Whamcloud - gitweb
fs/lustre-release.git
9 years agob=22560 fix "obd_connect_names[]" for "OBD_CONNECT_FULL20" introduced
Fan Yong [Thu, 20 May 2010 15:18:32 +0000 (23:18 +0800)]
b=22560 fix "obd_connect_names[]" for "OBD_CONNECT_FULL20" introduced

fix "obd_connect_names[]" for "OBD_CONNECT_FULL20" introduced

i=robert.read
i=andreas.dilger

9 years agob=21485 allocate lcd inside obd_init_export()
Mikhail Pershin [Thu, 20 May 2010 15:03:05 +0000 (19:03 +0400)]
b=21485 allocate lcd inside obd_init_export()

i=rread
i=zam

9 years agob=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated
Fan Yong [Thu, 20 May 2010 05:54:12 +0000 (13:54 +0800)]
b=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated

MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated.

i=robert.read
i=andreas.dilger

9 years agob=22827 restore locking for thread->t_flags
Alexander.Zarochentsev [Thu, 20 May 2010 07:33:48 +0000 (11:33 +0400)]
b=22827 restore locking for thread->t_flags

i=tappro
i=robert.read

9 years agob=20938 Add liblustreapi.a to dependencies lists
Robert Read [Thu, 13 May 2010 23:59:42 +0000 (16:59 -0700)]
b=20938 Add liblustreapi.a to dependencies lists

I think this is why we occasionally file to build lustre on LBATS
with a missing liblustreapi.a.

i=brian

9 years agob=22458 fix concurrent mgs lock revocation.
Eric Mei [Tue, 18 May 2010 14:59:35 +0000 (08:59 -0600)]
b=22458 fix concurrent mgs lock revocation.

r=nathan
r=rread

9 years agob=15587 ignore security.capability xattr on client side
Mikhail Pershin [Tue, 18 May 2010 05:52:31 +0000 (09:52 +0400)]
b=15587 ignore security.capability xattr on client side

i=adilger
i=johann

9 years agob=21681 Quiet bogus previously committed transno error
Mikhail Pershin [Tue, 18 May 2010 05:43:29 +0000 (09:43 +0400)]
b=21681 Quiet bogus previously committed transno error

i=zhang,panda

9 years agob=22455 remove "lnet." prefix from lctl params display
LiuYing [Tue, 18 May 2010 02:48:24 +0000 (10:48 +0800)]
b=22455 remove "lnet." prefix from lctl params display

remove "lnet." prefix from lctl params display and change one
"memused" to "lnet_memused".

o=adilger
i=johann
i=emoly.liu
i=rread

9 years agob=22731 server should not fall into LBUG if client send invalid parameter
Fan Yong [Tue, 18 May 2010 02:35:09 +0000 (10:35 +0800)]
b=22731 server should not fall into LBUG if client send invalid parameter

server should not fall into LBUG if client send invalid parameter

i=robert.read
i=di.wang

9 years agob=22560 introduce "OBD_CONNECT_FULL20" to distinguish 1.8 client from 2.0 one for...
Fan Yong [Tue, 18 May 2010 02:24:44 +0000 (10:24 +0800)]
b=22560 introduce "OBD_CONNECT_FULL20" to distinguish 1.8 client from 2.0 one for different checksum policy

Introduce "OBD_CONNECT_FULL20" to distinguish 1.8 client from 2.0 one for different checksum policy:
1) for 1.8 client, use fixed first 88 bytes of ptlrpc_body
2) for 2.0 client, use lm_buflens

i=andreas.dilger
i=robert.read

9 years agob=15253 add conf_param -d to remove permanent settings
Nathan Rutman [Fri, 14 May 2010 16:18:43 +0000 (09:18 -0700)]
b=15253 add conf_param -d to remove permanent settings

i=adilger
i=rread

9 years agob=22625 Fix libcfs_debug_file_path module option
Rahul Deshmukh [Fri, 14 May 2010 10:29:39 +0000 (15:59 +0530)]
b=22625 Fix libcfs_debug_file_path module option

i=rahul

Landing patch by Brian Behlendorf <behlendorf1@llnl.gov>,
fix libcfs_debug_file_path module option

9 years agob=21945 Adding WIRE_ATTR to lnet structures traversing network
Maxim Patlasov [Fri, 14 May 2010 09:11:39 +0000 (13:11 +0400)]
b=21945 Adding WIRE_ATTR to lnet structures traversing network

i=liang
i=isaac
LST passed some lnet structures via network neglecting the lack of WIRE_ATTR attribute. This resulted in incompatibility of LST-s running on different platforms.

