Whamcloud - gitweb
alex [Sun, 10 Jul 2005 23:10:17 +0000 (23:10 +0000)]
b=3984
- b_ldlm_newlocking landed. we need it to pass some CMD2 performance tests
NOTE: the new locking rules introduced by the patch are still experimental!
alex [Sun, 10 Jul 2005 23:05:06 +0000 (23:05 +0000)]
- drop reference for open replay data in mdc_commit_open() instead of
mdc_commit_close() as in some cases we may have no close at all
alex [Sun, 10 Jul 2005 22:12:32 +0000 (22:12 +0000)]
- out of date warning has been removed
alex [Sun, 10 Jul 2005 22:07:43 +0000 (22:07 +0000)]
- make 512-bytes inode size by default: with new EAs (fid, parent) we just
can't fit 256-bytes inodes
alex [Sun, 10 Jul 2005 20:38:42 +0000 (20:38 +0000)]
- convert CWARN() to CDEBUG()
alex [Sun, 10 Jul 2005 19:53:09 +0000 (19:53 +0000)]
- we don't need deadlock-monitor in general series
alex [Sun, 10 Jul 2005 19:52:22 +0000 (19:52 +0000)]
b=6817
- b_size_on_mds landed (prototype for CMD2 w/o recovery support on OST)
alex [Sun, 10 Jul 2005 18:36:24 +0000 (18:36 +0000)]
- minor debug level change to have only DEBUG_REQ with D_RPCTRACE
alex [Sun, 10 Jul 2005 18:31:06 +0000 (18:31 +0000)]
- update enqueue time upon lock conversion otherwise we get too long
enqueue wait time warning
ericm [Sun, 10 Jul 2005 05:17:51 +0000 (05:17 +0000)]
mds_open() better to check permission even it's open request for device
files, otherwise client vfs will check that and failed siliently which
cause request leak. (client's interaction with vfs seems not perfect --
can't handle all error cases)
ericm [Sat, 9 Jul 2005 19:33:55 +0000 (19:33 +0000)]
try to be a little bit more nice to remote use: as long as it be mapped
to an local user, the supp groups of target usr will take part in fs
access checking.
ericm [Sat, 9 Jul 2005 16:04:16 +0000 (16:04 +0000)]
adjust gss timeout valud, don't use fixed value.
wangdi [Sat, 9 Jul 2005 03:19:27 +0000 (03:19 +0000)]
Branch HEAD
fix a typo in sanity-cmobd.sh
yury [Fri, 8 Jul 2005 15:23:42 +0000 (15:23 +0000)]
- putting file fid into o_inline for cases when crow object may be created. It will be used later for storing into OSS object EA what is needed for quota.
tappro [Fri, 8 Jul 2005 13:57:25 +0000 (13:57 +0000)]
b=6285
- remove few debug and LBUG()
yury [Fri, 8 Jul 2005 10:35:04 +0000 (10:35 +0000)]
- cleanups in cobd and cmobd
- moved cmobd's level in lconf to be under mountpoint. This prevents oops in local config when both cobd and cmobd are used and refer to the same master device.
yury [Fri, 8 Jul 2005 07:29:45 +0000 (07:29 +0000)]
- using push_ctxt() and pop_ctxt() when setting uid/gid in filter_create_object() at Alex's suggest.
jacob [Thu, 7 Jul 2005 21:43:41 +0000 (21:43 +0000)]
- add things to /etc/modules.conf | /etc/modprobe.conf
- tweaks to init scripts for SLES 9
buffalo-pull [Thu, 7 Jul 2005 18:04:09 +0000 (18:04 +0000)]
uid == gid is no longer a valid assumption.
ericm [Thu, 7 Jul 2005 17:41:19 +0000 (17:41 +0000)]
land b_hd_sec on HEAD. various security fixes.
alex [Thu, 7 Jul 2005 16:29:19 +0000 (16:29 +0000)]
b=7039
- additional debug info: dump backtrace
alex [Thu, 7 Jul 2005 14:27:05 +0000 (14:27 +0000)]
- minor typo
alex [Thu, 7 Jul 2005 14:23:22 +0000 (14:23 +0000)]
b=6285
- sometimes mds_getattr_lock() is asked to refresh attributes by fid.
