Whamcloud - gitweb
grev [Tue, 6 Oct 2009 11:13:33 +0000 (11:13 +0000)]
b=20179
i=Robert.Read
process TESTSUITE log to fail if error
grev [Tue, 6 Oct 2009 11:06:18 +0000 (11:06 +0000)]
b=19737
i=Robert.Read
t-f NFSCLIENT changes
grev [Tue, 6 Oct 2009 11:03:12 +0000 (11:03 +0000)]
b=20698
i=Robert.Read
new base for new test_[a-z] tests
johann [Tue, 6 Oct 2009 08:49:19 +0000 (08:49 +0000)]
Branch b_release_1_8_1
b=18630
i=johann
don't LBUG on release quota error. Just a workaround until the problem is understood.
johann [Tue, 6 Oct 2009 08:35:33 +0000 (08:35 +0000)]
Branch b_release_1_8_1
b=18793
i=johann (patch from Tianzy)
i=panda
we should keep the same uid/gid for lquota_chkquota() and lquota_pending_commit()
johann [Tue, 6 Oct 2009 08:20:43 +0000 (08:20 +0000)]
Branch b_release_1_8_1
b=20903
i=rread
i=dmitry
Make obd_zombie_barrier() no-op if called from the zombie thread context.
johann [Mon, 5 Oct 2009 20:46:37 +0000 (20:46 +0000)]
Branch b_release_1_8_1
fix typo.
johann [Mon, 5 Oct 2009 20:37:21 +0000 (20:37 +0000)]
Branch b_release_1_8_1
b=20816
Revert patch from bug 13520.
johann [Mon, 5 Oct 2009 20:33:44 +0000 (20:33 +0000)]
Branch b_release_1_8_1
ad changelog entry for 18571
johann [Mon, 5 Oct 2009 20:22:59 +0000 (20:22 +0000)]
Branch b_release_1_8_1
remove 1.8.1.1 changelog section.
panda [Mon, 5 Oct 2009 15:43:39 +0000 (15:43 +0000)]
b=20575
o=Tom Wang
i=Johann Lombardi
i=Andrew Perepechko
avoid overflow in ll_readahead
panda [Fri, 2 Oct 2009 16:28:20 +0000 (16:28 +0000)]
b=20953
i=Elena Gryaznova
additional debug for test 30
johann [Thu, 1 Oct 2009 21:48:54 +0000 (21:48 +0000)]
Branch b_release_1_8_1
b=20122
i=adilger (patch from shadow)
i=panda
prevent req double free in ll_follow_link
girish [Wed, 30 Sep 2009 08:11:49 +0000 (08:11 +0000)]
b=18571
i=adilger
i=girish
o=vitaly
remove wrong get_bh() from journal checksum patch
yangsheng [Tue, 22 Sep 2009 12:10:52 +0000 (12:10 +0000)]
Branch b_release_1_8_1
b=20819
i=johann
Get rid of the warning messages for snmp build.
zhanghc [Mon, 21 Sep 2009 06:44:52 +0000 (06:44 +0000)]
b=20354
the client_uuid (lcd_uuid) stored in last_rcvd on server(MDT or OST)
could be corrupted, so it is needed to check during loading from disk
i=johann@sun.com
i=adilger@sun.com
i=tappro@sun.com
stinson1947 [Fri, 18 Sep 2009 22:11:42 +0000 (22:11 +0000)]
Updated for 1.8.1.1 RC1
johann [Fri, 18 Sep 2009 21:20:48 +0000 (21:20 +0000)]
Branch b_release_1_8_1
b=20205
disable write barrier for ext4/SLES11 since this is suspected
to hurt write performance as reported in the past on ext3/SLES10
yangsheng [Fri, 18 Sep 2009 17:03:52 +0000 (17:03 +0000)]
Branch b_release_1_8_1
b=20539
Add comment for OEL5.
yangsheng [Fri, 18 Sep 2009 16:40:34 +0000 (16:40 +0000)]
Branch b_release_1_8_1
b=19848
i=johann, bobijam
Update kernel to SLES11 2.6.27.29-0.1.1.