9 years agob=22455 add "list_param -R"
LiuYing [Fri, 14 May 2010 00:47:07 +0000 (08:47 +0800)]
b=22455 add "list_param -R"

list parameters recursively with the "-R" option

o=adilger
i=emoly.liu
i=nathan

9 years agoUpdated for build 42. 1.10.0.42 v1_10_0_42
Terry Rutledge [Thu, 13 May 2010 20:30:07 +0000 (13:30 -0700)]
Updated for build 42.

9 years agoRevert "b=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated"
Robert Read [Wed, 12 May 2010 19:07:34 +0000 (12:07 -0700)]
Revert "b=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated"

Hit ASSERTION(attr->la_blocks == 0), see bug 22802.

This reverts commit 33b4bafea13bd2cfe90dba3a8651a175683f3999.

9 years agob=22573 do not skip previously granted locks in osc_lock_enqueue_wait().
Vitaly Fertman [Wed, 12 May 2010 11:04:16 +0000 (15:04 +0400)]
b=22573 do not skip previously granted locks in osc_lock_enqueue_wait().

i=eric.mei
i=wangdi

as CLIO adds new locks to the tail, walk through the head of the queue
to cancel overlapping conflicting locks on enqueue.

9 years agob=22518 mount client2 at the start of tests, disable COS, fix test 10
Mikhail Pershin [Wed, 12 May 2010 16:25:20 +0000 (20:25 +0400)]
b=22518 mount client2 at the start of tests, disable COS, fix test 10

i=grev

9 years agob=18857 enhance seq allocation scalability by updating seq data asynchronously.
pravin [Wed, 12 May 2010 14:25:24 +0000 (19:55 +0530)]
b=18857 enhance seq allocation scalability by updating seq data asynchronously.

this patch also removes seq replay. ref bug for details.

a=pravin,tappro
i=tappro
i=alexander.zarochentsev
i=pravin

9 years agob=21140 Fix srv_threads_running counting.
Alexander.Zarochentsev [Wed, 12 May 2010 10:29:19 +0000 (14:29 +0400)]
b=21140 Fix srv_threads_running counting.

It was possible to overload n_active_request processing incoming requests and
break the thread reservation logic. Likely, it was responsible to the long
processing of requests.

The patch makes srv_threads_running to exactly count only running not sleeping
threads. All threads accounting and comparing/reservation of threads are done
under the service spinlock so it produce a reliable result. The thread
reservation logic is based on new srv_threads_running value and cannot be
confused by not active sleeping threads. The thread reservation logic is
concentrated now in one place, where the wakeup condition is checked (now in
ptlrpc_main_check_event), once a thread is woken up, it is counted as running
and does further work w/o additional checks.

i=zhen.liang
i=robert.read

9 years agob=22683 remove unnecessary check and assert in the cfs_hash function.
Wang Di [Tue, 11 May 2010 17:58:48 +0000 (13:58 -0400)]
b=22683 remove unnecessary check and assert in the cfs_hash function.

o=Eric.Barton
i=Robert.Read
i=Di.Wang

9 years agob=13698 allow e2fsck part of lfsck.sh to be run without lfsck
Vladimir Saveliev [Tue, 11 May 2010 06:40:46 +0000 (10:40 +0400)]
b=13698 allow e2fsck part of lfsck.sh to be run without lfsck

this combines initial patch from Andreas (https://bugzilla.lustre.org/attachment.cgi?id=29696)
and several necessary fixes (https://bugzilla.lustre.org/attachment.cgi?id=29747)

i=adilger

9 years agob=20562 increase LUSTRE_SEQ_META_WIDTH to keep FLD compact
Rahul Deshmukh [Tue, 11 May 2010 06:25:30 +0000 (11:55 +0530)]
b=20562 increase LUSTRE_SEQ_META_WIDTH to keep FLD compact

