Whamcloud - gitweb
Fan Yong [Mon, 20 Feb 2012 10:37:36 +0000 (18:37 +0800)]
LU-1070 agl: update lock state when AGL upcall
The AGL RPC sponsor may exits the cl_lock processing without
wait() called before related OSC lock upcall(). So when AGL
upcall(), it needs to update the cl_lock state according to
the enqueue result through the general cl_lock API wait().
Originally, it is done in "lov_lock_unuse()" against toplock.
But osc_lock upcall() is against sublock. So the sublock state
may be un-updated and inconsistent with the low layer osc_lock
state, then causes the sublock cannot be cached.
On the other hand, cl_lock::cll_descr::cld_enq_flags should not be
changed during unuse() for passing parameter, as the replacement,
it can be done through cl_lock::cll_flags.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: I4bc4c0ef0b93d5e0c2e59304158a35c57635ad70
Reviewed-on: http://review.whamcloud.com/2099
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Yu Jian [Wed, 7 Mar 2012 14:45:27 +0000 (22:45 +0800)]
LU-734 tests: add sub-tests into recovery-*-scale tests
This patch adds sub-tests into the recovery-*-scale tests
so that test results and logs could be gathered properly
and uploaded to Maloo.
The patch also does some cleanup works on the test scripts
and moves some common functions into test-framework.sh.
Signed-off-by: Yu Jian <yujian@whamcloud.com>
Change-Id: I2c23b641e62a56882171982b8a3157ba023ea36c
Reviewed-on: http://review.whamcloud.com/2265
Tested-by: Hudson
Reviewed-by: Cliff White <cliffw@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Mon, 12 Mar 2012 08:11:47 +0000 (16:11 +0800)]
LU-969 debug: reduce stack usage
1, libcfs_debug_vmsg2 to accept libcfs_debug_msg_data struture
to replace SUBSYSTEM, __FILE__, __FUNCTION__, __LINE__ and
cdls on the stack
2, CDEBUG, DEBUG_CAPA use static libcfs_debug_msg_data
3, remove the local variable in RETURN/GOTO/__CHECK_STACK
4, reduce stack in recovery thread by moving lu_env,
ptlrpc_thread to heap.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
Change-Id: I4a66d18d956ea4526ac85794d17305bde67f5634
Reviewed-on: http://review.whamcloud.com/2162
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Mon, 12 Mar 2012 17:17:19 +0000 (10:17 -0700)]
LU-1082 test: create test for mds-survey
Create a mds-survey.sh test with two test cases
1. test without stripe count
2. test with stripe count
Create test_224[a|b] for quick run on metadata survey
Notice, we remove all clients before testing
because the real client and echo_client share
the same client object memory
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: I453ac34d8b6019e109be5d89d482300d31571f93
Reviewed-on: http://review.whamcloud.com/2156
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Thu, 8 Mar 2012 07:29:09 +0000 (15:29 +0800)]
LU-1198 idl: move FID VER to DLM resource name[1]
Until Lustre 1.8.7/2.1.1 the FID version was packed into name[2].
However, this leaves very little room in the LDLM resource name
for other uses. The upcoming quota code needs to store another
FID into the LDLM resource to allow directory tree quotas, and
managed by the DLM.
The 32-bit VER, which is currently always 0, is moved into the high
bits of name[1] along with the 32-bit OID, to avoid consuming the
name[2] field. Since future use of the FID version (including
snapshots, pools, etc) will need changes on the client side anyway,
there will never be non-zero VER on an existing client.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: If1e500cfb277dfc25bc056bb0c5763e48e7dccdf
Reviewed-on: http://review.whamcloud.com/2271
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Thu, 8 Mar 2012 03:08:45 +0000 (11:08 +0800)]
LU-1190 test: Test failure on sanity, subtest test_51c.
Since LU-993 already add test_51ba for rmdir test, So remove test_51c.
Signed-off-by: yang sheng <ys@whamcloud.com>
Change-Id: Ib517f83964a877b687a332c7e1ff7a74c79836b6
Reviewed-on: http://review.whamcloud.com/2269
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Yu Jian [Mon, 5 Mar 2012 08:08:53 +0000 (16:08 +0800)]
LU-734 tests: save recovery-*-scale debug logs into $LOGDIR
This patch changes the specific debug log names of
recovery-*-scale tests and saves them into $LOGDIR.
Signed-off-by: Yu Jian <yujian@whamcloud.com>
Change-Id: Id1ad7dd3dde41bd49a53986945d15d80f91718e5
Reviewed-on: http://review.whamcloud.com/2261
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Cliff White <cliffw@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Fan Yong [Thu, 8 Mar 2012 02:47:07 +0000 (10:47 +0800)]
LU-1195 llite: error handling for statahead/AGL start failure
Statahead/AGL status will be checked when release ll_statahead_info.
But if some reasons cause statahead/AGL cannot be started, then the
status are undefined. So need proper error handling for such failure
cases.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: I4dec58862ad425c19e9c6284b7183b8bcaa449f6
Reviewed-on: http://review.whamcloud.com/2268
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Thu, 1 Mar 2012 00:32:23 +0000 (16:32 -0800)]
LU-1155 obdecho: Oops if target device can not be found.
Missing setting error value if target device can not be found,
which will cause Oops in some cases.
Signed-off-by: Wang Di <di.wang@whamcloud.com>
Change-Id: I5fc7cd376c8c7c276739692c7234ae142b4f6e23
Reviewed-on: http://review.whamcloud.com/2229
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Tue, 6 Mar 2012 07:54:03 +0000 (02:54 -0500)]
Master is now for upcoming 2.3.0 release
Updates changelog and lustre version.
New tag 2.2.50
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Change-Id: I8bbe086bbc8a579749eb7554b902dd83ca5670d0
Wally Wang [Wed, 22 Jun 2011 17:53:03 +0000 (10:53 -0700)]
LU-445 lnet: Send timestamps with LNet counters
We use a millisecond timer on a per-session basis. Each node then
sends the milliseconds since session start over with the LNet
counters. With 32bits and millisecond resolution, this should
allow ~49 days of session age before the timers would wrap.
This will change the wire attribute and not backward compatible.
Change-Id: I0dc6498d187ccc900b86d21b0460e1d419770033
Signed-off-by: Wally Wang <wang@cray.com>
Signed-off-by: Liang Zhen <liang@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1000
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: <bruce.korb@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Lai Siyao <laisiyao@whamcloud.com>
Niu Yawei [Fri, 24 Feb 2012 05:21:51 +0000 (21:21 -0800)]
LU-1128 ldlm: return -1 for server pool shrinker
For ldlm server pool shrinker, we just use it to decrease SLV,
but never reclaim any memory directly, so it should always return
-1 to inform the kernel to break the shrink loop.
