Greg Farnum
213ffe2079
mon: set mon_caps with authenticated monitor check instead of claimed one.
2010-03-25 11:42:45 -07:00
Greg Farnum
cd7edcafd4
mon: Fix all uses of Session and get_caps.
...
They now use MonSession and get_service()->caps, respectively.
2010-03-25 11:03:11 -07:00
Greg Farnum
5c9aa8804b
mon: Change name of Session to MonSession.
2010-03-25 11:02:21 -07:00
Greg Farnum
1045738077
msg: Add get_session function to PaxosServiceMessage
2010-03-25 11:02:21 -07:00
Greg Farnum
c69a03391d
Revert "msg: PaxosServiceMessage includes a MonCap*"
...
This reverts commit 167b28a0d4
2010-03-25 11:02:20 -07:00
Greg Farnum
ef5e8740a7
mon: MonCaps default initializes more variables
2010-03-25 10:55:02 -07:00
Sage Weil
4ecd615a22
osd: drop useless ENOMEM check
...
'new' will throw an exception if it fails, not return NULL.
2010-03-25 10:20:30 -07:00
Sage Weil
95eb2fc3c9
vstart.sh: include mon secret during mkfs
2010-03-25 10:20:27 -07:00
Sage Weil
33a02af0eb
mon: fix up authorizer generation/validation
...
Only generate authorizer for other mons. Only require authorizer for other
mons when CEPHX is enabled. Check it properly (before we were letting
mons connect without any authorizer!).
2010-03-25 10:20:22 -07:00
Sage Weil
3273581cc1
mds: fix bug in handle_discover on mdsdir
2010-03-24 16:02:40 -07:00
Sage Weil
59b4c60af1
mds: check myin, not stray, in rejoin_send_acks
2010-03-24 16:02:37 -07:00
Sage Weil
94794ab2e0
mon: change forwarding behavior somewhat
2010-03-24 15:50:31 -07:00
Sage Weil
d3d243c2bb
mon: initialize routed_request_tid
2010-03-24 15:24:07 -07:00
Sage Weil
9dfc4fb240
msgr: set m->connection for messages sent to self
2010-03-24 14:25:46 -07:00
Sage Weil
9656a758d8
osd: detect TMAPUP no-op and do no work
2010-03-24 11:55:17 -07:00
Sage Weil
1916e29308
debug: include timestamp + thread in _all_ debug output
2010-03-24 11:54:44 -07:00
Sage Weil
864b1a40b6
qa: fix loopall.sh mode
2010-03-24 11:50:23 -07:00
Sage Weil
8e069f64d2
todo
2010-03-24 11:50:17 -07:00
Sage Weil
c2d6abaa5e
todo mds bug
2010-03-23 16:47:59 -07:00
Greg Farnum
0eeec6f5d0
mon: Hack in setting monitor caps
2010-03-23 15:26:15 -07:00
Sage Weil
bf4a1b5656
fix up READMEs
2010-03-23 14:54:33 -07:00
Sage Weil
41f6adc9d4
mds: fix up client session importing
...
Keep import counter for each session, for overlapping imports.
Prevent a session close or kill during an import.
2010-03-23 14:54:13 -07:00
Greg Farnum
9733882e4b
mon: Fix initialization-crossing goto jump
2010-03-23 14:03:40 -07:00
Sage Weil
b932727617
mds: importing, ugh
2010-03-23 13:39:38 -07:00
Sage Weil
c99207df19
mds: clean up kill_session()
2010-03-23 13:32:58 -07:00
Sage Weil
74d638a59f
mds: possible session 'importing' state
2010-03-23 13:20:12 -07:00
Sage Weil
205e8836df
mds: drop purge step when closing client sessions
...
The purge step when closing client sessions is pointless. If the
MDS is alive, we know which inos were used and which weren't. It
is only the _journal_ that doesn't necessarily know which have
been recently allocated: any cleanup needs to happen on journal
replay during mds recovery.
Also clean up session states some more.
2010-03-23 13:15:58 -07:00
Greg Farnum
772daf213a
mon: Removing dead code; been commented out for 5 months
2010-03-23 12:40:55 -07:00
Greg Farnum
efd344d454
mon: remove dead 'magic' code
2010-03-23 12:27:16 -07:00
Sage Weil
d77185eb34
todo
2010-03-23 12:13:14 -07:00
Sage Weil
c965812f86
mds: fix bug in handle_discover on mdsdir
2010-03-23 12:13:14 -07:00
Sage Weil
f3ea95bd65
mds: check myin, not stray, in rejoin_send_acks
2010-03-23 12:13:14 -07:00
Sage Weil
328c9908ec
mds: fix eval_gather for sync->mix(2) and mix->sync(2)
...
These states never move to next state on the client on their own.
The auth mds needs to tell them to switch states.
2010-03-23 12:13:14 -07:00
Sage Weil
f1a59b0688
mds: encode fnode when replicating CDir
2010-03-23 12:13:14 -07:00
Greg Farnum
fbdbdc093f
mon: addition to MonCaps docs mentioning how perms are used
2010-03-23 11:41:52 -07:00
Greg Farnum
fdddc1e9ca
mon: remove unused label out, which was causing compiler warning
2010-03-23 11:41:52 -07:00
Greg Farnum
5fe3af8fd0
mon: remove now-unused macros.
2010-03-23 11:41:52 -07:00
Greg Farnum
99172e2776
mon: MRoutes are now checked by check_privileges instead of a macro
2010-03-23 11:41:52 -07:00
Greg Farnum
029a56bcf5
mon: MForwards are handled by check_privileges instead of a macro.
2010-03-23 11:41:52 -07:00
Greg Farnum
e4f623e08d
mon: handling MSG_CLASS now uses caps->check_privileges instead of macros.
...
Also, delete a leaked message.
2010-03-23 11:41:52 -07:00
Greg Farnum
7b11014102
mon: use check_privileges instead of macro for MSG_MON_ELECTION
2010-03-23 11:41:52 -07:00
Greg Farnum
7d7d2aad59
mon: MSG_MON_OBSERVE handled in-function instead of with macro, sends response
2010-03-23 11:41:52 -07:00
Greg Farnum
298073f5f3
mon: don't use macros in evaluating MSG_MON_PAXOS messages
2010-03-23 11:41:52 -07:00
Greg Farnum
45f8072b52
mon: MSG_LOG uses an in-handler cap check instead of a macro
2010-03-23 11:41:52 -07:00
Greg Farnum
ccb5cff408
mon: Pool ops are already cap-checked by the handler; remove macro
2010-03-23 11:41:52 -07:00
Greg Farnum
6818a01518
mon: add mon_caps used for unconnected monitor messages
2010-03-23 11:41:52 -07:00
Greg Farnum
aeab0f650e
mon: stat requests use caps->check_privileges instead of a macro
2010-03-23 11:41:52 -07:00
Greg Farnum
7e70afbe2a
mon: the MDS messages use caps->check_privileges instead of the macro
2010-03-23 11:41:51 -07:00
Greg Farnum
48e64a35ba
mon: MRemoveSnaps uses caps->check_privileges
2010-03-23 11:41:51 -07:00
Greg Farnum
1ad850f142
mon: OSD handling messages use caps->check_privileges instead of macros
2010-03-23 11:41:51 -07:00