Fixed the build error for fc11 and fc12 patchless client.

i=rread

9 years agob=22598 diagnostic patch for lock cancel callback error processing
Fan Yong [Tue, 11 May 2010 03:01:45 +0000 (11:01 +0800)]
b=22598 diagnostic patch for lock cancel callback error processing

diagnostic patch for lock cancel callback error processing.

i=robert
i=di.wang

9 years agob=19986 cleanup lock to eliminate former test cases effect before replay-single test_53
Fan Yong [Tue, 11 May 2010 01:55:54 +0000 (09:55 +0800)]
b=19986 cleanup lock to eliminate former test cases effect before replay-single test_53

cleanup lock to eliminate former test cases effect before replay-single test_53

i=robert
i=di.wang

9 years agob=18143 Make VBR compatible with pdirops.
Mikhail Pershin [Sun, 9 May 2010 09:09:23 +0000 (13:09 +0400)]
b=18143 Make VBR compatible with pdirops.

i=zam
i=bzzz

9 years agob=22283 clarify writeconf in man page
Nathan Rutman [Fri, 7 May 2010 22:33:29 +0000 (15:33 -0700)]
b=22283 clarify writeconf in man page

9 years agob=22671 Check for modules directly instead of keeping state
Nathan Rutman [Fri, 7 May 2010 19:29:07 +0000 (12:29 -0700)]
b=22671 Check for modules directly instead of keeping state

i=nico
i=rread

9 years agob=22581 LOADS env var in ncli.sh should allow overwrite
Elena Gryaznova [Fri, 7 May 2010 16:07:54 +0000 (20:07 +0400)]
b=22581 LOADS env var in ncli.sh should allow overwrite

i=Minh.Diep

9 years agob=22522 do not remove from res_list without locks
Oleg Drokin [Wed, 5 May 2010 23:35:58 +0000 (19:35 -0400)]
b=22522 do not remove from res_list without locks

Patch in bug 21501 moved list manipulation of res_list outside or res_lock
introducing a race window in flock code. Move it back under the lock.

i=rread
i=adilger

9 years agob=22669 fix fault page index handler in newer kernel.
Eric Mei [Wed, 5 May 2010 01:14:34 +0000 (19:14 -0600)]
b=22669 fix fault page index handler in newer kernel.

r=wangdi
r=rread

9 years agob=21502 symlink compatibility between 1.6 and 2.0
Rahul Deshmukh [Tue, 4 May 2010 15:16:01 +0000 (20:46 +0530)]
b=21502 symlink compatibility between 1.6 and 2.0

Fixed the sleep in spin lock, inode->i_sb->s_op-> dirty_inode(inode)
was called with spin lock held.

i=bzzz
i=pravin

9 years agob=22683 don't manipulate hash in lov_sub_enter/lov_sub_exit.
Eric Mei [Tue, 4 May 2010 14:30:09 +0000 (08:30 -0600)]
b=22683 don't manipulate hash in lov_sub_enter/lov_sub_exit.

r=wangdi
r=rread

9 years agoRevert "b=19427 correct lmm_object_id and reserve fids for fid-on-OST." 1.10.0.41a v1_10_0_41a
Robert Read [Fri, 30 Apr 2010 16:23:28 +0000 (09:23 -0700)]
Revert "b=19427 correct lmm_object_id and reserve fids for fid-on-OST."

This reverts commit 4c01e64e0a72c1682ebf0a8bd4cccf99fd04cd88.

This caused the interop issue seen in bug 22730.

9 years agoPrepare for Build 41 1.10.0.41 v1_10_0_41
Robert Read [Thu, 29 Apr 2010 22:09:34 +0000 (15:09 -0700)]
Prepare for Build 41

9 years agob=16680 remove some noisy debug messages
Fan Yong [Thu, 29 Apr 2010 06:06:37 +0000 (14:06 +0800)]
b=16680 remove some noisy debug messages

Remove some noisy debug messages.

i=robert.read
i=rahul.deshmukh

9 years agob=20326 Test suite for MMP feature
Jian Yu [Thu, 29 Apr 2010 05:59:13 +0000 (13:59 +0800)]
b=20326 Test suite for MMP feature

Tests for multiple mount protection (MMP) feature.