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: I17f51ac84eb0b8c70b2cee9ac7eeca34647c1990
Reviewed-on: http://review.whamcloud.com/2184
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Fri, 20 Jan 2012 20:30:37 +0000 (12:30 -0800)]
LU-983 llite: align readahead to 1M after ra_max adjustment
1. Align the readahead pages only if ria_start != 0, otherwise
the readahead pages will be cut to zero.
2. Add test_101e to verify small reads for small size files.
Add test_101f to verify max_read_ahead_whole_mb.
3. Port 101c from b1_8
Signed-off-by: Wang Di <di.wang@whamcloud.com>
Change-Id: Ia22434a5d546aa1154f6067c7db84743c4ff5a96
Reviewed-on: http://review.whamcloud.com/1998
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Gearing [Mon, 27 Feb 2012 10:59:26 +0000 (10:59 +0000)]
LU-164 test: Update testframework to restart sessions at next suite
Update the test framework so that it can append to yaml files rather
than begin a new series and allow auster to begin at a selected
test_suite.
auster has a new switch '-S suite-name' which allows the caller to
start a test group part way through.
The idea behind all of this is that a test session that fails at a
test suite can be restarted at the next suite.
Signed-off-by: Chris Gearing <chris@whamcloud.com>
Change-Id: Ieaa987874d69fa8c33c3af92957e99607a8a7be7
Reviewed-on: http://review.whamcloud.com/414
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Robert Read <rread@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Jinshan Xiong [Fri, 2 Mar 2012 05:46:18 +0000 (21:46 -0800)]
LU-1163 llite: never try to invalidate a dirty page
When ll_page_mkwrite() detects page's covering lock is being canceled,
it should write the page back before invalidating it. Otherwise, it
will cause data corruption.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I345728daacf50edc154cd434dea392410f8cc782
Reviewed-on: http://review.whamcloud.com/2243
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Andreas Dilger [Wed, 25 Jan 2012 23:52:52 +0000 (16:52 -0700)]
LU-459 tests: quiet spurious console/test messages
Quiet a number of frequent and useless messages printed to the test
output and/or the console. These distract from understanding what
is happening in the test and provide little, if any, value.
The "acl" mount option on the client has been deprecated since 1.8.0,
since it was added in mid-1.6, and is driven by the MDS config only.
Use textual debug/subsystem flags instead of numeric flags. This has
been possible through all of 1.8, but is deprecated in 2.x so that
we can move to a more sophisticated debug mechanism in the future.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I81bed13c22465dc179e19f5a6714b732d68c500c
Reviewed-on: http://review.whamcloud.com/2015
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Wed, 22 Feb 2012 11:47:28 +0000 (04:47 -0700)]
LU-1145 test: add Test-Parameters tag for autotest
Allow Test-Parameters line to specify extra tests that should be run
to properly validate a change made to code or a regression test.
The Test-Parameters line can be quite long, depending on the number
of parameters specified. A single set of test parameters can be
split over multiple lines by escaping the linefeed, though the
individual lines should still be kept below the 70-character limit.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I65e1d41ad0f487df3753f25d9111ade9a0af08b1
Reviewed-on: http://review.whamcloud.com/2176
Reviewed-by: <bruce.korb@gmail.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Chris Gearing <chris@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alexey Lyashkov [Thu, 26 Jan 2012 12:39:12 +0000 (16:39 +0400)]
LU-1039 ptlrpc: handle bulk IO errors correctly.
don't panic on incorrect bulk transfer,
correctly handle a bulk request reorder.
LustreError: 12927:0:(client.c:1696:ptlrpc_check_set()) LBUG
Pid: 12927, comm: ptlrpcd-brw
Call Trace:
[<
ffffffffa083c865>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
[<
ffffffffa083cea5>] lbug_with_loc+0x75/0xe0 [libcfs]
[<
ffffffffa0e2edd6>] ptlrpc_check_set+0x1a66/0x1b90 [ptlrpc]
[<
ffffffffa0e61006>] ? ptlrpcd_check+0x46/0x290 [ptlrpc]
[<
ffffffffa0e61006>] ? ptlrpcd_check+0x46/0x290 [ptlrpc]
[<
ffffffffa0e611b8>] ptlrpcd_check+0x1f8/0x290 [ptlrpc]
[<
ffffffffa0e6157b>] ptlrpcd+0x32b/0x3b0 [ptlrpc]
[<
ffffffff81056388>] ? finish_task_switch+0x48/0x110
[<
ffffffff8105f500>] ? default_wake_function+0x0/0x20
[<
ffffffffa0e61250>] ? ptlrpcd+0x0/0x3b0 [ptlrpc]
[<
ffffffff8100c2ca>] child_rip+0xa/0x20
[<
ffffffff81500b70>] ? _spin_unlock_irq+0x30/0x40
[<
ffffffff8100bc10>] ? restore_args+0x0/0x30
[<
ffffffffa0e61250>] ? ptlrpcd+0x0/0x3b0 [ptlrpc]
[<
ffffffff8100c2c0>] ? child_rip+0x0/0x20
Xyratex-bug-id: MRP-303
Change-Id: Ibb19c33e92dbecf5d029c7e6c567f65fb764f444
Signed-off-by: Alexey Lyashkov <alexey_lyashkov@xyratex.com>
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/2023
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
James Simmons [Thu, 16 Feb 2012 14:39:29 +0000 (09:39 -0500)]
LU-335 ldiskfs: SuSE 11 SP1 update for Lustre 2.X
The ext4 code difference between SuSE and RHEL6 is very small.
Because of to this ldiskfs support for SuSE servers is mostly
based the RHEL6 ldisk work. Recently Lustre support migrated
to RHEL6.2 support which required the SuSE port to be updated.
This patch provides this update which works for the SuSE
2.6.32.45 kernel as well as remove some no longer needed
ldsiskfs patches for older SuSE releases.
Signed-off-by: James Simmons <uja.ornl@gmail.com>
Change-Id: Ia6763e8a35698018bc84c95564ec9ea63d5f2c33
Reviewed-on: http://review.whamcloud.com/2138
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 2 Mar 2012 02:30:30 +0000 (19:30 -0700)]
LU-1146 build: batch update copyright messages
Update copyright messages for all files modified by Whamcloud.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I9654b780c6e4726d369e2a893f7e210d4078819b
Reviewed-on: http://review.whamcloud.com/2242
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Tue, 24 Jan 2012 08:08:46 +0000 (00:08 -0800)]
LU-991 llite: cleanup md_readpage interface
Use md_op_data to transfer the parameters of readpage on client
metadata stack.
