]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
15 years agoobjecter: optionally set priority on ObjectOperations
Sage Weil [Thu, 24 Sep 2009 21:56:39 +0000 (14:56 -0700)]
objecter: optionally set priority on ObjectOperations

This just sets the priority on the MOSDOp we send.

15 years agomds: do not attach mdr to NULL session
Sage Weil [Thu, 24 Sep 2009 21:10:57 +0000 (14:10 -0700)]
mds: do not attach mdr to NULL session

15 years agotodo
Sage Weil [Wed, 23 Sep 2009 17:11:56 +0000 (10:11 -0700)]
todo

15 years agomds: warn on dup inodes; don't crash
Sage Weil [Wed, 23 Sep 2009 17:00:32 +0000 (10:00 -0700)]
mds: warn on dup inodes; don't crash

We just ignore the second instance of the inode.  It can get
cleaned up transparently if/when the directory gets saved.

15 years agoobjecter: remove dead code
Sage Weil [Tue, 22 Sep 2009 22:12:18 +0000 (15:12 -0700)]
objecter: remove dead code

15 years agolibrados: pass resets to objecter, with proper locking
Sage Weil [Tue, 22 Sep 2009 22:11:47 +0000 (15:11 -0700)]
librados: pass resets to objecter, with proper locking

15 years agomds: pass resets to objecter, with proper locking
Sage Weil [Tue, 22 Sep 2009 22:11:34 +0000 (15:11 -0700)]
mds: pass resets to objecter, with proper locking

15 years agouclient: pass reset events to objecter, with proper locking
Sage Weil [Tue, 22 Sep 2009 22:11:25 +0000 (15:11 -0700)]
uclient: pass reset events to objecter, with proper locking

15 years agoobjecter: resend mon ops on handle mon connection reset
Sage Weil [Tue, 22 Sep 2009 22:11:10 +0000 (15:11 -0700)]
objecter: resend mon ops on handle mon connection reset

15 years agomds: fix inode load existence check
Sage Weil [Tue, 22 Sep 2009 22:37:51 +0000 (15:37 -0700)]
mds: fix inode load existence check

We can lookup using last snapid, not first.  That's how the
inode_map is indexed.

15 years agotodo
Sage Weil [Tue, 22 Sep 2009 21:18:47 +0000 (14:18 -0700)]
todo

15 years agouclient: don't readdir first frag on opendir
Sage Weil [Tue, 22 Sep 2009 21:17:50 +0000 (14:17 -0700)]
uclient: don't readdir first frag on opendir

This reorders things such that rm -r breaks.  It was an old old
optimization that doesn't make sense anymore anyway.

15 years agoobjecter: fix uninitialized variable last_seen_version
Sage Weil [Tue, 22 Sep 2009 20:17:03 +0000 (13:17 -0700)]
objecter: fix uninitialized variable last_seen_version

Fixes hang on 'rados df'

15 years agolibrados: fix dispatch
Sage Weil [Tue, 22 Sep 2009 20:09:09 +0000 (13:09 -0700)]
librados: fix dispatch

15 years agouclient: Better fix for insert_dentry_inode to work with pre-built Dentrys.
Greg Farnum [Tue, 22 Sep 2009 00:27:59 +0000 (17:27 -0700)]
uclient: Better fix for insert_dentry_inode to work with pre-built Dentrys.

15 years agouclient: get_or_create should return out-of-date lease Dentrys too
Greg Farnum [Tue, 22 Sep 2009 00:07:20 +0000 (17:07 -0700)]
uclient: get_or_create should return out-of-date lease Dentrys too

15 years agouclient: encode_cap_releases shouldn't update caps if inode isn't in the cache
Greg Farnum [Tue, 22 Sep 2009 00:06:14 +0000 (17:06 -0700)]
uclient: encode_cap_releases shouldn't update caps if inode isn't in the cache

15 years agouclient: First attempt at preemptive cache dropping in the functions.
Greg Farnum [Tue, 22 Sep 2009 00:04:32 +0000 (17:04 -0700)]
uclient: First attempt at preemptive cache dropping in the functions.

15 years agouclient: Don't *always* set_dentry_wanted() on MClientRequests.
Greg Farnum [Mon, 21 Sep 2009 21:13:19 +0000 (14:13 -0700)]
uclient: Don't *always* set_dentry_wanted() on MClientRequests.

