 2012 00:28:51 +0000 (17:28 -0700)]
rgw-admin: use correct modifier with strptime
Bug #2658: used %I (12h) instead of %H (24h)
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
John Wilkins [Tue, 26 Jun 2012 19:30:01 +0000 (12:30 -0700)]
doc: removing old 'todo' configuration reference.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Tue, 26 Jun 2012 19:24:23 +0000 (12:24 -0700)]
doc: converted daemon references from nasty tables to lines.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Tue, 26 Jun 2012 15:28:10 +0000 (08:28 -0700)]
doc: a few minor tweaks to radosgw to cleanup navigation.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Tue, 26 Jun 2012 15:21:05 +0000 (08:21 -0700)]
doc: Published RADOS GW docs and install as is for now. Needs more work.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
caleb miles [Tue, 26 Jun 2012 07:05:27 +0000 (00:05 -0700)]
CrushTester.cc: removed last broken bits of obsolete method
Signed-off-by: caleb miles <caleb.miles@inktank.com>
caleb miles [Tue, 26 Jun 2012 07:02:24 +0000 (00:02 -0700)]
CrushTester.cc: removed obsolete method for collapsing a CRUSH map
Signed-off-by: caleb miles <caleb.miles@inktank.com>
caleb miles [Tue, 26 Jun 2012 06:58:52 +0000 (23:58 -0700)]
CrushTester: eliminated compiler warning
Signed-off-by: caleb miles <caselim@gmail.com>
caleb miles [Tue, 26 Jun 2012 06:14:09 +0000 (23:14 -0700)]
Merge branch 'master' of github.com:ceph/ceph
Yehuda Sadeh [Thu, 21 Jun 2012 22:40:27 +0000 (15:40 -0700)]
rgw: send both swift x-storage-token and x-auth-token
older clients need x-storage-token, newer x-auth-token
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 21 Jun 2012 22:17:19 +0000 (15:17 -0700)]
rgw: radosgw-admin date params now also accept time
The date format now is "YYYY-MM-DD[ hh:mm:ss]". Got rid of
the --time param for the old ops log stuff.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Thu, 21 Jun 2012 20:14:47 +0000 (13:14 -0700)]
rgw-admin: fix usage help
s/show/trim
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Tommi Virtanen [Mon, 25 Jun 2012 22:14:33 +0000 (15:14 -0700)]
upstart: fix regex
Signed-off-by: Tommi Virtanen <tv@inktank.com>
Signed-off-by: Greg Farnum <greg@inktank.com>
Sage Weil [Mon, 25 Jun 2012 21:45:21 +0000 (14:45 -0700)]
Merge branch 'next'
Greg Farnum [Mon, 25 Jun 2012 20:04:15 +0000 (13:04 -0700)]
mon: don't tick the PaxosServices if we are currently slurping.
They aren't prepared to deal with the on-disk state being inconsistent.
Signed-off-by: Greg Farnum <greg@inktank.com>
Sage Weil [Wed, 20 Jun 2012 18:07:29 +0000 (11:07 -0700)]
objecter: do not feed session to op_submit()
The linger_send() method was doing this, but it is problematic because the
new Op doesn't get its pgid or acting vector set correctly. The result is
that the request goes to the right OSD, but has the wrong pgid, and makes
the OSD complain about misdirected requests and drop it on the floor. It
didn't affect the test results because we weren't testing whether the
watch was working in that case.
Instead, we'll just recalculate and get the same value the parent linger
op did. Which is fine, and goes through all the usual code paths so
nothing is missed.
Also, increment num_homeless_ops before we recalc_op_target(), so that we
don't (harmlessly, but confusingly) underflow.
Fixes: #2022
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Sun, 24 Jun 2012 20:30:53 +0000 (13:30 -0700)]
ObjectStore::Transaction: initialize pool_override in all constructors
use_pool_override and pool_override weren't initialized in these two
constructors.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Wed, 20 Jun 2012 18:07:29 +0000 (11:07 -0700)]
objecter: do not feed session to op_submit()
The linger_send() method was doing this, but it is problematic because the
new Op doesn't get its pgid or acting vector set correctly. The result is
that the request goes to the right OSD, but has the wrong pgid, and makes
the OSD complain about misdirected requests and drop it on the floor. It
didn't affect the test results because we weren't testing whether the
watch was working in that case.