Signed-off-by: Wang Di <di.wang@whamcloud.com>
Change-Id: I103a555094e678711aa2f2e5d73ec848a452aed0
Reviewed-on: http://review.whamcloud.com/2001
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Sat, 25 Feb 2012 01:17:55 +0000 (18:17 -0700)]
LU-1019 ptlrpc: fix ptlrpcd transfer message
Fix the debug message when transferring requests between ptlrpcd
threads to correctly identify the source and target thread index.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I330deb92a2ae0c9134dbf24ee6074e0cf9803d31
Reviewed-on: http://review.whamcloud.com/2195
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Wed, 21 Dec 2011 04:33:12 +0000 (20:33 -0800)]
LU-949 test: improve sanity test_42e
The sanity test_42e should check that the RPCs with less than 16 pages
are not more than 15%, instead of checking the 16-page RPCs are more
than 85%, because the pages could be merged into even larger(> 16 pages)
RPCs.
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: I603f9d278b55cd850d66466050745ecb80e2ff02
Reviewed-on: http://review.whamcloud.com/1897
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Andreas Dilger [Wed, 26 Oct 2011 07:37:59 +0000 (01:37 -0600)]
LU-433 ldiskfs: fix up patch comment for 2.2
The LU-433 patch wasn't landed until 2.2, so this patch is at least
needed to make 2.1->2.2 transition easier.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ia213ae92a8fe21f5153e22b588ff66c8f67439f2
Reviewed-on: http://review.whamcloud.com/1604
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andrew Perepechko [Fri, 24 Feb 2012 11:27:51 +0000 (15:27 +0400)]
LU-1133 tests: sanity-quota 33 fixes
sanity-quota test 33 must wait for removal completion
before verifying quota usage correctness
Signed-off-by: Andrew Perepechko <andrew_perepechko@xyratex.com>
Change-Id: Ia9994a41b734a19c0711c201c3b54a38480c33fd
Xyratex-bug-id: MRP-378
Reviewed-by: Alexander Zarochentsev <azarochentsev@xyratex.com>
Reviewed-on: http://review.whamcloud.com/2186
Tested-by: Hudson
Reviewed-by: Alexander Zarochentsev <alexander_zarochentsev@xyratex.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Denis Kondratenko [Fri, 10 Feb 2012 10:54:45 +0000 (12:54 +0200)]
LU-958 tests: debug_mb set incorrectly for smp or vm
For cpus with number of cores or for some VMs,
number of possible CPUs in the system could
be greater than number of cpu reported by getconf.
Added check for maximum debug buffer size.
Xyratex-bug-id: MRP-219 incorrect settings for debug_mb
Reviewed-by: Andrew Perepechko <Andrew_Perepechko@xyratex.com>
Reviewed-by: Alexey Lyashko <Alexey_Lyashko@xyratex.com>
Signed-off-by: Denis Kondratenko <Denis_Kondratenko@xyratex.com>
Change-Id: I7001af7b1c88d5be056734d7d73a0263cca01627
Reviewed-on: http://review.whamcloud.com/1912
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andrew Perepechko <andrew_perepechko@xyratex.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Lai Siyao [Thu, 23 Feb 2012 08:23:25 +0000 (16:23 +0800)]
LU-1014 mountconf: MGS should process parameter config
MGS doesn't have llog config of its own, but it should process
<profile>-params config which is global parameters for the whole
system.
Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
Change-Id: I62147dfff17f28c537d75652b49ed3582b0fa64c
Reviewed-on: http://review.whamcloud.com/2139
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Mike Pershin <tappro@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Tue, 28 Feb 2012 21:15:35 +0000 (14:15 -0700)]
LU-482 test: sync new fs before first replay test
LVM lacks barrier support for kernel < 2.6.33, so we need to send an
explicit sync and sleep to avoid the replay_barrier call in the first
test being run from discarding blocks from just-formatted filesystems.
Otherwise this incorrectly causes just-written metadata to be lost by
the block layer by the dev-read-only failover test code.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I69fc64298603ad75a4fcda09edde5bc7ded00c1e
Reviewed-on: http://review.whamcloud.com/2223
Reviewed-by: Lai Siyao <laisiyao@whamcloud.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Liang Zhen [Tue, 21 Feb 2012 04:40:25 +0000 (12:40 +0800)]
LU-78 o2iblnd: kiblnd_check_conns can deadlock
kiblnd_check_conns() called kiblnd_check_sends() with hold of global
rwlock, it's wrong because kiblnd_check_sends() could do many things:
- call lnet_finalize() which is not safe with hold of spinlock
- call kiblnd_close_conn() which requires to write_lock the same
global lock
- kiblnd_check_sends() might need to allocate NOOP message
It can be fixed by moving call of kiblnd_check_sends out from spinlock
This patch is from the fix of Bug 20288, with some small changes.
Signed-off-by: Liang Zhen <liang@whamcloud.com>
Change-Id: Icc9fedc70ecb25b0c41ebaf6d80c971f8281c9c6
Reviewed-on: http://review.whamcloud.com/2166
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Doug Oucharek <doug@whamcloud.com>
Reviewed-by: Lai Siyao <laisiyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Bobi Jam [Wed, 22 Feb 2012 06:30:41 +0000 (14:30 +0800)]
LU-1125 recovery: initial recovery thread's watchdog
Recovery thread does not have watchdog attached, correctly initialize
it.
Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: I6993c39bbf18f47e9ccd965a5d2ba1919cfb7736
Reviewed-on: http://review.whamcloud.com/2174
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Mike Pershin <tappro@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jinshan Xiong [Thu, 23 Feb 2012 19:57:54 +0000 (11:57 -0800)]
LU-1109 llite: do splice read stripe by stripe
If nfsd is reading an across stripe buffer, and if the first stripe
happens to be 64KB(PIPE_BUFFERS*PAGE_SIZE), then first read will
occupy all pipe buffers and this makes nfsd stuck if it reads the
next stripe immediately.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I13cb54b37f738ee3c081dff1929630ea523b77fd
Reviewed-on: http://review.whamcloud.com/2182
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Johann Lombardi [Wed, 13 Jul 2011 23:05:32 +0000 (01:05 +0200)]
LU-1132 protocol: reserve connect flags for new features
This patch reserves connect flags for several outstanding features:
* OBD_CONNECT_GRANT_PARAM for grant additional parameters to consume
grant more intelligently on the client side.
* OBD_CONNECT_JOBSTATS to pass a job id in ptlrpc_body.
* OBD_CONNECT_EINPROGRESS for clients that can deal with EINPROGRESS
on write.
This patch also reserves fields for the per-file data_version patch
from LU-827.
Change-Id: Icb55435bca443b4c2106e7eec9b368f94d1be08c
Signed-off-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1090
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Wed, 15 Feb 2012 05:57:50 +0000 (13:57 +0800)]
LU-1069 mdc: Avoid doing permission check again.
inode_setattr() replaced by simple_setattr(). The latter
will invoke inode_change_ok(). It causes a -EPERM error when
we want to update inode timestamps.