i=andreas.dilger
i=grev

9 years agob=22069 port "llapi_get_connect_flags()" API from b1_8 to master
Fan Yong [Thu, 29 Apr 2010 05:37:03 +0000 (13:37 +0800)]
b=22069 port "llapi_get_connect_flags()" API from b1_8 to master

Port "llapi_get_connect_flags()" API from b1_8 to master.

i=robert.read
i=landen

9 years agob=22075 buffalo-v2 should detect test timeouts
Manoj Joseph [Thu, 29 Apr 2010 03:27:21 +0000 (21:27 -0600)]
b=22075 buffalo-v2 should detect test timeouts

buffalo-v2 now detects test timeouts. It now generates status entries
in results.yml after a sub-test completes. If the test status is missing
a timeout is assumed to have occured.

i=robert.read
i=grev

9 years agob=21962 Quote the error message in results.yaml
Manoj Joseph [Thu, 29 Apr 2010 03:27:20 +0000 (21:27 -0600)]
b=21962 Quote the error message in results.yaml

Quote and escape the error message in results.yaml

i=robert.read
i=nicolas.williams

9 years agob=22582 remove leading / from fid2path results print full path if given mountpoint
Nathan Rutman [Wed, 28 Apr 2010 18:04:59 +0000 (11:04 -0700)]
b=22582 remove leading / from fid2path results print full path if given mountpoint

i=manoj
i=rread

9 years agob=22456 Remove files for unsupported kernels
Robert Read [Wed, 28 Apr 2010 17:06:59 +0000 (10:06 -0700)]
b=22456 Remove files for unsupported kernels

Removes support for fc3, fc5, rhel4, sles10, 2.6.18 vanilla and 2.6.22 vanilla.

i=adilger

9 years agob=18649 set wait_recovery_complete() MAX value to max recovery time estimated
Elena Gryaznova [Tue, 27 Apr 2010 15:32:56 +0000 (19:32 +0400)]
b=18649 set wait_recovery_complete() MAX value to max recovery time estimated

i=Mikhail.Pershin

9 years agob=20918 t-f max recovery time estimation
Elena Gryaznova [Tue, 27 Apr 2010 15:23:04 +0000 (19:23 +0400)]
b=20918 t-f max recovery time estimation

i=Nathan.Rutman
i=Brian.Murrell

inform user about the estimated maximum recovery time value
to help him to set the server FAILOVER period properly

9 years agob=22342 process racer condition between statahead and rename/unlink operation
Fan Yong [Tue, 27 Apr 2010 05:23:03 +0000 (13:23 +0800)]
b=22342 process racer condition between statahead and rename/unlink operation

1) process racer condition between statahead and rename/unlink operation
2) replace "lli_lock" with "lli_sa_lock" for statahead related processing

i=robert.read
i=tom.wang

9 years agob=22634 hold "mds_qonoff_sem" when call "lustre_read_quota()", and check parameter...
Fan Yong [Tue, 27 Apr 2010 03:36:53 +0000 (11:36 +0800)]
b=22634 hold "mds_qonoff_sem" when call "lustre_read_quota()", and check parameter properly in such function

1) replace "cfs_semaphore_t" with "cfs_rw_semaphore_t" for "mds_qonoff_sem" to enhance the parallel processing of quota related operations
2) hold "mds_qonoff_sem" when call "lustre_read_quota()", and check parameter properly in such function

i=robert.read
i=landen

9 years agob=22614 enlarge MDSSIZE/OSTSIZE to increase default journal size for conf-sanity
Fan Yong [Tue, 27 Apr 2010 02:51:27 +0000 (10:51 +0800)]
b=22614 enlarge MDSSIZE/OSTSIZE to increase default journal size for conf-sanity

1) enlarge MDSSIZE/OSTSIZE to increase default journal size for conf-sanity
2) journal handler error process in lustre_commit_dquot

i=robert.read
i=landen

9 years agob=21251 ha.sh: Fix ha_wait_loads and ha_dump_logs
Li Wei [Tue, 27 Apr 2010 02:28:09 +0000 (10:28 +0800)]
b=21251 ha.sh: Fix ha_wait_loads and ha_dump_logs

Report existence of hanging workloads.  Ignore "lctl dk" failures, since
some nodes may be down.