Instead, we'll just recalculate and get the same value the parent linger
op did. Which is fine, and goes through all the usual code paths so
nothing is missed.
Also, increment num_homeless_ops before we recalc_op_target(), so that we
don't (harmlessly, but confusingly) underflow.
Fixes: #2022
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Fri, 22 Jun 2012 00:08:20 +0000 (17:08 -0700)]
osd_types.cc: remove hobject_t decode asserts
These asserts were useful for ensuring that pool is passed
in in the correct places, but they prevent the encoder
testing from working.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Thu, 21 Jun 2012 14:33:47 +0000 (07:33 -0700)]
mon: note that monmap may be reencoded later
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 14:31:47 +0000 (07:31 -0700)]
mon: encoding new monmap using quorum feature set
It is probably unlikely that someone will expand the mon cluster with a
mixed feature set, but we know the quorum features here, so we should use
them.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 14:27:49 +0000 (07:27 -0700)]
mon: conditionally encode mon features for remote mon
The only time we encode these is when forwarding messages. Encoding using
the destination's feature set.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 14:23:56 +0000 (07:23 -0700)]
mon: conditionally encode PGMap[::Incremental] with quorum features
This allows a mon cluster to transition to the new encoding during a
rolling upgrade.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 03:41:17 +0000 (20:41 -0700)]
mon: conditionally encode auth incremental with quorum feature bits
If the quorum does not yet all have the MONENC feature, stick to the old
encoding.
It might be more polite to require a super-quorum before switching over,
and take note so that thereafter we can stick to the new encoding, but
that has more moving parts and I'm not sure it's worth the complexity.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 03:33:41 +0000 (20:33 -0700)]
mon: track intersection of quorum member features
When we form a quorum, also note the intersection of the quorum members'
feature bits. This will inform decisions about what encodings we use.
This is an imperfect strategy because the quorum may change, and we may
have a mon with old code join in and not understand what is going on.
However, it does ensure that a majority of the members run new code, so in
the absence of other failures we can make progress.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 02:08:34 +0000 (19:08 -0700)]
mon: conditionally encode old monmap when peer lacks feature
This allows a rolling upgrade from 0.47.2 to 0.48.
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Wed, 20 Jun 2012 19:55:38 +0000 (12:55 -0700)]
OSD,PG,ObjectStore: handle messages with old hobject_t encoding
Messages that embed an hobject_t need to have the pool field fixed
on messages from old peers.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Travis Rhoden [Thu, 21 Jun 2012 19:25:02 +0000 (15:25 -0400)]
doc: ceph osd crush add is now ceph osd crush set
Mailing list thread: http://www.spinics.net/lists/ceph-devel/msg06199.html
Signed-off-by: Travis Rhoden <trhoden@gmail.com>
Sage Weil [Thu, 21 Jun 2012 19:42:53 +0000 (12:42 -0700)]
logrotate: reload all upstart instances
upstart doesn't let you wildcard all instances of a given job, so we
slog through initctl list output, and reload any running daemons.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Tommi Virtanen <tv@inktank.com>
Yehuda Sadeh [Thu, 21 Jun 2012 18:34:20 +0000 (11:34 -0700)]
docs: clarify example in radosgw-admin
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Thu, 21 Jun 2012 18:17:34 +0000 (11:17 -0700)]
filestore: remove btrfs start/end transaction ioctl code
This was an ill-conceived approach to getting atomic transactions out of
btrfs. It doesn't offer rollback, which means that any error means we need
to wedge the file system and reboot in order to avoid corrupting the
data set. And that's silly!
Snapshots are more robust and only marginally slower (because we have to
quiesce our writes while waiting for the snap to start, and btrfs resume
work in-kernel slightly faster...maybe).
Fixes: #2623
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 18:16:39 +0000 (11:16 -0700)]
mon: make pool deletion dne error code consistently ENOENT
This matches the check in _prepare_remove_pool() now.