Signed-off-by: yang sheng <ys@whamcloud.com>
Change-Id: I29c58d7027df1bf94db58df82e5214fd12ea5c87
Reviewed-on: http://review.whamcloud.com/2145
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Bobi Jam [Tue, 21 Feb 2012 01:23:11 +0000 (09:23 +0800)]
LU-1098 debug: lower debug message level
File info read and unlink race is normal, we'd lower the debug message
level since a lot of unnecessary unmasked messages will be generated
if mdt_object_find() cannot find those deleted objects.
Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: I7630e6a1456ffb435c8e67cc626bf38547b840d0
Reviewed-on: http://review.whamcloud.com/2165
Tested-by: Hudson
Reviewed-by: Christopher J. Morrone <chris.morrone.llnl@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Nikitas Angelinas [Tue, 14 Feb 2012 09:07:54 +0000 (09:07 +0000)]
LU-719 ldiskfs: fix error message rate-limiting logic in kmmpd
Current logic would print an error message only once, and then
'failed_writes' would stay at 1. Rework the loop to increment
'failed_writes' and print the error message every
s_mmp_update_interval * 60 seconds, as intended according to the
comment.
Signed-off-by: Nikitas Angelinas <nikitas_angelinas@xyratex.com>
Change-Id: I0c8d90927bd618542026ab5fc70cf12b85898dc3
Reviewed-by: Andrew Perepechko <andrew_perepechko@xyratex.com>
Xyratex-bug-id: MRP-222
Reviewed-on: http://review.whamcloud.com/1420
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Andreas Dilger [Tue, 24 Jan 2012 20:22:08 +0000 (13:22 -0700)]
LU-1028 tests: re-enable sanityn.sh test_16 (fsx)
The sanityn.sh version of fsx in test_16 provides useful test
coverage because it is running an IO load that is doing unaligned
file access from multiple clients (mountpoints) and also verifying
data correctness. This has exposed several bugs in the orion branch,
and should also be enabled by default for all landings on master.
A longer version of the test is run in sanity-benchmark.sh for
better test coverage. Check that fsx completes successfully,
otherwise it may fail without reporting an error to the test
framework.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I26dd91cacc2596f6e84f4c32362befa1e339cab0
Reviewed-on: http://review.whamcloud.com/2007
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Sun, 19 Feb 2012 13:14:00 +0000 (21:14 +0800)]
LU-1120 kernel: kernel update [RHEL6.2 2.6.32-220.4.2]
Update RHEL6.2 kernel to 2.6.32-220.4.2.
Signed-off-by: yang sheng <ys@whamcloud.com>
Change-Id: If4d63c32776452fcb615b8ed196cc3f037b07dfa
Reviewed-on: http://review.whamcloud.com/2163
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Fri, 4 Nov 2011 06:00:05 +0000 (23:00 -0700)]
LU-819 utils: Fix lfs getstripe -M
1. Fix lfs getstripe -M problem: some typo in llapi_file_fget_mdtidx.
2. Add lfs find -mdt and lfs mdts.
3. Add sanity test to check lfs getstripe -M and lfs find -mdt.
Signed-off-by: wangdi <di.wang@whamcloud.com>
Change-Id: I3b327d33c49009149dbed23728416efecb67691c
Reviewed-on: http://review.whamcloud.com/1646
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Mon, 6 Feb 2012 06:02:07 +0000 (22:02 -0800)]
LU-1051 osd: reserve less credits for llog unlink
Reserve less credits for the llog write and catalog header udpate.
Since we can't get the llog inode in osd declare function, we use
a temporary solution: hack the 'size' parameter to indicate a llog
write or catalog header udpate.
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: Ic5df00a878cdbf64a44b22b684e2c6b2dff83b07
Reviewed-on: http://review.whamcloud.com/2100
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Oleg Drokin [Sat, 18 Feb 2012 05:44:13 +0000 (00:44 -0500)]
Tag 2.1.56
Change-Id: I87d5cf42ad646a393fde8743b7153251820c3719
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Christopher J. Morrone [Sat, 11 Feb 2012 01:34:32 +0000 (17:34 -0800)]
LU-1084 ptlrpc: Change CWARNs to CDEBUGs
These messages should not appear on the console. A sysadmin
will have no idea what to make of most of them.
Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
Change-Id: I58bbc1eca9f5082d08cee6c5a95793c0f64ef370
Reviewed-on: http://review.whamcloud.com/2135
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Wed, 7 Dec 2011 00:38:54 +0000 (16:38 -0800)]
LU-633 iokit: mds-survey script for MD echo client test
Create a mds-survey script to run echo client for MDS
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: I425438aa294d8d4360c28f686296ff7f26f847ea
Reviewed-on: http://review.whamcloud.com/1969
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: wangdi <di.wang@whamcloud.com>
yangsheng [Tue, 31 Jan 2012 01:25:07 +0000 (09:25 +0800)]
LU-1054 kernel: Kernel update [RHEL6.2 2.6.32-220.4.1.el6]
Update RHEL6.2 kernel to 2.6.32-220.4.1.el6.
Signed-off-by: yang sheng <ys@whamcloud.com>
Change-Id: Ife086ef0a8c521373434b02ff4e49dab85b9627c
Reviewed-on: http://review.whamcloud.com/2058
Tested-by: Hudson
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shuichi Ihara [Mon, 13 Feb 2012 16:47:38 +0000 (01:47 +0900)]
LU-1050 o2iblnd: fix checking order of rdma_create_id() argument
Replace rdma_create_id() with rdma_destroy_id() in
openib gen2 test and four argument check moves to
the back of openib test.
Signed-off-by: Shuichi Ihara <sihara@ddn.com>
Change-Id: I0782183f15f58647291518a4222610601083c369
Reviewed-on: http://review.whamcloud.com/2097
Tested-by: Hudson
Reviewed-by: Brian J. Murrell <brian@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Liang Zhen <liang@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Jinshan Xiong [Sat, 28 Jan 2012 22:43:47 +0000 (14:43 -0800)]
LU-1028 osc: fix grant checking on the osc side
If there is no available grant on the client side, but exists dirty
pages, we should issue BRW rpc and wait for more grants.
Another changes is about lost_grant: if dirty pages were truncated,
grants will be returned to client, instead of the target having
allocated them.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I50791554807dddfafafcd2b8d228958ebfa3ac16
Reviewed-on: http://review.whamcloud.com/2037
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: wangdi <di.wang@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Mon, 6 Feb 2012 23:02:56 +0000 (15:02 -0800)]
LU-1075 tests: auster detect directory as script name
Auster should use -f or detect the existing of
script name rather use -e which will treat directory
same as script
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: Idba7b5a25cd48a513e9db885fb7399deff62827f
Reviewed-on: http://review.whamcloud.com/2104
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Thu, 4 Aug 2011 03:36:23 +0000 (11:36 +0800)]
LU-427 test: Test failure on test suite lfsck
- Reset $MDSDB & $OSTDB in generate_db(). Else they will
stale if user redefine $SHARED_DIRECTORY.