i=robert.read

9 years agob=21251 Add lustre/tests/ha.sh
Li Wei [Tue, 27 Apr 2010 02:28:08 +0000 (10:28 +0800)]
b=21251 Add lustre/tests/ha.sh

This is a simple failover test script that works with configurations
controlled by a CRM and have multiple targets per server.

i=robert.read
i=grev

9 years agoRevert "b=21379 Fix orphans proceeding in osc_create"
Robert Read [Mon, 26 Apr 2010 22:11:52 +0000 (15:11 -0700)]
Revert "b=21379 Fix orphans proceeding in osc_create"

This reverts commit 2deb4f149f4601f9128fd39efd4705573520f277.

9 years agob=22458 move lcw_dump out of softirq context.
Eric Mei [Mon, 26 Apr 2010 14:59:52 +0000 (08:59 -0600)]
b=22458 move lcw_dump out of softirq context.

Now the message dump is done in thread context in lc_watchdogd.

r=rread
r=nathan

9 years agob=21128 run sync ldlm_bl_to_thread_list() in separate thread to save stack space.
pravin [Mon, 26 Apr 2010 13:15:26 +0000 (18:45 +0530)]
b=21128 run sync ldlm_bl_to_thread_list() in separate thread to save stack space.

i=oleg.drokin
i=rahul

9 years agob=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated
Fan Yong [Mon, 26 Apr 2010 07:38:37 +0000 (15:38 +0800)]
b=22637 MDS returns OBD_MD_FLSIZE to client only when no OSS object allocated

i=robert.read
i=landen

9 years agob=22513 Remove unecessary lock in read-ahead process.
Wang Di [Mon, 26 Apr 2010 03:57:44 +0000 (23:57 -0400)]
b=22513 Remove unecessary lock in read-ahead process.

i=Robert.Read
i=Eric.Mei

9 years agob=21938 use the same set during replay
hongchao.zhang [Wed, 21 Apr 2010 00:54:53 +0000 (08:54 +0800)]
b=21938 use the same set during replay

some requests use its own ptlrpc_request_set to process its requests, but Lustre
will use a specific ptlrpc_request_set to process the requests during recovery.
this patch fixes this problem to allow the requests to use its own set if it have
one

i=johann@sun.com
i=tappro@sun.com

9 years agob=15936 Unified target cleanups v2
Mikhail Pershin [Sun, 25 Apr 2010 12:09:48 +0000 (16:09 +0400)]
b=15936 Unified target cleanups v2

i=rread
i=andreas

9 years agob=20373 Putting parent lock for rep-ack on create is wasteful
Cliff White [Tue, 6 Apr 2010 06:40:15 +0000 (23:40 -0700)]
b=20373 Putting parent lock for rep-ack on create is wasteful

Do not put locks if no create.

i=robert.read
i=tappro

9 years agob=22310 add a little more comment.
Eric Mei [Fri, 23 Apr 2010 18:48:07 +0000 (12:48 -0600)]
b=22310 add a little more comment.

r=adilger

9 years agob=21678 Add more debug info to lnd_query code path
Isaac Huang [Fri, 23 Apr 2010 04:03:14 +0000 (00:03 -0400)]
b=21678 Add more debug info to lnd_query code path

The peer health code lacked some important debugging info in lnd_query
code paths. This patch added necessary debug prints, not just for bug
21678, but also for future troubleshooting.

i=liang
i=maxim

9 years agob=19427 correct lmm_object_id and reserve fids for fid-on-OST.
Wang Di [Fri, 23 Apr 2010 19:57:04 +0000 (12:57 -0700)]
b=19427 correct lmm_object_id and reserve fids for fid-on-OST.