Signed-off-by: Sage Weil <sage@inktank.com>
Yehuda Sadeh [Thu, 21 Jun 2012 17:46:55 +0000 (10:46 -0700)]
docs: update usage logging info
added some text under radosgw, examples under radosgw-admin
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Thu, 21 Jun 2012 15:20:17 +0000 (08:20 -0700)]
Merge remote-tracking branch 'gh/stable' into next
John Wilkins [Wed, 20 Jun 2012 22:10:23 +0000 (15:10 -0700)]
doc: Normalized shell script syntax. Added generic cookbook path.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Wed, 20 Jun 2012 21:29:58 +0000 (14:29 -0700)]
doc: Changed libvirt-dev to libvirt-bin, and cleaned up ./autogen.sh
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Tommi Virtanen [Wed, 20 Jun 2012 19:55:29 +0000 (12:55 -0700)]
doc: Typo.
Signed-off-by: Tommi Virtanen <tv@inktank.com>
Sage Weil [Wed, 20 Jun 2012 17:57:41 +0000 (10:57 -0700)]
v0.47.3
Sage Weil [Fri, 15 Jun 2012 17:00:54 +0000 (10:00 -0700)]
filestore: disable 'filestore fiemap' by default
We've seen this failing on both btrfs (Guido) and XFS (Oliver). This works
around #2535.
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Tue, 19 Jun 2012 21:29:48 +0000 (14:29 -0700)]
OSD: clear_temp: split delete into many transactions
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 19 Jun 2012 22:19:21 +0000 (15:19 -0700)]
Merge branch 'wip-clsrbd'
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Yehuda Sadeh [Tue, 19 Jun 2012 22:17:53 +0000 (15:17 -0700)]
doc: document usage log
Modify radosgw, radosgw-admin man page
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Tue, 19 Jun 2012 21:31:21 +0000 (14:31 -0700)]
cls_rbd: note overlap units
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 20:08:06 +0000 (13:08 -0700)]
cls_rbd: fix uninitialized var in cls_rbd_parent
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 14:50:46 +0000 (07:50 -0700)]
cls_rbd: clean up doxygen, whitespace a bit
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 00:41:04 +0000 (17:41 -0700)]
cls_rbd: include dumpers, test instances for our types
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 00:29:35 +0000 (17:29 -0700)]
cls_rbd: rename parent size -> overlap
The use of 'size' to describe the amount that we overlap with the parent
was confusing; overlap is more intuitive (to me, at least!).
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 00:27:25 +0000 (17:27 -0700)]
cls_rbd: handle parent overlap
Two things:
* when we set a parent, the overlap is the min of our size and the
parent's size.
* when we resize, our overlap may shrink.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 00:03:47 +0000 (17:03 -0700)]
cls_rbd: make snapshots inherit the head's parent
When you take a snapshot, the snap gets the current parent of the head, if
any.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 15 Jun 2012 17:42:49 +0000 (10:42 -0700)]
cls_rbd: check for LAYERING feature for parent methods
You need to set the feature on the image before you can use these methods.
We *could* allow get_parent, but that just means they'll get ENOENT instead
of ENOEXEC, and the latter is more informative.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 18 Jun 2012 21:22:53 +0000 (14:22 -0700)]
cls_rbd: implement get_parent, set_parent, remove_parent
Implement methods to get/set/remove the parent pointer on child images.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 15 Jun 2012 22:38:27 +0000 (15:38 -0700)]
cls_rbd_client: use snapid_t type
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 19 Jun 2012 21:29:43 +0000 (14:29 -0700)]
Merge remote-tracking branch 'gh/next'
Conflicts:
src/osd/PG.cc
Yehuda Sadeh [Mon, 18 Jun 2012 21:44:38 +0000 (14:44 -0700)]
rgw: set s->header_ended before flushing formatter
otherwise we don't account the formatter in s->bytes_sent.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Mon, 18 Jun 2012 21:28:25 +0000 (14:28 -0700)]
rgw: log user and not bucket owner for service operations
For operations that are done on the service (e.g., list buckets)
we need to log the user that did the operation, and not the bucket
owner.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Yehuda Sadeh [Mon, 18 Jun 2012 21:27:51 +0000 (14:27 -0700)]
rgw: initalize s->enable_usage_log
Missing initialization, we ended up not logging every operation.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
Sage Weil [Tue, 19 Jun 2012 17:12:40 +0000 (10:12 -0700)]
osd: use derr (instead of cerr) for convertfs
This will appear in the log *and* stderr (if we're running in the
foreground).