- Add a function check_shared_dir() to ensure
$SHARED_DIRECTORY is shared among tests nodes.
Change-Id: Idf2a3d75e46c4cf768419adfea627511c24c495c
Signed-off-by: Yang Sheng <ys@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1180
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Fan Yong [Tue, 31 Jan 2012 07:26:31 +0000 (15:26 +0800)]
LU-883 ldiskfs: drop unused diagnostic information
The debug informatin was introduced for diagnosing readonly
issues for bug 16680. It is unused any longer and makes the
console output some noisy. So drop.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: If950654d6517f317809bb0de1446e2377a524c77
Reviewed-on: http://review.whamcloud.com/2068
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mikhail Pershin [Mon, 9 Jan 2012 13:46:10 +0000 (17:46 +0400)]
LU-971 mdt: fix open resent issue with last_xid
req_xid_is_last() compares xid with one in client data,
but in master it is not updated in case of open without create.
Patch updates export lcd with last xid, transno and rc for open
requests.
Signed-off-by: Mikhail Pershin <tappro@whamcloud.com>
Change-Id: I9e769fecc1e93c7a556d4b9eeadc80a70369e901
Reviewed-on: http://review.whamcloud.com/1930
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Reviewed-by: Li Wei <liwei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Tue, 31 Jan 2012 00:39:43 +0000 (16:39 -0800)]
LU-1049 script: lc_net doesn't parse output correctly
Port the second fix from bz=23234
The output of ping contains two ip addresses. The regex
matching picks up both addresses. The fix is to add
-m 1 to stop at the first match
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: Ie145e710844d80587bf6826822c8db941a4a3bae
Reviewed-on: http://review.whamcloud.com/2057
Reviewed-by: Cliff White <cliffw@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Wed, 9 Nov 2011 20:06:30 +0000 (13:06 -0700)]
LU-601 mdd: cleanup error messages and code style
While looking at LU-601, I added comment blocks for several of
the functions in the PENDING directory cleanup path, along with
fixes to the code style and error messages. No functional changes
should be introduced by this code change.
The error message that was formerly printed by the 1.8 MDS in
mds_unlink_orphan() and checked by replay-single.sh test_37 no
longer exists in the 2.x MDS code. Re-add a message if there
is an error cleaning up directory orphans, and change test_37
to match the resulting error message (similar to the 1.8 code).
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ic97843e49643041513f3b2fc6a40f2be63ffddca
Reviewed-on: http://review.whamcloud.com/1679
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Brian J. Murrell [Fri, 10 Feb 2012 20:27:04 +0000 (15:27 -0500)]
LU-1091 fix the tag matching code in configure
Signed-off-by: Brian J. Murrell <brian@whamcloud.com>
Change-Id: I24dd4d0929364a2012afe725586626328ea584c0
Reviewed-on: http://review.whamcloud.com/2131
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Christopher J. Morrone [Fri, 10 Feb 2012 23:24:06 +0000 (15:24 -0800)]
LU-1095 mgs: remove message from console
There is no good reason for a sysadmin to see this message
on the console. Most of the time this will be a fluke
due to the vagarities of lnet networks (server decides
client is disconnected, but client doesn't know that yet,
messages arriving out of order, etc.).
Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
Change-Id: I0c18734f82a9c89a5e940ce4e2c602614e89ce26
Reviewed-on: http://review.whamcloud.com/2133
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Tue, 6 Dec 2011 21:24:59 +0000 (05:24 +0800)]
LU-709 build: LC_XATTR_ACL & LC_POSIX_ACL_XATTR_H cleanup
xattr_acl.h removed since 2.6.13 and posix_acl_xattr.h
exist since 2.6.11. So just use posix_acl_xattr.h for now.
And removed relate codes.
Change-Id: Ibefd4db87fa79ec9ee6365d48c83abd0dda0e063
Signed-off-by: yangsheng <ys@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1916
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Brian J. Murrell [Tue, 7 Feb 2012 22:15:53 +0000 (17:15 -0500)]
LU-1077 distribute missing file in build/autoconf
In landing
0d5cf708f60c1ed9233fd12f4326851ab412ef6a for ticket
LU-723 a defect was introduced where
build/autoconf/lustre-build-ldiskfs.m4 is missing from the
distribution tarball.
Signed-off-by: Brian J. Murrell <brian@whamcloud.com>
Change-Id: I34a1d7ad20229fb531eb4e76182647e708b300d4
Reviewed-on: http://review.whamcloud.com/2115
Tested-by: Hudson
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Prakash Surya <surya1@llnl.gov>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Tue, 7 Feb 2012 15:53:09 +0000 (07:53 -0800)]
LU-1029 build: build ofed 1.5.4 with qlgc_vnic failed
Ofed failed to include proper header files to support
qlgc_vnic. Since we are not using ib on vnic in the lab,
we'll remove it
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: Ibce110b44940fda5f13abab890f3e9176f55b9d2
Reviewed-on: http://review.whamcloud.com/2110
Tested-by: Hudson
Reviewed-by: Brian J. Murrell <brian@whamcloud.com>
Reviewed-by: Michael MacDonald <mjmac@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Thu, 12 Jan 2012 15:00:06 +0000 (10:00 -0500)]
LU-915 tests: MKFSOPTS fix for test suite
While testing LU-80 I had to provide my own MKFSOPT string
for the mkfsoptions in local.sh. In the conf-sanity test
the MKFSOPT provided were totally replaced by a new MKFSOPT
that contained only lazy_itable_init. This prevented me from
running any conf-sanity test bacause I couldn't format the
MDS. This patch instead appends lazy_itable_init to the
mkfsoptions string if it already exist.
Change-Id: If30d2a692431dcdb3631fc9c1e77da5ff2792881
Signed-off-by: James Simmons <uja.ornl@gmail.com>
Reviewed-on: http://review.whamcloud.com/1830
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Cliff White <cliffw@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Wed, 15 Feb 2012 18:41:03 +0000 (13:41 -0500)]
Revert "LU-1013 obdclass: lu_object_find miss to unlink object from LRU"
Reverting because in fact it overlaps with LU-1017
This reverts commit
b9ccecd1453c5c76fe135048c39f149c241650c6.
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Mikhail Pershin [Wed, 15 Feb 2012 08:39:53 +0000 (12:39 +0400)]
LU-966 mdd: revert commit and use original patch
Use original patch which replaces LASSERT with CERROR in MDD,
because checking object exists in MDT ruins VBR recovery.
It happens too early but VBR should check that too before reply
to the client.