1. Change lmm_object_id to fid.
2. Cleanup fid spaces reservation (for fid-on-OST).http://arch.lustre.org/index.php?title=Interoperability_fids_zfs#NEW.0
3. Rename group to Seq.

i= Andreas.diger
i= pravin.shelar

9 years agob=22615 fixes for regressions caused by 11063
Vladimir Saveliev [Thu, 22 Apr 2010 19:15:32 +0000 (12:15 -0700)]
b=22615 fixes for regressions caused by 11063

set atime to past under PW EOF extent lock
fix truncate in liblustre

i=vitaly
i=ericm

9 years agob=22507 rm -rf not replicated
Manoj Joseph [Thu, 22 Apr 2010 19:15:28 +0000 (12:15 -0700)]
b=22507 rm -rf not replicated

Support replication of recursive directory removal.

i=nathan.rutman
i=robert.read

9 years agob=22520 set the thread to be uninterrupt before add to waitq
Wang Di [Thu, 22 Apr 2010 19:15:24 +0000 (12:15 -0700)]
b=22520 set the thread to be uninterrupt before add to waitq

In lov_subobject_kill, if the thread needs to wait the object being
freed, it should set the thread to be uninterrupt, otherwise, the thread
might spin there.

i=Eric.mei
i=Robert

9 years agob=22296 Fix script problem for recovery-double-scale
Wang Di [Thu, 22 Apr 2010 19:15:20 +0000 (12:15 -0700)]
b=22296 Fix script problem for recovery-double-scale

Force the test threads stopped before shutdown the clients in
recovery-double-scale.

i=Jack.Chen
i=WangDi

9 years agob=22161 Use LCK_PW for parent lock in mdt_link(). Pdirops test set
Mikhail Pershin [Wed, 21 Apr 2010 18:43:20 +0000 (11:43 -0700)]
b=22161 Use LCK_PW for parent lock in mdt_link(). Pdirops test set

i=adilger
i=bzzz
i=rread

9 years agob=19919 Supply a absolute path.
yangsheng [Wed, 21 Apr 2010 18:43:12 +0000 (11:43 -0700)]
b=19919 Supply a absolute path.

i=andreas
i=johann

9 years agob=22190 Apply 19195 patch to add tls data for recovery thread.
Mikhail Pershin [Wed, 21 Apr 2010 18:43:11 +0000 (11:43 -0700)]
b=22190 Apply 19195 patch to add tls data for recovery thread.

This will be needed anyway when sync journal will be ported

i=zam
i=oleg

9 years agofixed for bug 22237
Rahul Deshmukh [Wed, 21 Apr 2010 18:42:26 +0000 (11:42 -0700)]
fixed for bug 22237

b=22237 replay-single test-13 mmp failure, BUG: warning at
fs/proc/generic.c:764/remove_proc_entry()

The proc entry EXT4_MAX_DIR_SIZE_NAME was not remove in cleanup path. It
is now fixed.

i=johann

9 years agob=22129 add debug patch for 22129
Wang Di [Wed, 21 Apr 2010 18:42:24 +0000 (11:42 -0700)]
b=22129 add debug patch for 22129

i=Robert
i=Fangyong

9 years agob=21379 Fix orphans proceeding in osc_create
Wang Di [Wed, 21 Apr 2010 18:42:20 +0000 (11:42 -0700)]
b=21379 Fix orphans proceeding in osc_create

in "osc_create", there is no need to proceed after the orphans are deleted, which cause the
oscc->oscc_next_id is increased by 1 and break the balance between it and the last objid stored
in "mds->mds_lov_page_array"

o=Zhang Hongchao
i=Johann
I=Tappro

9 years agob=22409 Spurious error messages from smp_processor_id() on preemptible kernel
Dmitry Zogin [Mon, 19 Apr 2010 23:29:32 +0000 (16:29 -0700)]
b=22409 Spurious error messages from smp_processor_id() on preemptible kernel

 Disable a preemption by grabbing the lock in fs_trace_get_tcd() first.
 The function fs_trace_get_tcd() was moved up.

 o=andreas.dilger
 i=johann
 i=dmitry.zogin
 i=nathan.rutman

9 years agob=20953 sanity-quota test 30 fixes
Andrew Perepechko [Mon, 19 Apr 2010 23:29:30 +0000 (16:29 -0700)]
b=20953 sanity-quota test 30 fixes

i=Yong Fan

9 years agob=22310 temporary fix: align readahead window end to 1M rpc boundary.
Eric Mei [Mon, 19 Apr 2010 23:29:27 +0000 (16:29 -0700)]
b=22310 temporary fix: align readahead window end to 1M rpc boundary.