15 years agouclient: Only make the MClientRequest in send_request itself.
Greg Farnum [Mon, 21 Sep 2009 20:46:37 +0000 (13:46 -0700)]
uclient: Only make the MClientRequest in send_request itself.
Add a few more functions to MetaRequest so we can use those in choose_mds.

15 years agouclient: More debugging. May want to back off the inode printout to
Greg Farnum [Mon, 21 Sep 2009 20:11:52 +0000 (13:11 -0700)]
uclient: More debugging. May want to back off the inode printout to
not include caps_issued.

15 years agouclient: changed link and insert_dentry_inode to behave with pre-built dentrys
Greg Farnum [Mon, 21 Sep 2009 18:13:55 +0000 (11:13 -0700)]
uclient: changed link and insert_dentry_inode to behave with pre-built dentrys

15 years agouclient: Attempt preemptive cap dropping on mknod.
Greg Farnum [Fri, 18 Sep 2009 22:44:45 +0000 (15:44 -0700)]
uclient: Attempt preemptive cap dropping on mknod.

15 years agouclient: dout statements in new functions.
Greg Farnum [Fri, 18 Sep 2009 21:21:58 +0000 (14:21 -0700)]
uclient: dout statements in new functions.

15 years agouclient: Fixes for get_or_create
Greg Farnum [Fri, 18 Sep 2009 22:20:29 +0000 (15:20 -0700)]
uclient: Fixes for get_or_create

15 years agouclient: Functions to create Dentrys prior to gettin caps from mds.
Greg Farnum [Fri, 18 Sep 2009 20:51:26 +0000 (13:51 -0700)]
uclient: Functions to create Dentrys prior to gettin caps from mds.

15 years agoremove extra whitespace
Sage Weil [Tue, 22 Sep 2009 19:29:33 +0000 (12:29 -0700)]
remove extra whitespace

15 years agokclient: fix bit test logic
Sage Weil [Tue, 22 Sep 2009 19:26:52 +0000 (12:26 -0700)]
kclient: fix bit test logic

This fixes the fault messages with <NULL> in them; we now print
the proper message, and in a timely fashion.

15 years agokclient: use pr_fmt to prefix module name
Sage Weil [Tue, 22 Sep 2009 19:15:33 +0000 (12:15 -0700)]
kclient: use pr_fmt to prefix module name

This doubles up the ceph:ceph: if we have CONFIG_DYNAMIC_DEBUG, but
that's not our fault.

15 years agokclient: prefix frag_ functions with ceph_
Sage Weil [Tue, 22 Sep 2009 18:50:29 +0000 (11:50 -0700)]
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.

15 years agokclient: kill out_qlen
Sage Weil [Tue, 22 Sep 2009 18:27:32 +0000 (11:27 -0700)]
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.

15 years agomake inline string functions non-inline
Sage Weil [Tue, 22 Sep 2009 18:17:22 +0000 (11:17 -0700)]
make inline string functions non-inline

15 years agokclient: whitespace
Sage Weil [Tue, 22 Sep 2009 17:17:40 +0000 (10:17 -0700)]
kclient: whitespace

15 years agov0.15 v0.15
Sage Weil [Tue, 22 Sep 2009 17:13:11 +0000 (10:13 -0700)]
v0.15

15 years agoMakefile: include ioctl-number.patch
Sage Weil [Tue, 22 Sep 2009 17:10:40 +0000 (10:10 -0700)]
Makefile: include ioctl-number.patch

15 years agomsgr: don't print source in msg debug output
Sage Weil [Tue, 22 Sep 2009 17:10:27 +0000 (10:10 -0700)]
msgr: don't print source in msg debug output

15 years agotodo
Sage Weil [Tue, 22 Sep 2009 17:00:31 +0000 (10:00 -0700)]
todo

15 years agomonclient: change pick_new_mon -> reopen_session
Sage Weil [Tue, 22 Sep 2009 16:59:23 +0000 (09:59 -0700)]
monclient: change pick_new_mon -> reopen_session

15 years agomonclient: avoid dup subscribe messages
Sage Weil [Tue, 22 Sep 2009 16:58:07 +0000 (09:58 -0700)]
monclient: avoid dup subscribe messages

15 years agoosd: subscribe to monmap
Sage Weil [Tue, 22 Sep 2009 16:55:21 +0000 (09:55 -0700)]
osd: subscribe to monmap