Signed-off-by: Mikhail Pershin <tappro@whamcloud.com>
Change-Id: If31586c9d1cc86c4c8059a7227c231acdc82a0f1
Reviewed-on: http://review.whamcloud.com/2148
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Tested-by: Yu Jian <yujian@whamcloud.com>
Jinshan Xiong [Wed, 25 Jan 2012 06:15:59 +0000 (22:15 -0800)]
LU-889 recovery: rework extend_recovery_timer()
Commit
8da33c6cc5192303fcd18f45892e1f115004e662 is buggy where
the time used to rearm the timer is wrong. In this patch, I
will reset all changes and use the same scheme as 2.1 release.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: Icd887687262644e1a1cfffd6f34eacc5bc418f84
Reviewed-on: http://review.whamcloud.com/1722
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Mike Pershin <tappro@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jinshan Xiong [Tue, 31 Jan 2012 21:30:21 +0000 (13:30 -0800)]
LU-1059 clio: to not try to discard freeing pages
This is a bug imported in LU-948. We should check if we have owned
the page successfully before trying to discard it.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I30631be98f1fcc1b98abe727c8c6984b918bfffd
Reviewed-on: http://review.whamcloud.com/2073
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Thu, 5 Jan 2012 21:28:51 +0000 (05:28 +0800)]
LU-969 debug: reduce stack usage
in ptlrpc_main, move struct lu_env to heap to save stack
in mdt_open.c, declare functions non-static to prevent inlining
Change-Id: I73c703c79b0539dc40036a5f9b131507adaff662
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/2070
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Andreas Dilger [Fri, 27 Jan 2012 21:12:06 +0000 (14:12 -0700)]
LU-1042 obdfilter: initialize new object timestamp
Initialize newly-created objects with a/m/ctime = 0, so that however
clients modify the timestamp it will be updated the first time. The
fsfilt_ext3_setattr() code will special case ctime = 0 for new files
and reset them to the ext4 s_mkfs_time stored in the superblock.
In LU-221 (commit
414251797ed178eec5d431e1f5aa4a889d2b159f) the
timestamps were initialized to (INT_MIN + 24 * 3600) (1901-12-14),
but the use of unsigned on-disk timestamps caused these negative
numbers to be transformed into large positive numbers (debugfs
reports them to be "Tue Jan 19 20:14:08 2038"). Add a fix so that
files with this exact timestamp have the timestamp reset to 0.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I22f2cd443402c0c524bafaf37c80cad7b6d00c1e
Reviewed-on: http://review.whamcloud.com/2036
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 3 Feb 2012 22:27:40 +0000 (15:27 -0700)]
LU-797 tests: skip ost-pools.sh 23b when SLOW=no
Running test_23b can take over 30 minutes, and should not be run on a
large LUN inside a VM. Print out some of the constants that determine
whether the test is run or not, for debugging purposes.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I7491002276e36994be4baf30a731576c8ced7530
Reviewed-on: http://review.whamcloud.com/2091
Tested-by: Hudson
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Niu Yawei [Tue, 14 Feb 2012 02:21:57 +0000 (18:21 -0800)]
LU-963 build: update Lustre ChangeLog
Update the recommended e2fsprogs version to 1.41.90.wc4 for 2.2.
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: I5516a0c228cca4aa36c68cc299d0127c9c6f5325
Reviewed-on: http://review.whamcloud.com/2140
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Fri, 3 Feb 2012 07:44:24 +0000 (23:44 -0800)]
LU-1055 remove improper assert in osd_declare_object_destroy
After recovery, orphan cleanup could race with the client unlink on
close, so the LASSERT on lu_object_is_dying() should be moved from
declare stage into the real destroy function.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: Ibc4ad0613d907ab98b653d5a64ce54e1f018dcb2
Reviewed-on: http://review.whamcloud.com/2083
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Mon, 23 Jan 2012 22:17:55 +0000 (15:17 -0700)]
LU-506 tests: modprobe exportfs for fsfilt_ldiskfs
When trying to load fsfilt_ldiskfs.ko from the build tree, it fails
to load due to the missing exportfs_decode_fh() symbol in RHEL6
kernels. Force loading of the exportfs.ko module if ldiskfs is used.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Iecb13155b69ffc3a227e3df770ca590277cd500c
Reviewed-on: http://review.whamcloud.com/2014
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Niu Yawei [Thu, 2 Feb 2012 14:32:02 +0000 (06:32 -0800)]
LU-1051 osd: improve quota credits calculation
Instead of reserving constant LDISKFS_QUOTA_INIT_BLOCKS for all kinds
of transactions, this patch tracks how many/which uid/gids are
affected in each declare function, then calculates the total quota
credits according to the affected uid/gids in osd_trans_start().
This patch also includes following credits improvement:
- For object creation, reserve less credits for the '.' and '..'
insertion, and don't reserve for file creation;
- Only reserve the credits for oi insertion for normal fids;
- Limit the credits to j_max_transaction_buffers in the
osd_trans_start();
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: Ifbcc26b670c4965f22cc96ae7db16154927f32a4
Reviewed-on: http://review.whamcloud.com/2082
Tested-by: Hudson
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Bobi Jam [Fri, 13 Jan 2012 05:46:07 +0000 (13:46 +0800)]
LU-985 lprocfs: verify user buffer access
In lprocfs_xxx_evict_client(), need verify user's buffer when access
it.
Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: I702e22f8d432edce200c6d91a0af8a1eac792008
Reviewed-on: http://review.whamcloud.com/1961
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Liang Zhen <liang@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Fan Yong [Thu, 2 Feb 2012 11:17:41 +0000 (19:17 +0800)]
LU-1061 agl: cl_locks_prune() waits for the last user
The AGL sponsor holds user reference count on the cl_lock before
triggering AGL RPC. The user reference count on the cl_lock will
be released by AGL RPC reply upcall. Such AGL mechanism conflict
with cl_locks_prune(), which requires no lock is in active using
when the last iput() called.
So the cl_locks_prune() should wait for the last user reference
count to be released by the enqueue reply upcall.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: I8998c0a247448b1b6c6e99995c9d956b1666279b
Reviewed-on: http://review.whamcloud.com/2079
Tested-by: Hudson
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Tue, 31 Jan 2012 06:06:47 +0000 (22:06 -0800)]
LU-1017 handle -EAGAIN properly in lu_object_find_try()
htable_lookup() could return -EAGAIN for dying object, we should
handle it properly in lu_object_find_try().
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: I616928cb983d4f64a64f25b7d296b6c9bd18e4ea
Reviewed-on: http://review.whamcloud.com/2066
Tested-by: Hudson
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jinshan Xiong [Wed, 25 Jan 2012 19:27:55 +0000 (11:27 -0800)]
LU-874 ptlrpc: handle in-flight hqreq correctly
If there are in-flight requests pending, we shouldn't timeout the
covering dlm locks; neither should we remove the requests from export
exp_hp_rpcs list until the requests are handled.