r=wangdi
r=rread

9 years agob=22310 do not break early in osc_io_submit() which might cause sub-optimal rpc size...
Eric Mei [Mon, 19 Apr 2010 23:29:25 +0000 (16:29 -0700)]
b=22310 do not break early in osc_io_submit() which might cause sub-optimal rpc size for read and synchronous write.

i=wangdi
i=rread

9 years agob=21949 echo code fixes.
Vitaly Fertman [Mon, 19 Apr 2010 23:29:21 +0000 (16:29 -0700)]
b=21949 echo code fixes.

i=eric.may
i=bzzz

1. access to a wrong page on echo server commit;
2. wrong lock type enqueue in echo client test_brw code;
3. wrong thread offset initialization in lctl test_brw code;

9 years agob=18478 Fix sanity test_180 to write to correct obj
Robert Read [Mon, 19 Apr 2010 18:24:25 +0000 (11:24 -0700)]
b=18478 Fix sanity test_180 to write to correct obj

Write small amount of data to the object that was created. This
also splits 180 into two tests for using echo client with an osc
and directly to an obdfilter.

i=vitaly

9 years agob=20647 Ensure all needed pids exits before failover in test set 53
Mikhail Pershin [Fri, 16 Apr 2010 20:31:21 +0000 (13:31 -0700)]
b=20647 Ensure all needed pids exits before failover in test set 53

i=rread
i=grev

9 years agob=21826 a small fix for quota files invalidation
Andrew Perepechko [Fri, 16 Apr 2010 20:31:18 +0000 (13:31 -0700)]
b=21826 a small fix for quota files invalidation

i=Johann Lombardi
i=ZhiYong Tian

9 years agob=20615 always send ZC_ACK on non-blocking connection
Liang Zhen [Fri, 16 Apr 2010 20:31:13 +0000 (13:31 -0700)]
b=20615 always send ZC_ACK on non-blocking connection

i=isaac
i=maxim

9 years agob=18948 Speedy recovery
Mikhail Pershin [Fri, 16 Apr 2010 20:30:30 +0000 (13:30 -0700)]
b=18948 Speedy recovery

Add hard and soft time limits for server recovery.

i=andrew.perepechko
i=Hongchao.zhang

9 years agoRevert "b=20647 Ensure all needed pids exits before failover in test set 53"
Robert Read [Thu, 15 Apr 2010 23:06:54 +0000 (16:06 -0700)]
Revert "b=20647 Ensure all needed pids exits before failover in test set 53"

The "Cleanup wait_..._import_state function" part of this patch appears to cause
bug 22594.

This reverts commit 8f6d85eebc312b64d8e8a35b0be3ae137a50a45c.

9 years agob=22147 do not reset owner for lockless lock.
Eric Mei [Thu, 15 Apr 2010 05:39:44 +0000 (22:39 -0700)]
b=22147 do not reset owner for lockless lock.

r=wangdi
r=vitaly

9 years agob=22056 hash_add not under spinlock.
Vitaly Fertman [Thu, 15 Apr 2010 05:39:41 +0000 (22:39 -0700)]
b=22056 hash_add not under spinlock.

a refcount is added to the hash, move hash_add from under spinlock with no race against hash_destroy.

i=green
i=tappro

9 years agoPrepare for build 40. 1.10.0.40 v1_10_0_40
Robert Read [Sun, 11 Apr 2010 18:27:47 +0000 (11:27 -0700)]
Prepare for build 40.

9 years agob=22190 return error on write if rdonly mode is on, don't reply if umount
Mikhail Pershin [Fri, 9 Apr 2010 01:24:07 +0000 (18:24 -0700)]
b=22190 return error on write if rdonly mode is on, don't reply if umount

i=oleg

9 years agob=22359 fix t_id for the recovery thread
Alexander.Zarochentsev [Fri, 9 Apr 2010 01:24:01 +0000 (18:24 -0700)]
b=22359 fix t_id for the recovery thread

arbitrary uninialized value in the recovery thread t_id
was the reason of filter_iobuf_get()) ASSERTION(thread_id <
filter->fo_iobuf_count) failure.

i=tappro
i=andrew.perepechko

9 years agob=11063 set mtime to past under PW EOF extent lock (v6)
Vladimir Saveliev [Fri, 9 Apr 2010 01:24:01 +0000 (18:24 -0700)]
b=11063 set mtime to past under PW EOF extent lock (v6)