15 years agotodo
Sage Weil [Mon, 21 Sep 2009 23:52:40 +0000 (16:52 -0700)]
todo

15 years agokclient: fix le encoding bugs (yay sparse)
Sage Weil [Mon, 21 Sep 2009 23:52:19 +0000 (16:52 -0700)]
kclient: fix le encoding bugs (yay sparse)

15 years agokclient: clean whitespace
Sage Weil [Mon, 21 Sep 2009 23:49:56 +0000 (16:49 -0700)]
kclient: clean whitespace

15 years agokclient: include object name in DATALOC ioctl result
Sage Weil [Mon, 21 Sep 2009 23:40:00 +0000 (16:40 -0700)]
kclient: include object name in DATALOC ioctl result

15 years agokclient: include ioctl-number doc file in patch set
Sage Weil [Mon, 21 Sep 2009 23:26:50 +0000 (16:26 -0700)]
kclient: include ioctl-number doc file in patch set

15 years agokclient: no need for static ceph_buffer
Sage Weil [Mon, 21 Sep 2009 23:26:35 +0000 (16:26 -0700)]
kclient: no need for static ceph_buffer

15 years agomds: fix mds standby map handling
Sage Weil [Mon, 21 Sep 2009 22:11:07 +0000 (15:11 -0700)]
mds: fix mds standby map handling

Borken by c6ddbdb4671b3594f9b4d5f54b65f8525935758b

15 years agomon: consolidate sub request msgs; cleaner api
Sage Weil [Mon, 21 Sep 2009 21:55:59 +0000 (14:55 -0700)]
mon: consolidate sub request msgs; cleaner api

Caller must call renew_subs() explicitly when they want to start
the subscription(s).

15 years agomon: allow sub for monmap
Sage Weil [Mon, 21 Sep 2009 21:45:45 +0000 (14:45 -0700)]
mon: allow sub for monmap

15 years agomon: fix up subscribe infrastructure to use msgr connection for sessions
Sage Weil [Mon, 21 Sep 2009 21:30:19 +0000 (14:30 -0700)]
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.

15 years agomsgr: pass Connection* to msgr error handlers
Sage Weil [Mon, 21 Sep 2009 21:24:35 +0000 (14:24 -0700)]
msgr: pass Connection* to msgr error handlers

This lets the error handlers get at session state (if any).

15 years agomon: fix client id preallocation
Sage Weil [Mon, 21 Sep 2009 21:06:15 +0000 (14:06 -0700)]
mon: fix client id preallocation

Broken by b2c7655093821710aac2b4d35ccb1740429eb402

15 years agomonclient: send a single mount request
Sage Weil [Mon, 21 Sep 2009 21:05:57 +0000 (14:05 -0700)]
monclient: send a single mount request

15 years agouclient: use ref in constructor for MClientRequest
Sage Weil [Fri, 18 Sep 2009 04:41:04 +0000 (21:41 -0700)]
uclient: use ref in constructor for MClientRequest

15 years agokclient: two msgpool modes: blocking (ala mempool_t) and !blocking (strict)
Sage Weil [Fri, 18 Sep 2009 23:41:38 +0000 (16:41 -0700)]
kclient: two msgpool modes: blocking (ala mempool_t) and !blocking (strict)

Only warn on !blocking pool depletion!

15 years agokclient: fix symlink kmalloc under spinlock
Sage Weil [Fri, 18 Sep 2009 23:32:26 +0000 (16:32 -0700)]
kclient: fix symlink kmalloc under spinlock

15 years agostrerror -> strerror_r
Sage Weil [Fri, 18 Sep 2009 23:17:04 +0000 (16:17 -0700)]
strerror -> strerror_r

15 years agokclient: rename msgpool type
Sage Weil [Fri, 18 Sep 2009 21:57:35 +0000 (14:57 -0700)]
kclient: rename msgpool type

15 years agokclient: use msgpools for osd requests, replies
Sage Weil [Fri, 18 Sep 2009 21:54:04 +0000 (14:54 -0700)]
kclient: use msgpools for osd requests, replies

15 years agouclient: fix null pointer ref
Sage Weil [Fri, 18 Sep 2009 21:18:43 +0000 (14:18 -0700)]
uclient: fix null pointer ref

