RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-12 06:00:46 +00:00
Code Issues Packages Projects Releases Wiki Activity
9,044 Commits 214 Branches 394 Tags 1.3 GiB
032d9751e0
Commit Graph

9044 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sage Weil
05bd3f712f mds: don't renew caps (or touch_session) on closed session 
Also assert that we're on the state list before touching.
 2009-11-24 19:02:06 -08:00
Sage Weil
e66778b7d6 mds: less --, more - 2009-11-24 19:02:06 -08:00
Sage Weil
cd6f381710 mds: update pending_for_mds during table server replay 
Otherwise we our replay of AGREE after recovery is wrong; we just replay
what was saved, without taking additional journaled items into account.
 2009-11-24 19:02:06 -08:00
Sage Weil
fd21693fa6 mon: add 'mds set_state <gid> <stateid>' 
For mdsmon disaster recovery :)
 2009-11-24 19:02:05 -08:00
Sage Weil
1ac56b7d7a mon: avoid mds state update when clearing laggy flag 
Do not clear laggy flag (or otherwise get into prepare_update) if the mds
hasn't seen the latest map.  Previously we could go to clear laggy and also
revert to an old mds state.
 2009-11-24 19:02:05 -08:00
Yehuda Sadeh
e6e8f87891 vstart.sh: fix -k for mds loading 2009-11-23 16:01:23 -08:00
Yehuda Sadeh
efe705cf9a vstart.sh: fix -k (cmds init) 2009-11-23 16:01:23 -08:00
Sage Weil
3c9fa270b6 msgr: fix bad unlock on tcp_write error 2009-11-23 14:23:13 -08:00
Sage Weil
cf28787c7b mds: fix replay of hard links 
Use dir->link_remote_inode, not raw function, so that the CDir accounting
gets set up.
 2009-11-23 13:52:07 -08:00
Sage Weil
6d10f068d9 mds: clear session before remove_session 
remove_session may drop the last ref.
 2009-11-20 22:02:38 -08:00
Sage Weil
1ee8308c74 mds: do not eval subtree root during rename_prepare 
The eval may lead to a scatter_writebehind, which screws up the journal
ordering.  Since the inode is also locked, we will end up evaling it
later anyway, so don't worry about it.

I'm a bit unsure about the adjust_subtree_after_rename... that is during
the rename_apply stage (after the log entry committed), so i think it's
safe to do a scatter_writebehind here.. since we're not between a predirty
and submit_entry.
 2009-11-20 21:42:04 -08:00
Sage Weil
b62daac29d todo 2009-11-20 16:22:34 -08:00
Sage Weil
83f560cc32 mds: fix client purge crash 
set_state moves the session to the per-session lists; do
this _before_ remove_session (which removes the session
entirely).  Otherwise we fail an assert when deleting
the session:

mds/SessionMap.h: In function 'virtual Session::~Session()':
mds/SessionMap.h:141: FAILED assert(!session_list_item.is_on_xlist())
 1: ./cmds(_Z18__ceph_assert_failPKcS0_iS0_+0x36) [0x97fc87]
 2: ./cmds(_ZN7SessionD0Ev+0x47) [0x73d817]
 3: ./cmds(_ZN16RefCountedObject3putEv+0xdb) [0x712c3b]
 4: ./cmds(_ZN10ConnectionD0Ev+0x3a) [0x713f88]
 5: ./cmds(_ZN10Connection3putEv+0x3d) [0x71294b]
 6: ./cmds(_ZN15SimpleMessenger4PipeD1Ev+0x7f) [0x7130e1]
 7: ./cmds(_ZN15SimpleMessenger6reaperEv+0x228) [0x7048ba]
 8: ./cmds(_ZN15SimpleMessenger4waitEv+0x28) [0x7059dc]
 9: ./cmds(main+0x305) [0x6f9345]
 10: /lib/libc.so.6(__libc_start_main+0xe6) [0x7f2091e715c6]
 11: ./cmds [0x6f8e39]
 NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
 2009-11-20 16:20:02 -08:00
Sage Weil
bf3cd2803c mon: fix use-after-free in mds mon blacklisting 2009-11-20 14:07:16 -08:00
Sage Weil
a7e5be7aa5 mds: warn if replay used ino doesn't match session's next 2009-11-20 13:40:38 -08:00
Sage Weil
29efb0b563 mds: fix reconnect race 
Don't ignore client replay or session msgs if we are
moving to reconnect state.  The client may get the mdsmap
before us and send things our way, so take them if we
are reconnect, or if we want reconnect.
 2009-11-20 13:40:38 -08:00
Sage Weil
780ee5a299 mds: remove dead session request list trim waiter code 
I don't even remember what this was originally for.
 2009-11-20 13:40:38 -08:00
Sage Weil
3c8defef2d mds: associate session with connection; clean up 
Sessions always created on connect.  May or may not be
in then sessionmap, depending on the session open/close
commands.
 2009-11-20 13:40:38 -08:00
Sage Weil
d7364a4810 vstart: include [mds.$name] for auth_none 2009-11-20 13:40:38 -08:00
Sage Weil
69ff349eb3 osd: don't crash on (and allow) messages sent to self 2009-11-20 13:38:08 -08:00
Yehuda Sadeh
32c81a2871 conf: don't crash if conf file not specified and not found 2009-11-20 10:05:47 -08:00
Sage Weil
dad2011d25 filestore: fix mkfs with no jourhal 2009-11-20 09:54:51 -08:00
Yehuda Sadeh
16c8c5b8b9 auth: change error code when not supporting authentication 2009-11-19 16:50:59 -08:00
Sage Weil
ddf47d8c82 rgw: ifdef out dead code 2009-11-19 16:15:45 -08:00
Sage Weil
e648986751 auth: set initial values in misc structs 
Fixes a bunch of valgrind warnings.
 2009-11-19 16:03:30 -08:00
Sage Weil
ba9fd9a8f1 sepia: no valgrind on osds for now 2009-11-19 15:32:28 -08:00
Sage Weil
b580653f20 mon: allow mds injectargs on recovering mdss 2009-11-19 15:31:42 -08:00
Sage Weil
23eaa6d28c mds: use global_id as key in MDSMap 
This makes the data structures less stupid (using an
entity_addr_t as a key is silly).

