Sage Weil
949f1bf48a
kclient: prefix frag_ functions with ceph_
...
It was suggested this would avoid any potential conflict with TCP.
It's a good idea to keep the namespace clean anyway.
2009-09-22 11:50:29 -07:00
Sage Weil
c6faf6a15f
kclient: kill out_qlen
...
This is apparently buggy (not immediately obvious why) but it's
only really useful for debugging anyway (it makes it easier to
match up message sequence numbers). Rip it out.
2009-09-22 11:27:32 -07:00
Sage Weil
1744628de2
make inline string functions non-inline
2009-09-22 11:17:22 -07:00
Sage Weil
c2b4dc3d05
kclient: whitespace
2009-09-22 10:17:40 -07:00
Sage Weil
ab3c421eb9
v0.15
2009-09-22 10:13:11 -07:00
Sage Weil
c6e5d6e432
Makefile: include ioctl-number.patch
2009-09-22 10:10:40 -07:00
Sage Weil
1f1ec71ab1
msgr: don't print source in msg debug output
2009-09-22 10:10:27 -07:00
Sage Weil
13e6f345cf
todo
2009-09-22 10:00:31 -07:00
Sage Weil
48866c2f94
monclient: change pick_new_mon -> reopen_session
2009-09-22 09:59:23 -07:00
Sage Weil
88fd161523
monclient: avoid dup subscribe messages
2009-09-22 09:58:07 -07:00
Sage Weil
d8c8da8649
osd: subscribe to monmap
2009-09-22 09:55:21 -07:00
Sage Weil
cf3192dd39
todo
2009-09-21 16:52:40 -07:00
Sage Weil
0a338c1d9d
kclient: fix le encoding bugs (yay sparse)
2009-09-21 16:52:19 -07:00
Sage Weil
f5b46bf0fa
kclient: clean whitespace
2009-09-21 16:49:56 -07:00
Sage Weil
ee04c5657c
kclient: include object name in DATALOC ioctl result
2009-09-21 16:40:00 -07:00
Sage Weil
1af730352d
kclient: include ioctl-number doc file in patch set
2009-09-21 16:26:50 -07:00
Sage Weil
c0c3444e96
kclient: no need for static ceph_buffer
2009-09-21 16:26:35 -07:00
Sage Weil
0f22324d86
mds: fix mds standby map handling
...
Borken by c6ddbdb467
2009-09-21 15:11:26 -07:00
Sage Weil
84c1a14488
mon: consolidate sub request msgs; cleaner api
...
Caller must call renew_subs() explicitly when they want to start
the subscription(s).
2009-09-21 14:55:59 -07:00
Sage Weil
98e6da670f
mon: allow sub for monmap
2009-09-21 14:45:45 -07:00
Sage Weil
c2c9ecd1dd
mon: fix up subscribe infrastructure to use msgr connection for sessions
...
Session state is now cleaned up when connections drop. More
efficient in-memory structures.
2009-09-21 14:30:19 -07:00
Sage Weil
f65d127ee3
msgr: pass Connection* to msgr error handlers
...
This lets the error handlers get at session state (if any).
2009-09-21 14:24:35 -07:00
Sage Weil
35dc047356
mon: fix client id preallocation
...
Broken by b2c7655093
2009-09-21 14:06:15 -07:00
Sage Weil
9a45979b08
monclient: send a single mount request
2009-09-21 14:05:57 -07:00
Sage Weil
316f716d05
uclient: use ref in constructor for MClientRequest
2009-09-21 12:01:06 -07:00
Sage Weil
5dfe8fae0c
kclient: two msgpool modes: blocking (ala mempool_t) and !blocking (strict)
...
Only warn on !blocking pool depletion!
2009-09-18 16:42:09 -07:00
Sage Weil
84aec7d312
kclient: fix symlink kmalloc under spinlock
2009-09-18 16:32:26 -07:00
Sage Weil
71b2ee0f2c
strerror -> strerror_r
2009-09-18 16:17:04 -07:00
Sage Weil
6f00f11155
kclient: rename msgpool type
2009-09-18 14:57:35 -07:00
Sage Weil
c517d7f53c
kclient: use msgpools for osd requests, replies
2009-09-18 14:54:04 -07:00
Sage Weil
ebc6e6cba4
uclient: fix null pointer ref
2009-09-18 14:18:43 -07:00
Sage Weil
61948498cf
kclient: fix up ceph_client init error handling
2009-09-18 14:15:44 -07:00
Sage Weil
f0f10d6309
kclient: fix patchset comment
2009-09-18 14:05:10 -07:00
Sage Weil
d0efb5a7f5
kclient: fall back to WARN + new allocation if msgpool is depleted
2009-09-18 14:04:50 -07:00
Sage Weil
3be2b62b3f
kclient: use msgpools for other monc messages
2009-09-18 13:08:25 -07:00
Sage Weil
d1ec687925
kclient: add back msgpools
...
This reverts commit b985882aa2
2009-09-18 13:01:22 -07:00
Sage Weil
dbfb0f0deb
kclient: add GET_DATALOC ioctl to get object location, osd addr
2009-09-18 12:45:48 -07:00
Greg Farnum
3fa407eae8
uclient: _removexattr should be naming the xattr to remove!
2009-09-17 18:29:24 -07:00
Greg Farnum
9843d1fdd4
uclient: Add functions/data members for preemptive cap dropping.
2009-09-17 18:29:24 -07:00
Greg Farnum
271de28237
MClientRequest: Releases should have a constructor
2009-09-17 18:29:24 -07:00
Greg Farnum
370223027b
uclient: First pass at encode_cap_release function; not used.
2009-09-17 18:29:24 -07:00
Greg Farnum
3393d55cc1
uclient: Move the MetaRequest to MClientRequest into its own code, and
...
cut out the payload encode/decode since MClientRequests are now made
from MetaRequests.
Oh, and there are no issues with rebuilding messages -- false alarm!
2009-09-17 18:29:24 -07:00
Greg Farnum
33c50adc74
uclient: The basic passing mechanism is now a MetaRequest. It seems to work.
...
More bug-testing is needed, though, as rebuilding messages in send_message
is finicky at best...
2009-09-17 18:29:24 -07:00
Greg Farnum
04f4911d4a
uclient: Make the uclient work properly with the new MetaRequest structure.
2009-09-17 18:29:24 -07:00
Greg Farnum
ae16b1cd15
uclient: MetaRequests are going to be how we pass requests around;
...
repurpose them so they don't require an MClientRequest.
ALSO, convert the mds member from a map to an int.
2009-09-17 18:29:24 -07:00
Greg Farnum
fabaeb99c2
Hadoop: Minor fixes and style changes to satisfy Hadoop's FindBugs system.
...
Remove a few deprecated methods.
2009-09-17 18:29:24 -07:00
Sage Weil
24c84ab494
kclient: fix osd request kicking
...
We often need to kick the request to the same osd. Keep track of
an osd incarnation, incremented each time we reset_osd or see the
connection drop, and kick requests when the incarnation doesn't
match.
2009-09-17 16:15:21 -07:00
Sage Weil
b2c7655093
mon: fix bug with client id assignment of -1
2009-09-17 15:57:06 -07:00
Sage Weil
28b2df36a2
kclient: avoid no-op dir cap release when no dentry lease
...
Make release logic a bit smarter: only force inclusion of dir
cap release if there is a dentry lease to start with.
2009-09-17 15:56:43 -07:00
Sage Weil
1d8a91ec68
kclient: make mds console msgs consistent
2009-09-17 15:29:09 -07:00