it doesn't take LOOKUP lock, thus has no protected from concurrent
unlink. this means mds_getattr_lock() can return already unlinked
(and being removed) inode. to avoid this we check i_nlink right
after we got UPDATE lock and return -ENOENT if nlink is zero
alex [Thu, 7 Jul 2005 13:38:55 +0000 (13:38 +0000)]
b=6913
- retract extents lock from clients even if OSS hasn't created object yet
alex [Thu, 7 Jul 2005 13:03:01 +0000 (13:03 +0000)]
- minor comment about openlock and test 24n from sanity.sh
tappro [Thu, 7 Jul 2005 12:21:56 +0000 (12:21 +0000)]
b=6285
- more debug info was added
tappro [Thu, 7 Jul 2005 08:33:07 +0000 (08:33 +0000)]
b=6285
- add LBUG() to get kernel-logs in case of this issue
yury [Thu, 7 Jul 2005 07:57:07 +0000 (07:57 +0000)]
- better to make vfs_create() to set correct uid and gid for new inode, as dirtying inode in ext3 is too expensive. Thanks to Andreas.
alex [Wed, 6 Jul 2005 21:53:09 +0000 (21:53 +0000)]
b=7023
- export can be failed asynchronously with regard to processing request.
this case cause open rpc to be replied to zero transno. zero transno
means no reference for request and open request gets freed after
ll_file_open() though mdc_open_data still has a reference on the
request. and this causes oops when mdc_close() tries to access the
request. the workaround is to take additional reference on the request
for mdc_open_data
alex [Wed, 6 Jul 2005 21:06:19 +0000 (21:06 +0000)]
- disable test 24n: it fails because of disabled opencache. current
mds_open() doesn't take lookup lock and client can't maintain
dcache in coherence
alex [Wed, 6 Jul 2005 20:13:41 +0000 (20:13 +0000)]
- make recovery-small.sh working correct with local configuration (1 mds)
yury [Wed, 6 Jul 2005 16:49:20 +0000 (16:49 +0000)]
- passing full options page to do_kern_mount() in confobd and smfs too like it is done in MDS and OST
yury [Wed, 6 Jul 2005 12:41:36 +0000 (12:41 +0000)]
- removed not used vars in client_obd_setup()
- many fixes and improvements suggested by Andreas
- removed not used function inode_update_time()
niu [Wed, 6 Jul 2005 07:53:00 +0000 (07:53 +0000)]
minor fix in lmv_get_mea_and_update_object().
niu [Wed, 6 Jul 2005 07:48:59 +0000 (07:48 +0000)]
convert to u64 in case of overflow.
- b 6450
adilger [Tue, 5 Jul 2005 23:46:12 +0000 (23:46 +0000)]
Branch b1_4
Fix catamount build breakage, may be able to remove entirely at some point.
b=6923
brian [Tue, 5 Jul 2005 21:31:00 +0000 (21:31 +0000)]
Need to get values from netstat "unresolved" because the port numbers returned
are used in numerical comparisons.
brian [Tue, 5 Jul 2005 18:56:54 +0000 (18:56 +0000)]
Optionally use RUNAS_GID to runas command to set group to setregid() to.
This is needed to prevent errors such as:
mds blocked setgid attempt (1000/1000 -> 100)
alex [Tue, 5 Jul 2005 10:35:06 +0000 (10:35 +0000)]
- fix for test 9 from b1_4
alex [Tue, 5 Jul 2005 10:12:19 +0000 (10:12 +0000)]
- minor cleanup to fix warnings with disabled openlock
alex [Tue, 5 Jul 2005 09:59:56 +0000 (09:59 +0000)]
- handle copy_from_user() error
adilger [Tue, 5 Jul 2005 09:48:44 +0000 (09:48 +0000)]
Branch b1_4
Add some better error messages to acceptor to see if we can identify
ongoing startup problem. Also, remove pidfile at clean exit.
adilger [Tue, 5 Jul 2005 09:19:19 +0000 (09:19 +0000)]
Branch HEAD
Don't tag all of the backup files created by quilt for ldiskfs.