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 19 Jun 2012 17:11:01 +0000 (10:11 -0700)]
osd: close stderr on daemonize
This spams stderr in an ugly way. Users should look at the logs.
In particular, filestore upgrades spam the console, which is unpleasant.
Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Tue, 19 Jun 2012 16:11:57 +0000 (09:11 -0700)]
PG: improve find_best_info
07f853db3982e68b952a337cf91cbf7ec0709de9 is actually too conservative,
it suffices to find any info with a last_update of at least the least
last_update from the last period to go active. An info from a previous
interval is acceptable if the last interval never reported a commited
operation and thus still has the same last_update.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Tue, 19 Jun 2012 16:39:01 +0000 (09:39 -0700)]
osdcap: make grammar build on spirit 2.2/boost 1.42
This is a stupid workaround, but it makes the compilation error go away.
Signed-off-by: Sage Weil <sage@inktank.com>
John Wilkins [Tue, 19 Jun 2012 17:15:35 +0000 (10:15 -0700)]
doc: Added a snapshot page with outline. Needs additional info.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
John Wilkins [Tue, 19 Jun 2012 17:10:51 +0000 (10:10 -0700)]
doc: Added prerequisite of running Ceph cluster.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Samuel Just [Mon, 18 Jun 2012 16:26:12 +0000 (09:26 -0700)]
PG: reg_last_pg_scrub on pg resurrection
This may solve the unreg_last_pg_scrub assert.
see #2453.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just [Mon, 18 Jun 2012 21:02:28 +0000 (14:02 -0700)]
ceph_osd: move auto-upgrade to after fork
Signed-off-by: Samuel Just <sam.just@inktank.com>
John Wilkins [Mon, 18 Jun 2012 21:29:27 +0000 (14:29 -0700)]
Merge branch 'master' of github.com:ceph/ceph
Sage Weil [Mon, 18 Jun 2012 21:11:37 +0000 (14:11 -0700)]
librbd: simplify refresh sequence checks
- incremental seq each request
- set last to the last completed seq
- drop the now-redundant bool
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
John Wilkins [Mon, 18 Jun 2012 21:29:04 +0000 (14:29 -0700)]
doc: first cut of OpenStack/RBD w/out cephx. Authentication coming soon.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Sage Weil [Mon, 18 Jun 2012 21:07:20 +0000 (14:07 -0700)]
filestore: make disk format upgrade warning less scary, more informative
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Sam Just <sam.just@inktank.com>
Sage Weil [Mon, 18 Jun 2012 21:02:29 +0000 (14:02 -0700)]
mon: include quorum in ceph status
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 18 Jun 2012 21:00:06 +0000 (14:00 -0700)]
mon: gracefully handle slow 'ceph -w' clients
If we are sending log updates to a client (ceph -w), and they are far
enough behind to drop behind first_committed, include a friendly message
in their stream but continue.
Drop useless return value from _create_sub_incremental(). Assert that we
can read the state file.
Signed-off-by: Sage Weil <sage@inktank.com>
John Wilkins [Mon, 18 Jun 2012 20:22:18 +0000 (13:22 -0700)]
doc: Added libvirt installation.
Signed-off-by: John Wilkins <john.wilkins@inktank.com>
Samuel Just [Sat, 16 Jun 2012 00:09:42 +0000 (17:09 -0700)]
PG: best_info must have a last_epoch_started as high as any other info
We disregard incomplete infos during find_best_info, but we can't an
info with a last_epoch_started less that of the incomplete info.
This should avoid cases like #2462. In that case, it appears that
a peer with empty info/log was chosen as authoritative even though
there was a non-empty incomplete peer.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Tommi Virtanen [Mon, 18 Jun 2012 17:36:37 +0000 (10:36 -0700)]
doc: btrfs and parted chef recipes are not needed.
They were removed in ceph-cookbooks.git commit
a77b418b95df8f2b6f56d9a90e47dc62100cdf4d .