In this patch, the following things are improved:
1. leave IO rpcs in export's hp list until they are handled;
2. using interval tree to find rpc overlapped locks;
3. refresh the lock again after IO rpcs are finished to leave a time
window for clients to cancel covering dlm locks;
4. rework repbody in ost_handler.c so as to not modify original obdo
5. cleanup the code.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I33e2d113d7929a56389741c06dffb5efb6bf28a3
Reviewed-on: http://review.whamcloud.com/1918
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: <alexander_boyko@xyratex.com>
Jinshan Xiong [Tue, 10 Jan 2012 07:34:06 +0000 (23:34 -0800)]
LU-874 osc: prioritize writeback pages
When a lock is being canceled, we should prioritize those covering
pages which have already been submitted by page writeback daemon;
otherwise, this client may be evicted because there is no active IO
for that lock for a long time.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: If14eff6361f55d2b2eeb2db7146789dda4c32060
Reviewed-on: http://review.whamcloud.com/1938
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Sat, 4 Feb 2012 08:12:48 +0000 (00:12 -0800)]
LU-1067 obdecho: Recheck client env ctx for echo md client.
During echo md test, if there are real clients being mounted
at the same time, the cl env it gotten from the cache might
contain the wrong context, so we need recheck the ctx and
refill the env forcedly.
Signed-off-by: Wang Di <di.wang@whamcloud.com>
Change-Id: Iddbceee80966b5ca9284c886731386a97d089d53
Reviewed-on: http://review.whamcloud.com/2092
Tested-by: Hudson
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alexander Zarochentsev [Sat, 11 Feb 2012 07:19:31 +0000 (11:19 +0400)]
LU-1073 llite: no CERROR on unknown flock type
Xyratex-bug-id: MRP-215
Signed-off-by: Alexander Zarochentsev <alexander_zarochentsev@xyratex.com>
Change-Id: I435c3282bc87750f6ebd19e618a1ee8e229834d2
Reviewed-on: http://review.whamcloud.com/2109
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Fan Yong [Sat, 11 Feb 2012 01:26:29 +0000 (09:26 +0800)]
LU-1013 obdclass: lu_object_find miss to unlink object from LRU
There is race condition between lu_object_find and lu_object_put.
For the case of two threads trying to find some object with the
same FID concurrently, and the object is not allocated in memory
yet, if the first thread adds the object into LRU list before the
second thread re-searching object hash table for inserting the
new object created by itself, then the second thread will find
the object created by the first thread. Under such case, the
second thread should unlink the object from LRU.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: Iadec96c27d5285f8b859b0060a6f611e87585789
Reviewed-on: http://review.whamcloud.com/2134
Tested-by: Hudson
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Thu, 26 Jan 2012 10:06:30 +0000 (03:06 -0700)]
LU-687 ldiskfs: resolve section dynlocks mismatch
Fix __init/__exit section mismatch.
WARNING: lustre-2.1.0/ldiskfs/ldiskfs/ldiskfs.o(.init.text+0x1bc):
Section mismatch in reference from the function init_module() to the
function .exit.text:dynlock_cache_exit(). An __init init_module()
function references a function __exit dynlock_cache_exit().
This is often seen when error handling in the init function uses
functionality in the exit path. The fix is often to remove the
__exit annotation of dynlock_cache_exit() so it may be used outside
an exit section.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ie0bdde7f78c18bca1127151175cf56bfa6ad500c
Reviewed-on: http://review.whamcloud.com/2019
Tested-by: Hudson
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Fan Yong [Tue, 31 Jan 2012 05:51:41 +0000 (13:51 +0800)]
LU-1046 ldlm: hold ldlm_cb_set_arg until the last user exit
There is race condition between ldlm_cb_interpret() and
ldlm_run_ast_work() when test/release "ldlm_cb_set_arg":
ldlm_run_ast_work() can exit before ldlm_cb_interpret()
processing "ldlm_cb_set_arg::waitq", then causes invalid
memory accessing.
Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
Change-Id: Idfb3f41739e02a79fe4310ecbc7bb842e75a82d8
Reviewed-on: http://review.whamcloud.com/2065
Tested-by: Hudson
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Mon, 16 Jan 2012 06:52:12 +0000 (22:52 -0800)]
LU-993 osd: code cleanup for directory nlink count
- LDISKFS_LINK_MAX is ldiskfs unique, we should not handle it in the mdd
layer. It's moved down to the osd layer in this patch.
- Remove the declared operation count assertion in OSD_EXEC_OP(), since
the undo operations are not declared, if there some operation failed,
the undo operation will trigger the assertion unnecessarily.
- Restore the sub-directory count to 70000 for the sanity test_51b, and
add test_51ba for testing unlink > LDISKFS_LINK_MAX sub-directories.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: Id02e7d2ac3a7664a55566b1de783f0a73162339b
Reviewed-on: http://review.whamcloud.com/1971
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Andreas Dilger [Thu, 15 Dec 2011 08:24:33 +0000 (01:24 -0700)]
LU-928 llite: fix comment for lustre_client_ocd
ll_ocd_update() was renamed to cl_ocd_update() before 2.0.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I834935d2dc9ef0edebbd5badb4bc54283a325730
Reviewed-on: http://review.whamcloud.com/1868
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Sun, 29 Jan 2012 07:20:45 +0000 (00:20 -0700)]
LU-1053 build: allow deleted/chmod in commit-msg
Allow files that were deleted or chmod when verifying a commit
message that contains a diff generated with "commit -v". The
checks added for excluding the diff were previously too strict.
Add a test case for each of these situations, as well as a case
for a commit message that contains "diff" at the start of a line
in the middle of a commit message.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I74fe98682342e445da7fe64fd42e91df4564500c
Reviewed-on: http://review.whamcloud.com/2053
Reviewed-by: Bruce Korb <bruce_korb@xyratex.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alexander.Boyko [Fri, 3 Feb 2012 07:20:01 +0000 (11:20 +0400)]
LU-1025 checksum: add final bit inversion for crc32c
The linux kernel implementations of crc32c perform final bit
inversion after loop calculation of checksum.
Signed-off-by: Alexander Boyko <alexander_boyko@xyratex.com>
Change-Id: I5fa6af60c51f6f86f394f3cc71aa2672be614f7b
Reviewed-on: http://review.whamcloud.com/2018
Tested-by: Hudson
Reviewed-by: Shuichi Ihara <sihara@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Shuichi Ihara [Mon, 7 Nov 2011 09:45:33 +0000 (18:45 +0900)]
LU-817 lustre-iokit: sgpdd-survey is encountering r/w errors on arrays using 2TB drives
There are two fix/improvements in sgpdd-survey.
(1) adding --lba option to sg_readcap for large LUNs.
(2) support configurable boundadry block size between
concurrent regions per device.