(according to HLD: https://bugzilla.lustre.org/attachment.cgi?id=15441).

new IO type (CIT_SETATTR) replaces CIT_TRUNC

Eariler patches
         https://bugzilla.lustre.org/attachment.cgi?id=7933
         https://bugzilla.lustre.org/attachment.cgi?id=7882
are reversed.

i=ericm
i=vitaly

9 years agob=11063 mtime tests
Vladimir Saveliev [Fri, 9 Apr 2010 01:24:00 +0000 (18:24 -0700)]
b=11063 mtime tests

few tests from 11063 and 21114

i=grev

9 years agob=22526 rename num_possible_cpus() to cfs_num_possible_cpus()
Robert Read [Thu, 8 Apr 2010 00:39:06 +0000 (17:39 -0700)]
b=22526 rename num_possible_cpus() to cfs_num_possible_cpus()

a=kenh@cmf.nrl.navy.mil
i=rread

9 years agob=19505 changelogs - distinguish between SETATTR and [acm]time changes
Nathan Rutman [Wed, 13 Jan 2010 22:21:18 +0000 (14:21 -0800)]
b=19505 changelogs - distinguish between SETATTR and [acm]time changes

i=breitz
i=LEIBOVICI Thomas <thomas.leibovici@cea.fr>

9 years agob=21966 can not mount mdt after umount
Rahul Deshmukh [Wed, 7 Apr 2010 17:18:18 +0000 (10:18 -0700)]
b=21966 can not mount mdt after umount

i=johann

9 years agofix typo in recovery-*-scale.sh script
Johann Lombardi [Wed, 7 Apr 2010 08:17:59 +0000 (10:17 +0200)]
fix typo in recovery-*-scale.sh script

9 years agob=22505 Initialize cui_tot_nregs for liblustre
Robert Read [Sat, 3 Apr 2010 02:01:53 +0000 (19:01 -0700)]
b=22505 Initialize cui_tot_nregs for liblustre

Change 910bf010 inadvertantly broke liblustre because cio->cui_tot_nregs
was not initilized, and ccc_io_update_iov was setting cui_nregs
incorrectly.

i=eric.mei
i=yong.fan

9 years agob=22312 sanity test 33c has problems parsing llobdstat when write_bytes is large
root [Tue, 6 Apr 2010 22:23:06 +0000 (15:23 -0700)]
b=22312 sanity test 33c has problems parsing llobdstat when write_bytes is large

i=robert.read@sun.com

9 years agob=21502 symlink compatibility between 1.6 and 2.0
Rahul Deshmukh [Mon, 5 Apr 2010 18:50:40 +0000 (11:50 -0700)]
b=21502 symlink compatibility between 1.6 and 2.0

Fixed the symlink compatibility problem between 1.6/1.8 and 2.0 related to short symlinks.

i=bzzz
i=andreas
i=pravin

9 years agob=20339 Remove LPSZ definition.
pravin [Mon, 5 Apr 2010 18:50:34 +0000 (11:50 -0700)]
b=20339 Remove LPSZ definition.

a=andreas.dilger
i=alexey.lyashkov
i=pravin

9 years agob=20562 Increasing super sequence width so that there will be lesser update to fld...
pravin [Mon, 5 Apr 2010 18:50:26 +0000 (11:50 -0700)]
b=20562 Increasing super sequence width so that there will be lesser update to fld table.

We can not increase LUSTRE_SEQ_META_WIDTH, meta sequence is allocated to client
and client can loose it in case of crash and there is no recovery mechanism.
MDT can not loose super sequence as it is stored on MDT disk.

i=andreas.dilger
i=tappro

9 years agob=21476 fix race between mdt_finish and __mds_lov_synchronize
pravin [Mon, 5 Apr 2010 18:49:56 +0000 (11:49 -0700)]
b=21476 fix race between mdt_finish and __mds_lov_synchronize

i=FanYong
i=Rahul

9 years agob=20647 Ensure all needed pids exits before failover in test set 53
Mikhail Pershin [Mon, 5 Apr 2010 18:49:50 +0000 (11:49 -0700)]
b=20647 Ensure all needed pids exits before failover in test set 53

Cleanup wait_..._import_state function.

i=rread
i=grev