niu [Tue, 5 Jul 2005 09:18:16 +0000 (09:18 +0000)]
add ioctl for audit syslog interface
alex [Mon, 4 Jul 2005 16:23:18 +0000 (16:23 +0000)]
- give 1MB RPCs a run on 256MB mountain's client
alex [Mon, 4 Jul 2005 14:27:07 +0000 (14:27 +0000)]
- don't pass "prefer mergability over length" hint to mballoc
adilger [Mon, 4 Jul 2005 11:10:21 +0000 (11:10 +0000)]
Branch b1_4
Clean up CVS history files ala .#some_old_file.c.1.2.3.4.5
alex [Mon, 4 Jul 2005 10:08:24 +0000 (10:08 +0000)]
- disable "too long searching" in mballoc
alex [Mon, 4 Jul 2005 08:13:53 +0000 (08:13 +0000)]
- add mballoc to ldiskfs series
- make extents,mballoc a default options for OST
alex [Mon, 4 Jul 2005 06:28:56 +0000 (06:28 +0000)]
- CROW-related fixes from b_hd_mdref
alex [Mon, 4 Jul 2005 06:14:45 +0000 (06:14 +0000)]
- define NOHIGHMEM for recently added highmem-split patch
alex [Mon, 4 Jul 2005 05:59:51 +0000 (05:59 +0000)]
- make mdc_close() to use regular recovery mechanism
alex [Sun, 3 Jul 2005 22:49:44 +0000 (22:49 +0000)]
b=5888
- should exit immediately, if we aren't a holder of size semaphore
alex [Sun, 3 Jul 2005 20:23:57 +0000 (20:23 +0000)]
- highmem-split-2.6.10-fc3.patch in order to make 3GB memory on mountain's
nodes available
wangdi [Sun, 3 Jul 2005 14:51:58 +0000 (14:51 +0000)]
Branch: HEAD
minor fix for cmobd test script
tappro [Sun, 3 Jul 2005 06:49:34 +0000 (06:49 +0000)]
b=6285
- debug info
alex [Sat, 2 Jul 2005 06:15:47 +0000 (06:15 +0000)]
- statmany needs more arguments
alex [Sat, 2 Jul 2005 05:40:21 +0000 (05:40 +0000)]
- remove non-needed check
alex [Sat, 2 Jul 2005 05:38:47 +0000 (05:38 +0000)]
- minor change to prevent warnings
eeb [Fri, 1 Jul 2005 09:34:26 +0000 (09:34 +0000)]
* Changed default vib tx descriptor pool sizes
eeb [Wed, 29 Jun 2005 20:38:06 +0000 (20:38 +0000)]
* Removed misleading comment about the # of tx descs
* Don't force FMR unmap on error if it wasn't mapped in the first place.
alex [Wed, 29 Jun 2005 16:17:27 +0000 (16:17 +0000)]
b=6285
- protect i_nlink in rmdir/mkdir case because with pdirops it can be
modified simultaneously
alex [Wed, 29 Jun 2005 15:51:29 +0000 (15:51 +0000)]
b=6962
- lov_setattr() doesn't need OBD_MD_FLGENER
alex [Wed, 29 Jun 2005 15:15:09 +0000 (15:15 +0000)]
b=6816
- don't release intent, because ->lookup() can be called later
wangdi [Wed, 29 Jun 2005 13:51:44 +0000 (13:51 +0000)]
Branch: HEAD
some fix and cleanup for WB test script
wangdi [Wed, 29 Jun 2005 13:29:26 +0000 (13:29 +0000)]
Branch HEAD
remove obsolete cmobd test script
wangdi [Wed, 29 Jun 2005 13:25:56 +0000 (13:25 +0000)]
Branch: HEAD
add cfg_instance string in setup lcfg for WB zconf.
alex [Tue, 28 Jun 2005 21:10:11 +0000 (21:10 +0000)]
b=6975
- concurrent creations are valid since CROW merging
adilger [Tue, 28 Jun 2005 21:09:58 +0000 (21:09 +0000)]
Branch b1_4
Move some more verbose router messages to D_INFO.
alex [Tue, 28 Jun 2005 16:10:08 +0000 (16:10 +0000)]
- minor optimization: we already have fid in reply, no need to fetch it
eeb [Tue, 28 Jun 2005 09:46:16 +0000 (09:46 +0000)]
* vibnal: ensure FMR flushed on error
alex [Mon, 27 Jun 2005 15:52:45 +0000 (15:52 +0000)]
b=6885
- disable opencache feature in order to pass #3
alex [Mon, 27 Jun 2005 15:48:56 +0000 (15:48 +0000)]
b=6885
- make lustre_fid smaller to fit to 256 bytes inode
NOTE: improves test #3 significantly
alex [Mon, 27 Jun 2005 15:45:59 +0000 (15:45 +0000)]
b=6885
- reservation feature in ext3 hurts test #3: disable it
alex [Mon, 27 Jun 2005 15:43:28 +0000 (15:43 +0000)]
b=6885
- we don't need BKL in {set|get}_attr routines in 2.6
tappro [Sat, 25 Jun 2005 22:10:50 +0000 (22:10 +0000)]
b=5881
- protect smfs plugins operations
- alloc plugin before registration
tappro [Sat, 25 Jun 2005 22:05:52 +0000 (22:05 +0000)]
b=6285
r=alex
smfs should care about nlinks handling instead of just copying it from/to bottom fs.
tappro [Sat, 25 Jun 2005 08:15:32 +0000 (08:15 +0000)]
- pevious commit was incomplete.