yangsheng [Fri, 18 Sep 2009 13:20:23 +0000 (13:20 +0000)]
Branch b_release_1_8_1
b=20539
Change lbuild-oel5 same as rhel5.
yangsheng [Fri, 18 Sep 2009 12:27:56 +0000 (12:27 +0000)]
Branch b_release_1_8_1
b=20539
i=johann, bobijam
Add support for OEL5.
johann [Thu, 17 Sep 2009 21:01:06 +0000 (21:01 +0000)]
Branch b_release_1_8_1
b=20518
i=rread
i=tappro
Make sure all exports have been properly destroyed by the zombie
thread processed before stopping the target.
brian [Thu, 17 Sep 2009 18:28:21 +0000 (18:28 +0000)]
b=20640
i=wangyb
i=yangsheng
Only apply the asm-x86_64 workaround (filed as RH bug 491775) to RHEL
releases that actually need it.
shadow [Thu, 17 Sep 2009 09:27:06 +0000 (09:27 +0000)]
Don't allow make backward step on assiging osc next id.
Branch b_release_1_8_1
b=19557
i=tappro
i=jay
johann [Thu, 17 Sep 2009 09:15:30 +0000 (09:15 +0000)]
Branch b_release_1_8_1
b=18618
i=johann (patch from Nathan)
i=adilger
if a client doesn't respond to a blocking callback within the
adaptive ldlm enqueue timeout, don't adjust the adaptive estimate
when the lock is next granted.
shadow [Thu, 17 Sep 2009 06:32:14 +0000 (06:32 +0000)]
Don't panic on shrink getattr reply about special file after error.
Branch b_release_1_8_1
b=20020
i=johann
ericm [Wed, 16 Sep 2009 19:08:04 +0000 (19:08 +0000)]
branch: b_release_1_8_1
mgs initialize export with exp_connecting set.
b=18849
r=rread
r=fanyong
johann [Wed, 16 Sep 2009 14:41:38 +0000 (14:41 +0000)]
Branch b_release_1_8_1
b=19551
i=tianzy
i=adilger
don't print message on the console when ->adjust_qunit fails.
bobijam [Wed, 16 Sep 2009 11:21:23 +0000 (11:21 +0000)]
ranch b_release_1_8_1
b=20560
o=johann
i=alex.zhuravlev
i=zhenyu.xu (bobijam)
Discard pages if somehting failed after filter_preprw_write().
bobijam [Wed, 16 Sep 2009 02:07:12 +0000 (02:07 +0000)]
Branch b_release_1_8_1
b=20560
o=johann
i=alex.ahuravlev
i=zhenyu.xu (bobijam)
don't poison pages for evicted clients
manoj [Tue, 15 Sep 2009 19:40:51 +0000 (19:40 +0000)]
b=19781
r=alex.zhuravlev
r=grev
r=robert.read
Tests for the OSS Read Cache feature.
yangsheng [Tue, 15 Sep 2009 15:33:31 +0000 (15:33 +0000)]
Branch b_release_1_8_1
b=20400
i=shadow, bobijam
Update kernel to RHEL5 2.6.18-128.7.1.el5.
grev [Tue, 15 Sep 2009 15:15:58 +0000 (15:15 +0000)]
b=20237
i=Manoj.Joseph
i=Robert.Read
gather and archive the logs
panda [Tue, 15 Sep 2009 14:15:39 +0000 (14:15 +0000)]
b=16267
i=Johann Lombardi
i=Vitaly Fertman
llapi man pages
yangsheng [Tue, 15 Sep 2009 12:01:28 +0000 (12:01 +0000)]
Branch b_release_1_8_1
b=20758
Update kernel to SLES10 SP2 2.6.16.60-0.42.4.
johann [Tue, 15 Sep 2009 09:44:31 +0000 (09:44 +0000)]
Branch b_release_1_8_1
b=20550
i=adilger (patch from Brian)
i=wangyb
i=yangsheng
This patch fixes the problem of make dist not creating a tree_status when
the tree it's being run in is not from an sccs checkout.
