Whamcloud - gitweb
bobijam [Fri, 6 Mar 2009 05:47:16 +0000 (05:47 +0000)]
Branch b1_6
b=18154
i=alexey.lyashkov (shadow)
i=johann
always calls ptlrpc_import_recovery_state_machine() in
invalidate_import_thread() because otherwise threads sleeping on
the import's imp_recovery_waitq could be never woken up.
bobijam [Fri, 6 Mar 2009 03:52:54 +0000 (03:52 +0000)]
Branch b1_6
b=17536
o=johann
i=zhenyu.xu (bobijam)
i=adilger
MDS create should not wait for statfs RPC while holding DLM lock.
bobijam [Fri, 6 Mar 2009 02:32:40 +0000 (02:32 +0000)]
Branch b1_6
b=17682
i=johann
i=vitaly.fertman (vitaly)
Improve debug info for OSS.
green [Thu, 5 Mar 2009 03:05:37 +0000 (03:05 +0000)]
b=17614
r=adilger,shadow
Do not put cancelled locks into replay list, hold references on locks in replay list
zhanghc [Tue, 3 Mar 2009 17:47:38 +0000 (17:47 +0000)]
b=17491
Speedup the insertion of ptlrpc_request in
function "ptlrpc_at_add_timed" to fix up
"Slow req_in handling" problem in
"ptlrpc_server_handle_req_in"
i=nathan.rutman
i=johann
shadow [Tue, 3 Mar 2009 17:11:14 +0000 (17:11 +0000)]
forget about one configure check
Branch b1_6
b=14250
zhanghc [Tue, 3 Mar 2009 16:10:57 +0000 (16:10 +0000)]
b=17757
clear DISP_OPEN_LOCK flag in "reconstruct_open"
if the corresponding lock was canceld.
i=Green
i=Johann
johann [Tue, 3 Mar 2009 13:47:52 +0000 (13:47 +0000)]
Branch b1_6
b=18374
i=adilger
i=shadow
use memmove instead of memcpy when shrinking the reply since the two memory
areas can overlap.
grev [Tue, 3 Mar 2009 13:04:45 +0000 (13:04 +0000)]
b=17524
reset_enospc fn: fix defect of previous commit
yangsheng [Mon, 2 Mar 2009 06:22:21 +0000 (06:22 +0000)]
Branch b1_6
b=17954
i=bobijam, deen
Coverity fixes.
Author: Jim Garlick(LLNL)
grev [Fri, 27 Feb 2009 14:04:37 +0000 (14:04 +0000)]
b=17524
i=Adilger
exhaust_precreations, reset_enospc, test_27n, test_27s fixes
grev [Fri, 27 Feb 2009 00:17:53 +0000 (00:17 +0000)]
fix typo in b=17839 commit (made while ported from HEAD)
yangsheng [Thu, 26 Feb 2009 18:21:53 +0000 (18:21 +0000)]
Branch b1_6
b=18289
i=shadow, johann
Update to RHEL5U3 kernel-2.6.18-128.1.1.el5.
girish [Thu, 26 Feb 2009 06:20:24 +0000 (06:20 +0000)]
b=17846
i=girish
i=h.huang
o=adilger
ignore stripe index 0 when setting striping via setxattr
anserper [Wed, 25 Feb 2009 22:54:51 +0000 (22:54 +0000)]
b=14929
i=Alexey Lyashkov
i=Johann Lombardi
remove duplicate code in filter_get_info and add pop_ctxt into error path
anserper [Wed, 25 Feb 2009 22:49:33 +0000 (22:49 +0000)]
b=14929
i=alexey.lyashkov
i=alexander.zarochentsev
do late push_ctxt so that pop_ctxt was never omitted
grev [Wed, 25 Feb 2009 16:57:23 +0000 (16:57 +0000)]
b=18468
o=Adilger
i=Oleg.Drokin
i=grev
make racer harder
grev [Wed, 25 Feb 2009 15:20:43 +0000 (15:20 +0000)]
b=18468
i=Adilger
runracer multidir changes
grev [Wed, 25 Feb 2009 11:20:04 +0000 (11:20 +0000)]
b=18514
i=Adilger
test_170: check that log files were created
fanyong [Wed, 25 Feb 2009 01:49:17 +0000 (01:49 +0000)]
Branch b1_6
b=18294
i=tianzy
i=andrew.perepechko
1) cleanup garbage files by force after test_8 finished
2) replace unnecessary error message with log
wangdi [Tue, 24 Feb 2009 23:58:45 +0000 (23:58 +0000)]
Branch: b1_6
b=12069
shrink client grant cache when ost does not have enough space left.