Signed-off-by: Tommi Virtanen <tv@inktank.com>
Sage Weil [Sun, 17 Jun 2012 21:00:02 +0000 (14:00 -0700)]
Merge branch 'next'
Laszlo Boszormenyi (GCS) [Sat, 16 Jun 2012 20:49:41 +0000 (13:49 -0700)]
debian: fix python-ceph depends
Signed-off-by: Laszlo Boszormenyi (GCS) <gcs@debian.hu>
Laszlo Boszormenyi (GCS) [Sat, 16 Jun 2012 20:39:20 +0000 (13:39 -0700)]
debian: update homepage url
Signed-off-by: Laszlo Boszormenyi (GCS) <gcs@debian.hu>
Sage Weil [Sun, 17 Jun 2012 20:20:59 +0000 (13:20 -0700)]
filestore: fix 'omap' collection skipping
The if/else if/... structure was skipping this test if the file system
didn't support d_type.
Fixes: #2598
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 15 Jun 2012 21:48:22 +0000 (14:48 -0700)]
run-cli-test: use new pip incantation
http://www.pip-installer.org/en/latest/news.html#id1
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 15 Jun 2012 21:48:22 +0000 (14:48 -0700)]
run-cli-test: use new pip incantation
http://www.pip-installer.org/en/latest/news.html#id1
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 17 Jun 2012 16:07:41 +0000 (09:07 -0700)]
cls_rbd: do not pass snapid_t to vargs
On squeeze,
warning: cls_rbd.cc:534: cannot pass objects of non-POD type ‘struct snapid_t’ through ‘...’; call will abort at runtime
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sun, 17 Jun 2012 03:09:04 +0000 (20:09 -0700)]
msg: fix buffer overflow in ipv6 addr parsing
Noticed because of failing i386 unit tests for long addrs; x86_64 passed
fine. Sigh. FTR, the failing address was
2001:0db8:85a3:0000:0000:8a2e:0370:7334
Sadly the full length addrs don't turn it up on x86_64, still, nor does
valgrind notice. But, this fixes it on i386.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Sat, 16 Jun 2012 21:50:29 +0000 (14:50 -0700)]
ceph.newdream.net -> ceph.com
wiki url -> docs url
Sage Weil [Sat, 16 Jun 2012 14:42:17 +0000 (07:42 -0700)]
Merge branch 'next'
Conflicts:
src/cls_rbd.cc
Sage Weil [Sat, 16 Jun 2012 14:33:19 +0000 (07:33 -0700)]
cls_rbd: drop useless snapshot metadata helpers
Now that cls_rbd_snap is encodable, we don't need these helpers; get_key()
will suffice.
Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 14 Jun 2012 23:12:49 +0000 (16:12 -0700)]
cls_rbd: use encode macros for on-disk snap metadata
This will let us version this encoding later when we add new information
and features, like a per-snap parent.
Signed-off-by: Sage Weil <sage@inktank.com>
Greg Farnum [Sat, 16 Jun 2012 00:39:48 +0000 (17:39 -0700)]
Merge branch 'wip-rbd-locking'
Conflicts:
src/cls_rbd.cc
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Samuel Just [Sat, 16 Jun 2012 00:09:42 +0000 (17:09 -0700)]
PG: best_info must have a last_epoch_started as high as any other info
We disregard incomplete infos during find_best_info, but we can't an
info with a last_epoch_started less that of the incomplete info.
This should avoid cases like #2462. In that case, it appears that
a peer with empty info/log was chosen as authoritative even though
there was a non-empty incomplete peer.
Signed-off-by: Samuel Just <sam.just@inktank.com>
Greg Farnum [Sat, 16 Jun 2012 00:26:04 +0000 (17:26 -0700)]
librbd: remove the set-r-and-return idiom from some functions.
Josh hates this, and I have to admit it seems silly.
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum [Thu, 14 Jun 2012 21:18:25 +0000 (14:18 -0700)]
librbd: add cooperative image locking to the C and C++ interfaces
While we're at it, increment the version number.
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum [Thu, 14 Jun 2012 20:43:29 +0000 (13:43 -0700)]
librbd: include the image lockers in the ImageCtx mutable metadata
Signed-off-by: Greg Farnum <greg@inktank.com>
Greg Farnum [Tue, 12 Jun 2012 22:12:49 +0000 (15:12 -0700)]
test: add rbd image locking tests to test_cls_rbd
Signed-off-by: Greg Farnum <greg@inktank.com>