johann [Tue, 15 Sep 2009 09:36:53 +0000 (09:36 +0000)]
Branch b_release_1_8_1
b=20318
add changelog entry.
grev [Tue, 15 Sep 2009 09:27:42 +0000 (09:27 +0000)]
b=17951
i=Robert.Read
i=Alexey.Lyashkov
test_46a: use all ost-s available
grev [Tue, 15 Sep 2009 09:23:14 +0000 (09:23 +0000)]
b=20331
i=Mikhail.Pershin
i=Robert.Read
i=Alexey.Lyashkov
recovery-mds-scale: wait clients import FULL state
bobijam [Mon, 14 Sep 2009 02:11:07 +0000 (02:11 +0000)]
Branch b_release_1_8_1
b=20560
i=adilger
i=zhenyu.xu (bobijam)
Fix for crashes when disabling writethrough_cache_enable.
grev [Fri, 11 Sep 2009 16:08:43 +0000 (16:08 +0000)]
b=18623
o=Ed.Giesen
i=grev
move unset ENABLE_QUOTA to after config file read
bobijam [Fri, 11 Sep 2009 01:31:07 +0000 (01:31 +0000)]
Branch b_release_1_8_1
b=20560
i=johann
i=alexander.zarochentsev
Description: File checksum failures with OST read cache on
Details : Clear page cache's uptodate bit when its contents are cleared.
bobijam [Wed, 9 Sep 2009 09:04:29 +0000 (09:04 +0000)]
Branch b_release_1_8_1
b=20318
o=johann
i=adilger
i=andrew.perepechko (panda)
keep same lsm in place.
bobijam [Wed, 9 Sep 2009 08:59:19 +0000 (08:59 +0000)]
Branch b_release_1_8_1
b=20318
o=johann
i=adilger
i=andrew.perepechko (panda)
keep same lsm in place.
dzogin [Sun, 6 Sep 2009 16:24:26 +0000 (16:24 +0000)]
Branch b_release_1_8_1
b=20533
i=johann
i=adilger
----------------------------------------------------------------------
Modified Files:
Tag: b_release_1_8_1
lustre/ChangeLog
lustre/kernel_patches/patches/raid5-large-io-rhel5.patch
----------------------------------------------------------------------
Description: Changes in raid5-large-io-rhel5.patch to calculate sectors properly
dzogin [Sun, 6 Sep 2009 16:02:30 +0000 (16:02 +0000)]
Branch b_release_1_8_1
b=20533
i=johann
i=adilger
----------------------------------------------------------------------
Modified Files:
Tag: b_release_1_8_1
lustre/ChangeLog lustre/kernel_patches/series/2.6-rhel5.series
lustre/kernel_patches/series/2.6-sles11.series
Added Files:
Tag: b_release_1_8_1
lustre/kernel_patches/patches/blkdev_tunables-2.6-rhel5.patch
lustre/kernel_patches/patches/blkdev_tunables-2.6-sles11.patch
===================================================================
manoj [Fri, 4 Sep 2009 18:40:00 +0000 (18:40 +0000)]
b=19780
r=nathan.rutman
r=grev
r=edg
Tests for the OST Pools feature
dzogin [Fri, 4 Sep 2009 17:15:46 +0000 (17:15 +0000)]
Branch b_release_1_8_1
b=20482
i=robert.read
i=adilger
----------------------------------------------------------------------
Modified Files:
Tag: b_release_1_8_1
lustre/ChangeLog lustre/lov/lov_request.c
lustre/osc/osc_request.c
----------------------------------------------------------------------
Index: lustre/ChangeLog
===================================================================
Severity : normal
Bugzilla : 20482
Description: Error handling in osc_statfs_interpret() has been improved.
Details : Check in osc_statfs_interpret() for EBADR.
Severity : normal
Bugzilla : 20482
Description: Do not send statfs() requests to OSTs disabled by administrator.
Details : Check in lov_prep_statfs_set() for non-NULL ltd_exp.