This is a protocol and disk format change.
 2009-11-19 15:30:27 -08:00
Sage Weil
f35f8fef15 filestore: don't magically enable journal if ${osd_data}.journal exists 2009-11-19 14:37:23 -08:00
Sage Weil
bf58196eeb rados: kill unused ceph_fsid_set 2009-11-19 14:37:23 -08:00
Yehuda Sadeh
fec31964a1 auth: when renewing session, encrypt ticket 2009-11-19 14:20:20 -08:00
Sage Weil
6a34c6a282 rgw: fix base64 warnings (move impl to .c file) 2009-11-19 12:44:55 -08:00
Sage Weil
b05e184c07 mon: only remove_session if !s->closed 2009-11-19 12:41:04 -08:00
Sage Weil
beac937bc9 atomic: assert valid ref counts 
Make sure we catch ref counting errors!
 2009-11-19 12:41:04 -08:00
Sage Weil
0b85f09b9d auth: initialize EntityName type in constructor 2009-11-19 12:41:04 -08:00
Sage Weil
bf709f5425 mon: fix session ref leak 2009-11-19 12:41:04 -08:00
Sage Weil
2cc368c58b monc: restart auth handshake each time we connect to mon 2009-11-19 12:42:29 -08:00
Sage Weil
68b5da9e10 mds: don't change directory size/rbytes from client cap update 
This is wrong, and screws up the recursive accounting.
 2009-11-19 11:24:48 -08:00
Sage Weil
8f38d0e98b auth: add ull to large constant 
so that we build on 32bit platforms
 2009-11-19 11:05:17 -08:00
Sage Weil
8696cd90f4 mds: fix inode rstat debug output 2009-11-19 11:05:04 -08:00
Sage Weil
dda61fe1c3 auth: add missing virtual destructors 2009-11-19 10:57:01 -08:00
Sage Weil
55d7bf7194 osdc: use __u64, not size_t 
size_t is too small on 32 bit platforms.
 2009-11-19 10:11:19 -08:00
Sage Weil
d8f3c4fcca mds: do not crash if m->connection == NULL 2009-11-19 09:44:47 -08:00
Sage Weil
34a57c154c mds: do not send MMDSMap to ourselves 2009-11-19 09:44:47 -08:00
Sage Weil
082aa7c6ac add ceph_fsid_set helper 2009-11-19 09:46:08 -08:00
Sage Weil
2cb286f993 filejournal: use single shared zeroed buffer for alignment padding 
Avoids lots of mallocs and memsets (one for every journal event).
 2009-11-18 16:25:35 -08:00
Sage Weil
9a2606d1a2 buffer: add create_static constructor 2009-11-18 16:25:35 -08:00
Yehuda Sadeh
9a4083e142 vstart.sh: fix -k 2009-11-18 16:20:18 -08:00
Sage Weil
2e2eaa6b1f Merge commit 'd00b917143e2afd72851bfd0400b2c2a1f3fc018' into auth 2009-11-18 16:01:20 -08:00
Sage Weil
053db04667 vstart: fix errors on auth none 2009-11-18 16:00:42 -08:00