15 years agokclient: fix up ceph_client init error handling
Sage Weil [Fri, 18 Sep 2009 21:15:44 +0000 (14:15 -0700)]
kclient: fix up ceph_client init error handling

15 years agokclient: fix patchset comment
Sage Weil [Fri, 18 Sep 2009 21:05:10 +0000 (14:05 -0700)]
kclient: fix patchset comment

15 years agokclient: fall back to WARN + new allocation if msgpool is depleted
Sage Weil [Fri, 18 Sep 2009 21:04:50 +0000 (14:04 -0700)]
kclient: fall back to WARN + new allocation if msgpool is depleted

15 years agokclient: use msgpools for other monc messages
Sage Weil [Fri, 18 Sep 2009 20:08:25 +0000 (13:08 -0700)]
kclient: use msgpools for other monc messages

15 years agokclient: add back msgpools
Sage Weil [Fri, 18 Sep 2009 20:01:22 +0000 (13:01 -0700)]
kclient: add back msgpools

This reverts commit b985882aa27fea1d1ed92b09b6057be961c56c0a.

15 years agokclient: add GET_DATALOC ioctl to get object location, osd addr
Sage Weil [Fri, 18 Sep 2009 19:45:48 +0000 (12:45 -0700)]
kclient: add GET_DATALOC ioctl to get object location, osd addr

15 years agouclient: _removexattr should be naming the xattr to remove!
Greg Farnum [Fri, 18 Sep 2009 00:25:04 +0000 (17:25 -0700)]
uclient: _removexattr should be naming the xattr to remove!

15 years agouclient: Add functions/data members for preemptive cap dropping.
Greg Farnum [Fri, 18 Sep 2009 00:22:42 +0000 (17:22 -0700)]
uclient: Add functions/data members for preemptive cap dropping.

15 years agoMClientRequest: Releases should have a constructor
Greg Farnum [Thu, 17 Sep 2009 18:16:33 +0000 (11:16 -0700)]
MClientRequest: Releases should have a constructor

15 years agouclient: First pass at encode_cap_release function; not used.
Greg Farnum [Wed, 16 Sep 2009 00:32:01 +0000 (17:32 -0700)]
uclient: First pass at encode_cap_release function; not used.

15 years agouclient: Move the MetaRequest to MClientRequest into its own code, and
Greg Farnum [Tue, 15 Sep 2009 20:08:25 +0000 (13:08 -0700)]
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!

15 years agouclient: The basic passing mechanism is now a MetaRequest. It seems to work.
Greg Farnum [Tue, 15 Sep 2009 18:40:52 +0000 (11:40 -0700)]
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...

15 years agouclient: Make the uclient work properly with the new MetaRequest structure.
Greg Farnum [Mon, 14 Sep 2009 23:31:35 +0000 (16:31 -0700)]
uclient: Make the uclient work properly with the new MetaRequest structure.

15 years agouclient: MetaRequests are going to be how we pass requests around;
Greg Farnum [Mon, 14 Sep 2009 22:40:14 +0000 (15:40 -0700)]
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.

15 years agoHadoop: Minor fixes and style changes to satisfy Hadoop's FindBugs system.
Greg Farnum [Mon, 14 Sep 2009 17:46:39 +0000 (10:46 -0700)]
Hadoop: Minor fixes and style changes to satisfy Hadoop's FindBugs system.
Remove a few deprecated methods.

15 years agokclient: fix osd request kicking
Sage Weil [Thu, 17 Sep 2009 23:15:21 +0000 (16:15 -0700)]
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.

15 years agomon: fix bug with client id assignment of -1
Sage Weil [Thu, 17 Sep 2009 22:57:06 +0000 (15:57 -0700)]
mon: fix bug with client id assignment of -1

15 years agokclient: avoid no-op dir cap release when no dentry lease
Sage Weil [Thu, 17 Sep 2009 22:56:43 +0000 (15:56 -0700)]
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.