adilger [Sat, 25 Jun 2005 06:00:09 +0000 (06:00 +0000)]
Branch b1_4
Shouldn't be able to run land2.sh with uncommitted changes.
tappro [Thu, 23 Jun 2005 20:32:16 +0000 (20:32 +0000)]
b=6285
- make smfs_rmdir behaves like vfs_rmdir
ericm [Thu, 23 Jun 2005 17:45:32 +0000 (17:45 +0000)]
land b_hd_sec on HEAD: lctl flush stuff, plus various client gss
cache fixes.
yury [Thu, 23 Jun 2005 14:52:51 +0000 (14:52 +0000)]
- numerous fixes from b_cray, b_1_4 and Andreas' comments.
buffalo-pull [Thu, 23 Jun 2005 13:25:12 +0000 (13:25 +0000)]
Bring these two configs into parity with each others' recent changes.
alex [Thu, 23 Jun 2005 12:26:15 +0000 (12:26 +0000)]
b=6942
- don't take inode->i_sem. otherwise we deadlock because right ordering
is i_sem then journal_strart. as we're creating the file, nobody can
access it. so it's safe to set fid w/o i_sem
jacob [Wed, 22 Jun 2005 17:52:22 +0000 (17:52 +0000)]
add directory commands to gdb script
yury [Wed, 22 Jun 2005 11:33:48 +0000 (11:33 +0000)]
- CROW (CReate On Write) (precreation is removed)
- setting owner/group on OSS (needed for quotas)
- some fixes:
- deadlock on soc client lock.
- idr_remove() issue.
- forwarding to wrong MDS in lmv_link()
- f_dput(dentry) before using it in filter_lvbo_update()
- filter_sync() assert(group > 0) due to missed GROUP flag oa->o_valid in ll_fsync().
- cleanups, comments and more debugging messages.
- jump to wrong label in mds_create_objects() on error path.
- few missed/wrong ENTRY/RETURN things
nic [Tue, 21 Jun 2005 17:41:09 +0000 (17:41 +0000)]
move awk script into bash, as strtonum() is broken on some systems
add a bit more help when we hit errors too
alex [Tue, 21 Jun 2005 17:39:23 +0000 (17:39 +0000)]
- enable spinlock debugging for fc3 kernel
NOTE: it seems do_gettimeofday() may sleep in UML, so don't enable
CONFIG_DEBUG_SPINLOCK_SLEEP
ericm [Tue, 21 Jun 2005 16:31:14 +0000 (16:31 +0000)]
sometime we don't need gracefully flush client credential.
nikita [Tue, 21 Jun 2005 09:57:36 +0000 (09:57 +0000)]
added b_ioprovement branch
alex [Tue, 21 Jun 2005 06:16:47 +0000 (06:16 +0000)]
- declare export non-connected in filter_init_server_data()
alex [Tue, 21 Jun 2005 05:33:31 +0000 (05:33 +0000)]
b=6087
- don't confuse OST by non-zero connect flags from clients. OST expects
group number while only MDS can have a group
ericm [Mon, 20 Jun 2005 22:18:48 +0000 (22:18 +0000)]
land b_hd_sec onto HEAD: various security related fixes.
alex [Mon, 20 Jun 2005 13:12:16 +0000 (13:12 +0000)]
- let show_task() to show backtrack for currently running process (for LBUG)
adilger [Sun, 19 Jun 2005 09:18:22 +0000 (09:18 +0000)]
Land b_release_1_4_3 onto HEAD (20050619_0305)
b=6411 : enable rate-limiting of console error messages, and some console
errors now go only to the kernel log. Use CERROR/CWARN where
appropriate to allow rate-limiting of these messages.
b=1693 : add /proc/sys/portals/catastrophe entry which will report if
that node has previously LBUGged.
alex [Sat, 18 Jun 2005 13:39:35 +0000 (13:39 +0000)]
b=6871
- drop fail_loc to 0 in test 20
brian [Sat, 18 Jun 2005 03:15:29 +0000 (03:15 +0000)]
Allow sanity.sh to work with zconf.
Taken from b1_4 branch. Thanx nic and adilger.
yury [Fri, 17 Jun 2005 16:44:54 +0000 (16:44 +0000)]
- fix in lmv.sh to use it along with sanityN.sh
eeb [Fri, 17 Jun 2005 10:24:41 +0000 (10:24 +0000)]
* Backed out vib magic change; it's not required since the acceptor
isn't involved.
eeb [Fri, 17 Jun 2005 10:04:35 +0000 (10:04 +0000)]
* Make vibnal's magic unique before binary compatibility becomes an issue