===================================================================
ericm [Fri, 4 Sep 2009 14:34:51 +0000 (14:34 +0000)]
branch: b_release_1_8_1
OBD_FAIL_PTLRPC_PAUSE_REQ to ignore ping.
b=13520
r=nathan
r=wangdi
tianzy [Thu, 3 Sep 2009 02:25:46 +0000 (02:25 +0000)]
Branch b_release_1_8_1(1.8.1.1)
fix a rare error in test_25 of sanity-quota.sh
b=20433
i=yong.fan
i=panda
ericm [Wed, 2 Sep 2009 17:29:04 +0000 (17:29 +0000)]
branch: b_release_1_8_1
fix safe type conversion.
b=20392
r=nathan
r=wangdi
grev [Wed, 2 Sep 2009 16:31:57 +0000 (16:31 +0000)]
b=20186
i=Adilger
conf-sanity test_32* 1.6->1.8 upgrade tests
dzogin [Wed, 2 Sep 2009 14:28:24 +0000 (14:28 +0000)]
Branch b_release_1_8_1
b=20146
i=andrew.perepechko
i=adilger
Modified Files:
Tag: b_release_1_8_1
lustre/ChangeLog lustre/mds/mds_reint.c
Description: Do not update ctime for the deleted inode.
Details : Check in mds_reint_unlink() before calling fsfilt_setattr().
brian [Wed, 2 Sep 2009 13:56:20 +0000 (13:56 +0000)]
b=20595
i=yangsheng
Build support for OFED 1.4.2.
vs [Wed, 2 Sep 2009 13:23:56 +0000 (13:23 +0000)]
b=17807
i=nikita
i=huanghua
ll_revalidate_nd: do not return 0 in case of error
w.li [Wed, 2 Sep 2009 13:15:52 +0000 (13:15 +0000)]
Branch b_release_1_8_1
b=20347
i=tappro
i=yong.fan
In llite's SETFLAGS ioctl, the local i_flags should be updated even when
"lsm == NULL".
zhanghc [Mon, 24 Aug 2009 07:08:48 +0000 (07:08 +0000)]
b=19870
fix a bug of processing export->exp_rpc_count during recovery
i=tappro@sun.com
i=alexey.lyashkov@sun.com
zhanghc [Mon, 24 Aug 2009 05:38:03 +0000 (05:38 +0000)]
b=19791
obdclass/llog_lvfs.c: llog_lvfs_create
l_dentry_open will call dput if there is an erro, then
there should be no similar call(l_dput) if error returned
i=adilger@sun.com
i=alexey.lyashkov@sun.com
johann [Fri, 21 Aug 2009 15:47:34 +0000 (15:47 +0000)]
Branch b_release_1_8_1
b=20008
Move changelog entry to correct place.
johann [Fri, 21 Aug 2009 15:40:10 +0000 (15:40 +0000)]
Branch b_release_1_8_1
b=20321
i=bzzz
i=dmitry
Take i_alloc_sem in filter_destroy() when truncating the object & fix inode mutex leak
dzogin [Thu, 20 Aug 2009 22:41:06 +0000 (22:41 +0000)]
b=20146
i=alex.zhuravlev
i=johann
Description: Increase of the size of the LDLM resource hash.
Details : Bump up RES_HASH_BITS=12.
Modified Files:
Tag: b_release_1_8_1
lustre/ChangeLog lustre/include/lustre_dlm.h
anserper [Thu, 20 Aug 2009 18:56:13 +0000 (18:56 +0000)]
b=16890
i=Johann Lombardi
i=ZhiYong Tian
do not reset grace times if setquota is given only hardlimit params
shadow [Thu, 20 Aug 2009 16:00:52 +0000 (16:00 +0000)]
correctly send lsm on open replay
Branch b_release_1_8_1
b=19934
i=tappro
i=panda
brian [Wed, 19 Aug 2009 17:34:44 +0000 (17:34 +0000)]
b=20478
i=wangyb
i=adilger
Fix bug where version_tag.pl failed when there was no kernel source
available.
