Commit Graph

24378 Commits

Author SHA1 Message Date
Danny Al-Gaaf
a68411bd52 os/FileStore.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 16:04:24 +01:00
Danny Al-Gaaf
f1f1c77697 SimpleMessenger.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 15:54:21 +01:00
Danny Al-Gaaf
5009c77d60 msg/Pipe.cc: reduce scope of some variables
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 15:40:51 +01:00
Danny Al-Gaaf
bb172dd467 mon/Paxos.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 15:09:37 +01:00
Danny Al-Gaaf
8bfe3c6f89 mon/PGMonitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:37:29 +01:00
Danny Al-Gaaf
5376a139fe mon/PGMap.cc: remove useless assigment to 'first', it's never used
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:33:39 +01:00
Danny Al-Gaaf
e13da07a5f mon/OSDMonitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:32:09 +01:00
Danny Al-Gaaf
9b73448b07 mon/MonmapMonitor.cc: remove unused variable
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:25:55 +01:00
Danny Al-Gaaf
4d5e3dd167 mon/MonmapMonitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:22:20 +01:00
Danny Al-Gaaf
e2fe6f00ec mon/Monitor.cc: reduce scope of reuse_caps in Monitor::_ms_dispatch()
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:18:04 +01:00
Danny Al-Gaaf
0b73c31814 mon/Monitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:15:08 +01:00
Danny Al-Gaaf
0213806233 mon/MonClient.cc: use static_cast instead of C-Style cast
Use static_cast<T*>(m) instead of C-Style cast, add missing
include needed for static_cast.

Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 14:02:15 +01:00
Danny Al-Gaaf
f26fad2ab8 mon/MDSMonitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:57:39 +01:00
Danny Al-Gaaf
956408eaba mon/LogMonitor.cc: move unused variable into related code block
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:53:54 +01:00
Danny Al-Gaaf
b232621bc1 mon/LogMonitor.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:51:57 +01:00
Danny Al-Gaaf
17bd157632 mon/Elector.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:49:37 +01:00
Danny Al-Gaaf
00b309cf93 mon/AuthMonitor.cc: remove unused 'std::stringstream ss'
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:47:10 +01:00
Danny Al-Gaaf
17aa1b4092 mds/flock.cc: remove unused variable 'crossed_waiting_locks'
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:41:48 +01:00
Danny Al-Gaaf
eedb463846 mds/Server.cc: reduce scope of some variables
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:38:11 +01:00
Danny Al-Gaaf
116d8d1fc0 mds/Server.cc: : use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:33:12 +01:00
Danny Al-Gaaf
146e457502 mds/Migrator.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:27:03 +01:00
Danny Al-Gaaf
cb31159de3 mds/MDS.cc: remove unused variable 'oldresolve'
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:20:39 +01:00
Danny Al-Gaaf
7def48c306 mds/MDS.cc: : use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:20:03 +01:00
Danny Al-Gaaf
926e9960a8 mds/MDCache.cc: remove unused variables recover_q and check_q
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:10:32 +01:00
Danny Al-Gaaf
87a1773cf6 mds/MDCache.cc: reduce scope of 'in' in MDCache::open_root_inode()
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:05:32 +01:00
Danny Al-Gaaf
ec0e6cb64d mds/MDCache.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 13:03:19 +01:00
Danny Al-Gaaf
f40932d670 mds/MDBalancer.h: init some double variables in constructor
Initialize double variables my_load and target_load with 0.0 in
the constructor.

Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 12:49:37 +01:00
Danny Al-Gaaf
a8be638f7d mds/MDBalancer.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 12:40:01 +01:00
Danny Al-Gaaf
eabd0b8831 ESession.h: initialze 'open' in the constructor with false
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 12:33:37 +01:00
Danny Al-Gaaf
b6305c1e94 mds/LogEvent.cc: use static_cast instead of C-Style cast
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 12:15:13 +01:00
Danny Al-Gaaf
82f233b574 mds/Locker.cc: reduce scope of 'pool' in Locker::issue_client_lease()
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 12:11:01 +01:00
Danny Al-Gaaf
c02e317f53 mds/Locker.cc: use static_cast instead of C-Style cast
Use static_cast<T>() instead of C-Style cast.

Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 11:41:06 +01:00
Danny Al-Gaaf
95cfffb260 mds/CDir.h: use static_cast instead of C-Style cast
Use static_cast<T>() instead of C-Style cast.

Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-03-01 11:22:23 +01:00
Gary Lowell
96e9da1f54 ceph.spec.in: gdisk for for rpm based systems instead of gptfdisk.
Signed-off-by: Gary Lowell  <gary.lowell@inktank.com>
2013-02-28 22:31:47 -08:00
Sage Weil
862c761554 Merge branch 'next' 2013-02-28 16:58:02 -08:00
Sage Weil
0f42eddef5 msgr: drop messages on cons with CLOSED Pipes
Back in commit 6339c5d439, we tried to make
this deal with a race between a faulting pipe and new messages being
queued.  The sequence is

- fault starts on pipe
- fault drops pipe_lock to unregister the pipe
- user (objecter) queues new message on the con
- submit_message reopens a Pipe (due to this bug)
- the message managed to make it out over the wire
- fault finishes faulting, calls ms_reset
- user (objecter) closes the con
- user (objecter) resends everything

It appears as though the previous patch *meant* to drop *m on the floor in
this case, which is what this patch does.  And that fixes the crash I am
hitting; see #4271.

Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
2013-02-28 16:57:42 -08:00
Sage Weil
2a84437f86 osd: add 'osd debug op order' to track and assert order of client ops
Keep a big fat slow map in memory to ensure that client ops are applied to
objects in the order the clien sent them.  This will make it easier to
catch ordering bugs due to requeuing as soon as they happen, when the logs
are easier to wade through, and not later when the client notices something
is awry.

Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
2013-02-28 16:22:46 -08:00
Sage Weil
2ed9f5a96f osd: include osdmap epoch in osd_op message operator<<
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
2013-02-28 16:22:46 -08:00
Sage Weil
0ad717aa4d osd: useful debugging for when we drop requests
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Samuel Just <sam.just@inktank.com>
2013-02-28 16:22:46 -08:00
Samuel Just
d6724525e1 crush-map.rst: fix chooseleaf syntax in ssd example
Fixes: 4304
Backport: bobtail
Reviewed-by: Greg Farnum <greg@inktank.com>
Signed-off-by: Samuel Just <sam.just@inktank.com>
2013-02-28 15:50:48 -08:00
Sage Weil
6a9cda7ec4 Merge pull request #84 from dalgaaf/wip-da-sca-style
Some more smaller fixes

Reviewed-by: Sage Weil <sage@inktank.com>
2013-02-28 14:59:19 -08:00
Sage Weil
ffc0ff68e0 init-ceph: kick off ceph-create-keys when starting mons
We don't have a particularly tidy way to kick off ceph-create-keys from
sysvinit.  So, do something simple, ugly, and functional: if we are
starting a mon, and it looks like a 'new' type that is in /var/lib/ceph,
then start ceph-create-keys too.

Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Gary Lowell <gary.lowell@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
2013-02-28 13:45:04 -08:00
Sage Weil
1e0b2d8952 init-ceph: fix default admin socket path used for status
Signed-off-by: Sage Weil <sage@inktank.com>
2013-02-28 13:44:52 -08:00
Sage Weil
0cbe406f93 osd: show retry attempt in MOSDOp operator<<
Signed-off-by: Sage Weil <sage@inktank.com>
2013-02-28 13:33:40 -08:00
Sage Weil
c47c02dd87 msg/Pipe: allow tuning of TCP receive buffer size
Performance tests on high-end machines have indicated the Linux autotuning
of the receive buffer sizes can cause throughput collapse.  See bug
#2100, and this email discussion:

   http://marc.info/?l=ceph-devel&m=133009796706284&w=2

Initially default to 0, which leaves us with the default.  We may adjust
the default in the future.

Tested-by: Jim Schutt <jaschut@sandia.gov>
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
2013-02-28 13:01:27 -08:00
Sage Weil
e10c1d1453 msg/Pipe: move setting of socket options into a common method
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
2013-02-28 13:00:19 -08:00
Samuel Just
5d54ab154c FileJournal::wrap_read_bl: adjust pos before returning
Otherwise, we may feed an offset past the end of the journal to
check_header in read_entry and incorrectly determine that the entry is
corrupt.

Fixes: 4296
Backport: bobtail
Backport: argonaut
Reviewed-by: Sage Weil <sage@inktank.com>
Signed-off-by: Samuel Just <sam.just@inktank.com>
2013-02-28 11:08:27 -08:00
Danny Al-Gaaf
6cba92450f librbd/internal.cc: remove unused variable total_write
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-02-28 19:45:29 +01:00
Danny Al-Gaaf
c12d91e9e2 librbd/internal.cc: reduce scope of 'int r'
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-02-28 19:41:37 +01:00
Danny Al-Gaaf
bed61fcacb librados/librados.cc: don't check unsigned 'len' against less than zero
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2013-02-28 19:34:47 +01:00