i=adilger,Johann,Oleg
grev [Tue, 24 Feb 2009 22:54:33 +0000 (22:54 +0000)]
b=17839
i=Brian
cmd3-17 port to acc-sm
vs [Tue, 24 Feb 2009 12:51:20 +0000 (12:51 +0000)]
b=18428
i=shadow,tappro,rread
do not send requests to inactive targets
tianzy [Tue, 24 Feb 2009 07:43:28 +0000 (07:43 +0000)]
Branch b1_6
add more debug info for AT and fix a minor error in replay-single.sh
b=18426
i=nathan
tianzy [Tue, 24 Feb 2009 07:07:22 +0000 (07:07 +0000)]
Branch b1_6
add a test for 18491
b=18491
i=johann
i=panda
anserper [Tue, 24 Feb 2009 00:15:30 +0000 (00:15 +0000)]
b=16666
i=Alexey Lyashkov
i=Johann Lombardi
fix targets online check
grev [Fri, 20 Feb 2009 22:27:27 +0000 (22:27 +0000)]
b=18443
i=Adilger
remove obsolete tests scripts
yangsheng [Fri, 20 Feb 2009 10:15:02 +0000 (10:15 +0000)]
Branch b1_6
b=18289
i=shadow, johann, yibin.yang
Kernel update to rhel5.3(2.6.18.92-128.1.1). Enable the in-kernel OFED
since haven't a separate OFED release support this kernel.
grev [Thu, 19 Feb 2009 22:33:16 +0000 (22:33 +0000)]
b=18521
i=Alex.Zhuravlev
sanity test_80, t-f localize var fix
anserper [Thu, 19 Feb 2009 18:56:42 +0000 (18:56 +0000)]
b=18233
i=Oleg Drokin (green)
i=Johann Lombardi (johann)
implemet rw lock so that vmtruncate could not race with reads
tianzy [Wed, 18 Feb 2009 07:52:29 +0000 (07:52 +0000)]
Branch b1_6
Releasing quota between lquota_chkquota() and filter_direct_io()
could make false -EDQUOT
b=18491
i=johann
i=andrew.perepechko
girish [Mon, 16 Feb 2009 07:47:53 +0000 (07:47 +0000)]
b=18470
i=johann
i=adilger
Correct error handling
jxiong [Sun, 15 Feb 2009 02:33:48 +0000 (02:33 +0000)]
b=15428
r=adilger
Fixed a crash issue for raid56-zerocopy
grev [Fri, 13 Feb 2009 19:52:53 +0000 (19:52 +0000)]
b=18333
i=Oleg.Drokin
add flock to default mount opts
johann [Fri, 13 Feb 2009 19:46:20 +0000 (19:46 +0000)]
Branch b1_6
Fix typo in previous shadow's commit.
grev [Fri, 13 Feb 2009 13:41:43 +0000 (13:41 +0000)]
fix typo for
b=16488
i=Oleg.Drokin
RACER acc-sm test suit
yangsheng [Fri, 13 Feb 2009 11:18:39 +0000 (11:18 +0000)]
Branch b1_6
b=18462
i=adilger, johann
Remove the linux/fs.h changes from kernel patch.
shadow [Fri, 13 Feb 2009 10:20:41 +0000 (10:20 +0000)]
fix typo
Branch b1_6
grev [Thu, 12 Feb 2009 18:54:12 +0000 (18:54 +0000)]
b=17839
i=Brian
recovery-mds-scale for SLOW=no
grev [Thu, 12 Feb 2009 16:16:00 +0000 (16:16 +0000)]
revert b=17839 commit
grev [Thu, 12 Feb 2009 15:28:48 +0000 (15:28 +0000)]
b=17839
i=Brian
recovery-mds-scale for SLOW=no
bobijam [Thu, 12 Feb 2009 06:43:06 +0000 (06:43 +0000)]
Branch b1_6
b=18015
sles9 doesn't define SLAB_DESTROY_BY_RCU.
girish [Wed, 11 Feb 2009 18:07:40 +0000 (18:07 +0000)]
b=18417
o=h.huang
i=girish
Fix typo from last commit which caused test-framework to exit wrongly
shadow [Wed, 11 Feb 2009 08:42:10 +0000 (08:42 +0000)]
update client to support vanila kernels up to 2.6.27.