15 years agokclient: make mds console msgs consistent
Sage Weil [Thu, 17 Sep 2009 22:29:09 +0000 (15:29 -0700)]
kclient: make mds console msgs consistent

15 years agomonc: pick monitor before sending initial mount
Sage Weil [Thu, 17 Sep 2009 21:56:34 +0000 (14:56 -0700)]
monc: pick monitor before sending initial mount

15 years agouclient: implement basic I_COMPLETE support
Sage Weil [Thu, 17 Sep 2009 21:54:36 +0000 (14:54 -0700)]
uclient: implement basic I_COMPLETE support

15 years agouclient: factor out command cap issue checks
Sage Weil [Thu, 17 Sep 2009 21:25:55 +0000 (14:25 -0700)]
uclient: factor out command cap issue checks

rdcache_gen, shared_gen

15 years agomds: minor comments
Sage Weil [Thu, 17 Sep 2009 21:16:05 +0000 (14:16 -0700)]
mds: minor comments

15 years agokclient: move some inode update code inside i_lock
Sage Weil [Thu, 17 Sep 2009 21:15:53 +0000 (14:15 -0700)]
kclient: move some inode update code inside i_lock

15 years agokclient: use shared_gen, not cache_gen, for I_COMPLETE tracking
Sage Weil [Thu, 17 Sep 2009 21:15:39 +0000 (14:15 -0700)]
kclient: use shared_gen, not cache_gen, for I_COMPLETE tracking

15 years agotodo
Sage Weil [Thu, 17 Sep 2009 20:47:35 +0000 (13:47 -0700)]
todo

15 years agomonclient: small cleanup
Sage Weil [Wed, 16 Sep 2009 19:12:43 +0000 (12:12 -0700)]
monclient: small cleanup

15 years agotodo
Sage Weil [Wed, 16 Sep 2009 19:12:34 +0000 (12:12 -0700)]
todo

15 years agokclient: fix console msg
Sage Weil [Thu, 17 Sep 2009 20:15:29 +0000 (13:15 -0700)]
kclient: fix console msg

15 years agokclient: move caps stats into client debugfs dir; fix error checking
Sage Weil [Thu, 17 Sep 2009 19:04:28 +0000 (12:04 -0700)]
kclient: move caps stats into client debugfs dir; fix error checking

15 years agokclient: put fsid in debugfs dir name
Sage Weil [Thu, 17 Sep 2009 18:28:53 +0000 (11:28 -0700)]
kclient: put fsid in debugfs dir name

15 years agomds: make parallel_fetch work with hard/remote links
Sage Weil [Thu, 17 Sep 2009 18:58:06 +0000 (11:58 -0700)]
mds: make parallel_fetch work with hard/remote links

We need to open remote ino's, and possibly add to gather (like
a fetch).  Required some refactoring.  Killed/renamed
otherwise unused traverse_to_dir.

15 years agocosd.conf: fix date stamp
Sage Weil [Thu, 17 Sep 2009 17:50:14 +0000 (10:50 -0700)]
cosd.conf: fix date stamp

15 years agomsgr: fix null deref setting message connection ref
Sage Weil [Thu, 17 Sep 2009 17:41:07 +0000 (10:41 -0700)]
msgr: fix null deref setting message connection ref

15 years agoconfigure: don't link expat, fcgi to everything
Sage Weil [Wed, 16 Sep 2009 17:34:04 +0000 (10:34 -0700)]
configure: don't link expat, fcgi to everything

The default AC_CHECK_LIB action is to add -lfoo to the linker args,
which we don't want!

15 years agoMakefile: fix mode on installed headers
Sage Weil [Wed, 16 Sep 2009 04:27:41 +0000 (21:27 -0700)]
Makefile: fix mode on installed headers

15 years agomsgr: change dispatcher interfaces; some monclient cleanups
Sage Weil [Tue, 15 Sep 2009 22:02:17 +0000 (15:02 -0700)]
msgr: change dispatcher interfaces; some monclient cleanups

15 years agouclient: do not do locking for 'atomic' writes
Sage Weil [Tue, 15 Sep 2009 20:14:01 +0000 (13:14 -0700)]
uclient: do not do locking for 'atomic' writes

They aren't tested and aren't safe (client may fail; there's no cleanup).

15 years agomds: flush log in caller(s), not scatter_writebehind
Sage Weil [Mon, 14 Sep 2009 22:34:47 +0000 (15:34 -0700)]
mds: flush log in caller(s), not scatter_writebehind

Otherwise scatter_tick(), in particular, will call scatter_writebehind
a zillion times and generate lots of small log flushes.  Instead, flush
at the end of the whole tick.

15 years agocheck_mds_request_latency: fix parsing for current debug output
Sage Weil [Mon, 14 Sep 2009 22:07:30 +0000 (15:07 -0700)]
check_mds_request_latency: fix parsing for current debug output