stinson1947 [Wed, 19 Aug 2009 17:02:36 +0000 (17:02 +0000)]
Corrected placement of a bug fix from 1.8.1 to 1.8.1.1.
anserper [Mon, 17 Aug 2009 19:21:50 +0000 (19:21 +0000)]
ChangeLog
anserper [Mon, 17 Aug 2009 18:16:51 +0000 (18:16 +0000)]
b=20008
i=Oleg Drokin
i=Alex Tomas
pin the last page before starting transaction in setattr to avoid GFP_FS allocation under transaction
isaac [Wed, 12 Aug 2009 18:15:15 +0000 (18:15 +0000)]
b=19782,i=isaac:
- fixed a typo in LST sample test script.
stinson1947 [Thu, 6 Aug 2009 18:24:06 +0000 (18:24 +0000)]
Updated ChangeLog for next possible release, 1.8.1.1.
bobijam [Mon, 3 Aug 2009 01:57:58 +0000 (01:57 +0000)]
Branch b_release_1_8_1
b=20219
o=johann
i=andrew.perepechko (panda)
i=alexander.zarochentsev (zam)
Release lock ref for lockless i/o if allocation failure.
dzogin [Sun, 2 Aug 2009 15:54:54 +0000 (15:54 +0000)]
b=20321
i=dmitry.zogin
i=adilger
Severity : normal
Bugzilla : 20321
Description: Deadlock between filter_destroy() and filter_commitrw_write().
Details : filter_destroy() does not hold the DLM lock over the whole
operation. If the DLM lock is dropped, filter_commitrw() can go
through, causing the deadlock between page lock and i_mutex.
anserper [Fri, 31 Jul 2009 16:29:33 +0000 (16:29 +0000)]
b=20314
i=Johann Lombardi
i=Oleg Drokin
properly handle rc<0 from osc_brw_prep_request
anserper [Fri, 31 Jul 2009 15:53:42 +0000 (15:53 +0000)]
b=20317
i=Johann Lombardi
i=Alexander Zarochentsev
do not update kms without extent locks on DIO path
grev [Thu, 30 Jul 2009 19:20:45 +0000 (19:20 +0000)]
b=20259
i=Brian
get rid of useless rm PI*
johann [Fri, 24 Jul 2009 20:28:40 +0000 (20:28 +0000)]
Branch b_release_1_8_1
b=20020
i=adilger
i=johann (patch from shadow)
call lustre_shrink_reply() only on real reply.
johann [Fri, 24 Jul 2009 20:06:02 +0000 (20:06 +0000)]
Branch b_release_1_8_1
b=20261
i=johann
i=shadow
Use current when task == NULL.
grev [Fri, 24 Jul 2009 17:58:18 +0000 (17:58 +0000)]
b=20124
i=Jian.Yu
wait_for host for HARD failure mode
grev [Fri, 24 Jul 2009 13:54:12 +0000 (13:54 +0000)]
b=20164
o=Zhenyu.Xu
i=Adilger
i=grev
when echo client attached to OST device, o_brw_async() ops is not supported, we turn to o_brw()
instead.
maxim [Fri, 24 Jul 2009 10:05:28 +0000 (10:05 +0000)]
b=20254
i=johann
minor bugfix of rewritten router_proc.c (DECLARE_LL_PROC_PPOS_DECL should precede assignments with *ppos)
yangsheng [Fri, 24 Jul 2009 07:58:53 +0000 (07:58 +0000)]
Branch b_release_1_8_1
b=19847
Update kernel to SLES10 SP2 2.6.16.60-0.39.3.
maxim [Thu, 23 Jul 2009 13:48:46 +0000 (13:48 +0000)]
b=18102
i=shadow
i=isaac
proc-entries registered by lnet/lnet/router_proc.c were
re-implemented using modern sysctl approach, similarly
to libcfs/linux/linux-proc.c
johann [Thu, 23 Jul 2009 13:01:05 +0000 (13:01 +0000)]
Branch b_release_1_8_1
b=20219
Revert recent dio improvements because of bug 20219.