Branch b1_6
b=14250
i=green
i=deen
girish [Wed, 11 Feb 2009 06:54:36 +0000 (06:54 +0000)]
b=18417
i=kalpak
i=girish
o=adilger
cleanup correctly after sanity 102 tests
jxiong [Tue, 10 Feb 2009 06:50:02 +0000 (06:50 +0000)]
b=15428
r=adilger
Fix a crash issue for raid5 zerocopy.
fanyong [Tue, 10 Feb 2009 04:58:56 +0000 (04:58 +0000)]
Branch b1_6
b=18294
i=tianzy
i=panda
Diagnostic patch to catch what's the garbage left in the system when quota error.
tianzy [Tue, 10 Feb 2009 04:29:16 +0000 (04:29 +0000)]
Branch b1_6
add an entry under /proc to support what policy(RR or QOS) will be used.
b=18334
i=nathan
i=adilger
grev [Tue, 10 Feb 2009 00:22:41 +0000 (00:22 +0000)]
b=18219
i=Adilger
test_70b one client fix
grev [Mon, 9 Feb 2009 23:40:53 +0000 (23:40 +0000)]
b=18366
i=Johann
test_5 ENOSPC fix
johann [Mon, 9 Feb 2009 15:18:15 +0000 (15:18 +0000)]
Branch b1_6
b=18374
i=johann (patch from Tappro)
i=shadow
Return only valid cookies, empty slot for failed ones.
This fixes a memory corruption causing random oops.
johann [Mon, 9 Feb 2009 15:16:39 +0000 (15:16 +0000)]
Branch b1_6
b=18374
i=shadow
i=tappro
Wait for mds_fl_synced to be set during recovery before allowing clients
to reconnect. This is done to make sure that llog is properly set up.
johann [Mon, 9 Feb 2009 15:15:00 +0000 (15:15 +0000)]
Branch b1_6
b=18067
i=johann
i=adilger
Change oap magic LASSERT into LASSERTF to aid debugging.
anserper [Sun, 8 Feb 2009 23:35:45 +0000 (23:35 +0000)]
kernel support clarification
green [Sun, 8 Feb 2009 22:36:41 +0000 (22:36 +0000)]
b=16823
r=adilger,shadow
Update offset calculations for 32 bit platforms
anserper [Sun, 8 Feb 2009 10:16:39 +0000 (10:16 +0000)]
warning clarification
huanghua [Fri, 6 Feb 2009 02:27:56 +0000 (02:27 +0000)]
Branch b1_6
b=3633
i=robert.read
i=oleg.drokin
i=tappro
clear the "hold for replay" flag for open request once a file is closed
bobijam [Fri, 6 Feb 2009 02:00:43 +0000 (02:00 +0000)]
Branch b1_6
b=17011
i=tom.wang
i=joahnn
remove redundant brw_stats on obdfilter.
shadow [Thu, 5 Feb 2009 11:05:48 +0000 (11:05 +0000)]
handle MDS connect in failover as real connect.