The problem will be worked out for 1.8.2.
grev [Thu, 23 Jul 2009 11:08:37 +0000 (11:08 +0000)]
b=19415
i=Robert.Read
restart_client_loads fix related to new client load var name
liangzhen [Thu, 23 Jul 2009 04:44:28 +0000 (04:44 +0000)]
Branch b_release_1_8_1
ordering write_lock to avoid deadlock
b=19518
i=adilger
i=zhenyu.xu
tianzy [Wed, 22 Jul 2009 13:40:38 +0000 (13:40 +0000)]
Branch b_release_1_8_1
fix "Cannot mount mdt or osts if --param xxx.quota_type in mkfs command"
b=20236
i=johann
brian [Tue, 21 Jul 2009 20:09:55 +0000 (20:09 +0000)]
b=19248
i=rread
We cannot use make style variables in the CPPFLAGS variable because that
variable is used later on in the configure script, which is bash, not make.
grev [Tue, 21 Jul 2009 13:06:08 +0000 (13:06 +0000)]
b=19248
o=Robert.Read
i=Brian
write_append_truncate build fix
stinson1947 [Fri, 17 Jul 2009 23:36:52 +0000 (23:36 +0000)]
Updated the release date.
johann [Fri, 17 Jul 2009 21:29:58 +0000 (21:29 +0000)]
Branch b_release_1_8_1
b=20139
i=johann (patch from fanyong)
i=bobijam
grab dentry reference in parent process.
brian [Fri, 17 Jul 2009 18:06:27 +0000 (18:06 +0000)]
b=19975
i=yangsheng
i=wangyb
Because we now build RHEL5 and SLES10&11 kernels from their own spec files,
we don't produce a kernel-source for RHEL5. RH's story to people who want
source is to unpack and patch the kernel SRPM. So we should provide the
SRPM for people who want kernel source.
SLES just comes along for the ride (actually, as do RHEL4 and SLES9 but we
don't support them for server on 1.8, so no need to distribute any kernel
[S]RPMs for them whatsoever.
anserper [Fri, 17 Jul 2009 16:04:06 +0000 (16:04 +0000)]
b=18801
i=Johann Lombardi
i=Girish Shilamkar
mtime fix for dio
girish [Thu, 16 Jul 2009 16:14:43 +0000 (16:14 +0000)]
b=20150
i=adilger
i=johann
fix locking typo - from upstream
not present in ext3-ldiskfs
grev [Thu, 16 Jul 2009 15:41:22 +0000 (15:41 +0000)]
b=19415
i=Robert.Read
fixes related to new load var
tianzy [Thu, 16 Jul 2009 09:55:54 +0000 (09:55 +0000)]
Branch b_release_1_8_1
fix a typo for LQUOTA_QAQ_CEATE_LQS
b=20031
yujian [Thu, 16 Jul 2009 01:59:07 +0000 (01:59 +0000)]
Branch b_release_1_8_1
b=20047
i=robert.read
i=yong.fan
reset debug level if lustre was already mounted
w.li [Thu, 16 Jul 2009 01:12:20 +0000 (01:12 +0000)]
Branch b_release_1_8_1
b=19999
i=grev
i=johann
Remove PID reading from multiop_bg_pause. Turn to $! for multiop PID in
runmultiop_bg_pause. The result looks more consistent with HEAD.
shadow [Wed, 15 Jul 2009 18:47:01 +0000 (18:47 +0000)]
correctly shrink reply for avoid send too big message to client.
Branch b_release_1_8_1
b=20020
i=adilger
i=tappro
fanyong [Wed, 15 Jul 2009 15:06:33 +0000 (15:06 +0000)]
Branch b_release_1_8_1
b=20002
i=tianzy
i=robert.read
Check "auto_quota_on()" return value before "build_lqs()" called.
isaac [Wed, 15 Jul 2009 14:54:22 +0000 (14:54 +0000)]
i=liangzhen,b=18075:
- an assortment of LNet selftest fixes and enhancements.
git://git.whamcloud.com / fs / lustre-release.git / log