Branch b1_6
b=18304
i=johann
i=green
bobijam [Thu, 5 Feb 2009 02:04:56 +0000 (02:04 +0000)]
Branch b1_6
b=18225
o=Nic Henke(nic@cray.com)
i=johann
i=adilger
use obd_brw_async in echo_client_kbrw.
bobijam [Wed, 4 Feb 2009 01:35:38 +0000 (01:35 +0000)]
Branch b1_6
b=18015
i=johann
i=sheng.yang
add SLAB_DESTROY_BY_RCU to ldlm_lock_slab's flags for ldlm_locks are freed
with RCU mechanism.
girish [Tue, 3 Feb 2009 19:24:46 +0000 (19:24 +0000)]
b=18417
i=adilger
i=kalpak
missed hunk from last commit
girish [Tue, 3 Feb 2009 19:09:44 +0000 (19:09 +0000)]
b=18417
i=adilger
i=kalpak
remove star tests from sanity.sh
grev [Tue, 3 Feb 2009 13:39:55 +0000 (13:39 +0000)]
b=17899
i=Johann
test_53* fix: rm dirs used at start
tianzy [Tue, 3 Feb 2009 06:44:26 +0000 (06:44 +0000)]
Branch b1_6
use QUOTA_REQ_RETURNED to replace -EAGAIN when a quota req is finished.
b=18307
i=johann
i=panda
fanyong [Tue, 3 Feb 2009 02:34:54 +0000 (02:34 +0000)]
Branch b1_6
b=18135
i=tianzy
i=andrew.perepechko
Wake up all wait threads on the "lq_waitq" when "lqc_import" is invalid.
shadow [Mon, 2 Feb 2009 17:32:13 +0000 (17:32 +0000)]
don't call obd_disconnect under lov_lock.
Branch b1_6
b=17310
i=johann
i=rread
grev [Mon, 2 Feb 2009 13:55:41 +0000 (13:55 +0000)]
remove debug line
grev [Mon, 2 Feb 2009 12:49:04 +0000 (12:49 +0000)]
b=16818
i=Nathan
get rid of set_obd_timeout(), improve test_26*
kalpak [Mon, 2 Feb 2009 07:17:47 +0000 (07:17 +0000)]
b=17895
i=adilger
i=alex
While using HA for Lustre servers with Linux RAID, it is possible that MMP will not detect multiple mounts. To make this work we need to unplug the device queue in RAID when the MMP block is being written. Also while reading the MMP block, we should read it from disk and not the cached one.
shadow [Wed, 28 Jan 2009 09:35:24 +0000 (09:35 +0000)]
fix typo
shadow [Wed, 28 Jan 2009 09:14:36 +0000 (09:14 +0000)]
don't lose wakeup for imp_recovery_waitq
Branch b1_6
b=18154
i=nathan
i=rread
girish [Wed, 28 Jan 2009 03:52:57 +0000 (03:52 +0000)]
b=16895
i=adilger (o=johann)
i=girish
dont assert on lvfs_check_rdonly()
yury [Tue, 27 Jan 2009 14:27:23 +0000 (14:27 +0000)]
- added a comment about LDLM_FL_* flags at Adilger's request.
yury [Tue, 27 Jan 2009 12:38:05 +0000 (12:38 +0000)]
b=17282
r=johann,shadow
- simplifies a lot lru resize memory presssure handling. No estimation pass is needed right now as it was too CPU consuming.
johann [Mon, 26 Jan 2009 13:58:26 +0000 (13:58 +0000)]
Branch b1_6
b=18304
i=shadow
i=nathan
MSG_CONNECT_INITIAL is not set on the initial MDS->OST connect.
As a conseqence, the patch from bug 18224 is not operational
and the MDS export cannot be reused on the OSTs until it gets
evicted.
johann [Mon, 26 Jan 2009 13:45:28 +0000 (13:45 +0000)]
Branch b1_6
b=18307
i=adilger
i=shadow
send OST_CONNECT to OST_REQUEST_PORTAL instead of OST_IO_PORTAL (was used
for compatibility with releases < 1.4.6)
yangsheng [Sat, 24 Jan 2009 17:28:51 +0000 (17:28 +0000)]
Branch b1_6
b=18091
i=adilger, tappro
Clean up IOC_{INIT, PACK, UNPACK} macros.
grev [Sat, 24 Jan 2009 09:19:53 +0000 (09:19 +0000)]
b=18045
i=Adilger
-- get rid of Hendrix rates values
-- create NUM_FILES -le IFree
-- use mdsrate for cleanup instead of rm
yangsheng [Fri, 23 Jan 2009 17:22:26 +0000 (17:22 +0000)]
Branch b1_6
b=17464
yangsheng [Fri, 23 Jan 2009 17:04:41 +0000 (17:04 +0000)]
Branch b1_6
b=17464
Recommit lost part(I wonder why it lost).
girish [Fri, 23 Jan 2009 16:27:51 +0000 (16:27 +0000)]
b=15524
i=huanghua
i=kalpak
fix test 102f
shadow [Fri, 23 Jan 2009 14:29:22 +0000 (14:29 +0000)]
add workaround for leak space in ESTALE case
Branch b1_6
grev [Fri, 23 Jan 2009 13:44:33 +0000 (13:44 +0000)]
b=18279
i=Nathan
AT tests cleanup: run setup/cleanup only if AT tests run
vitaly [Fri, 23 Jan 2009 10:18:33 +0000 (10:18 +0000)]
Branch b1_6
b=17748
i=grev
i=adilger
sanityN test fix
tianzy [Fri, 23 Jan 2009 09:50:21 +0000 (09:50 +0000)]
Branch b1_6
when there is any inactive ost, refuse to do quotacheck
b=18151
i=johann
i=andrew.perepechko
yangsheng [Fri, 23 Jan 2009 09:16:11 +0000 (09:16 +0000)]
Branch b1_6
b=17464
i=johann, bobijam
Swab lov_mds_md when ioctl be called directly in liblustre.
tianzy [Fri, 23 Jan 2009 06:21:11 +0000 (06:21 +0000)]
Branch b1_6
fix the test issue of test_12 in sanity-quota.sh
b=18151
i=johann
i=panda
girish [Wed, 21 Jan 2009 18:38:40 +0000 (18:38 +0000)]
b=16895
i=adilger (o=johann)
i=girish
don't LASSERT on read-only device
yury [Wed, 21 Jan 2009 16:43:35 +0000 (16:43 +0000)]
b=17310
r=johann
- tmp fix for long connect rpc in sending list on umount.
fanyong [Wed, 21 Jan 2009 03:28:20 +0000 (03:28 +0000)]
Branch b1_6
b=15691
i=huanghua
i=yep
Another diagnostic patch to collect more log when error.
tianzy [Wed, 21 Jan 2009 02:56:54 +0000 (02:56 +0000)]
Branch b1_6
take refcount of export when quotacheck is going on
b=18126
i=shadow
i=panda
shadow [Tue, 20 Jan 2009 13:04:37 +0000 (13:04 +0000)]
always handle mds connect as reconnect.
Branch b1_6
b=18224
i=johhan
i=green
i=adilger
i=wangdi
yangsheng [Tue, 20 Jan 2009 10:27:35 +0000 (10:27 +0000)]
Branch b1_6
b=17153
i=adilger, tappro
Remove 2.4 compatibility from b1_6.
shadow [Tue, 20 Jan 2009 08:57:00 +0000 (08:57 +0000)]
don't confuse mds_finish_transno() with PTR_ERR(-ENOENT)
Branch b1_6
b=18238
i=green
i=umka
kalpak [Tue, 20 Jan 2009 06:43:01 +0000 (06:43 +0000)]
b=17942
i=adilger
i=girish
update fiemap structures and API to match upstream
fanyong [Tue, 20 Jan 2009 01:43:21 +0000 (01:43 +0000)]
Branch b1_6
b=18182
i=tianzy
i=panda
Set stripe count as 1 and stripte index as 0 to make sure the file is created on first ost.
shadow [Mon, 19 Jan 2009 21:18:09 +0000 (21:18 +0000)]
release lock refernce only after add page to pages list.
Branch b1_6
b=17972
i=shadow
i=johann
johann [Mon, 19 Jan 2009 12:17:34 +0000 (12:17 +0000)]
Branch b1_6
b=18044
i=vitaly
i=bobijam
copy DISCARD_DATA flag if cp ast includes bl ast.
zhanghc [Sun, 11 Jan 2009 15:59:24 +0000 (15:59 +0000)]
b=17887
1, print the actual time to wait for
lock cancellation in "waiting_locks_callback"
2, compare resoure in "ost_rw_hpreq_lock_match"
3, replace "LDLM_ERROR" with "LDLM_WARN" if a lock
is completed after waiting more than obd_timeout
i=vitaly.fertman
i=johann