Signed-off-by: Shuichi Ihara <sihara@ddn.com>
Change-Id: Ib3cdb051cf55e096919ad63a42640aaacfe511c4
Reviewed-on: http://review.whamcloud.com/1658
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Cliff White <cliffw@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Lai Siyao [Fri, 4 Nov 2011 15:08:25 +0000 (08:08 -0700)]
LU-746 test: obdfilter-survey FAIL hndls expected >8, have 2
obdfilter-survey.sh checks jbd proc stats after survey, but it
doesn't take obd cleanup time into account, so the stats collected
may not reflect the data of test. To fix this, save run time of
survey test, and collect stats of (run_time/4) ago.
Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
Change-Id: I597a6279353c203b50b1c2b17543bd34d86d8806
Reviewed-on: http://review.whamcloud.com/1645
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Liang Zhen <liang@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
wangdi [Tue, 24 Jan 2012 03:56:09 +0000 (19:56 -0800)]
LU-1023 utils: Time counting fix for obdfilter-survey
1. Assigning the start and end time of test_brw so it
can get correct time usage.
2. In mds_survey, it adds a new output line for threas utils,
(Total: total xxxx threads X sec ...), and we should skip
this line in get_stats(obdfilter-survey), so it can get
correct min/max time.
Signed-off-by: Wang Di <di.wang@whamcloud.com>
Change-Id: I58c13bde299c254ed10e1405bfed3e1dce4ef216
Reviewed-on: http://review.whamcloud.com/1999
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jinshan Xiong [Thu, 22 Dec 2011 00:33:51 +0000 (16:33 -0800)]
LU-874 ldlm: prioritize LDLM_CANCEL requests
If a lock canceling request has already reached server, there is no
reason of evicting the client when the waiting list times out.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I1a300916b4b3b592ca565ffc06cb3658d699d7a0
Reviewed-on: http://review.whamcloud.com/1900
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Niu Yawei <niu@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Thu, 12 Jan 2012 13:29:00 +0000 (21:29 +0800)]
LU-980 llog: cleanup return value in llog_client_create
in llog_client_create, the newly allocated llog_handle is
return by parameter res, but it doesn't be cleaned up
if the following operations failed and the corresponding
llog_handle is already freed.
Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
Change-Id: I8b59dfde91da2c20881f29e8ff46a0a93f0ee1b2
Reviewed-on: http://review.whamcloud.com/1958
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Lai Siyao [Wed, 18 Jan 2012 07:46:24 +0000 (15:46 +0800)]
LU-998 acl: declare acl operation for setattr
Setattr on ATTR_MODE may set acl if acl is enabled, and it should be
declared in advance.
Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
Change-Id: Id9c94c5498f8fee0a79986bb424444f658c98e60
Reviewed-on: http://review.whamcloud.com/1984
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Li Wei <liwei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Sat, 28 Jan 2012 01:26:53 +0000 (20:26 -0500)]
LU-1048 ldiskfs: fix dynlock cache entry freeing
Update to rhel6.2 broke dynlocks patch and the last hunk was
ignored by quilt/patch.
This commit rediffs the proper patch against 2.6.32-220.el6
Change-Id: If9020458d07c7c2dc714b3e38587f66c4846f806
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/2034
Tested-by: Hudson
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Andreas Dilger [Thu, 26 Jan 2012 12:39:06 +0000 (05:39 -0700)]
LU-797 tests: speed up ost-pools tests
The test time of the ost-pools subtests is unreasonably long.
test_14 fills an OST to 90% full, regardless of the OST size.
Skip the test if the amount of data to be written is too large
to run in a practical time.
test_18 creates 3x3x30000 files to compare performance with/without
pools enabled. Instead of creating a fixed number of files, use
createmany to run for a specific (short) time to measure performance.
test_23 tried to fill all OSTs 100% full. Split this test into two:
- test_23a to test quota with a file in a pool
- test_23b to test OOS with a file striped over pool
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ib851939a210ab68f52da1ac777781c2a922c500c
Reviewed-on: http://review.whamcloud.com/2028
Tested-by: Hudson
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Tue, 10 Jan 2012 21:16:17 +0000 (13:16 -0800)]
LU-972 ldlm: reduce server log message everytime client connect
Changed to D_HA to reduce the messages
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: Ief4507ef484369b6a7aafb7ce9b1f1e0b1f0967e
Reviewed-on: http://review.whamcloud.com/1944
Tested-by: Hudson
Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Mikhail Pershin <tappro@whamcloud.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Wed, 25 Jan 2012 14:49:22 +0000 (09:49 -0500)]
Tag 2.1.55
Change-Id: I7f6bf0b88a7dc5c5a2eb08ccff090c2a17dfce22
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Alexander.Boyko [Fri, 6 Jan 2012 06:38:00 +0000 (10:38 +0400)]
LU-898 ptlrpc: fix ptlrpc request race.
Allow request reorder from export, only if request has been
added to the list.
Race condition with req error handle at ptlrpc_server_handle_req_in().
1. req is added to rq_export->exp_queued_rpc by ptlrpc_hpreq_init()
2. ptlrpc_server_request_add() returns error ( ost_validate_obdo() whith
one of this condition !(fid_seq_is_rsvd(oa->o_seq) ||
fid_seq_is_idif(oa->o_seq))
3. ptlrpc_server_drop_request(req) drops request (disconnect export),
but req is in rq_export->exp_queued_rpc
4. ldlm_server_blocking_ast handle rq_export->exp_queued_rpc and fail
because req->rq_export is NULL.
Fix allows request reorder from export only if request has been added to
the list. So workaround is not need for situation when request from export
was processed before ptlrpc_server_request_add.
Signed-off-by: Alexander Boyko <alexander_boyko@xyratex.com>
Reviewed-by: Vitaly Fertman <vitaly_fertman@xyratex.com>
Reviewed-by: Denis Kondratenko <denis_kondratenko@xyratex.com>
Reviewed-by: Andrew Perepechko <Andrew_Perepechko@xyratex.com>
Xyratex-bug-id: MRP-284
Change-Id: I5f763b3c4f19b6af5f803b50b43a5570dab3dc76
Reviewed-on: http://review.whamcloud.com/1799
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andrew Perepechko <andrew_perepechko@xyratex.com>
Reviewed-by: Denis Kondratenko <Denis_Kondratenko@xyratex.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Tue, 17 Jan 2012 04:10:25 +0000 (12:10 +0800)]
LU-931 mdd: store lu_fid instead of pointer in md_capainfo
in md_capainfo, mc_fid contains at most 5 pointers to lu_fid,
and if the corresponding lu_fid is freed, the pointer isn't notified
about it, then the pointer will point to freed memory!
Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
Change-Id: I00088cbfeb145ceac0477467a8b2436f6cf1e530
Reviewed-on: http://review.whamcloud.com/1979
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Mikhail Pershin <tappro@whamcloud.com>
Reviewed-by: wangdi <di.wang@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Niu Yawei [Wed, 11 Jan 2012 03:59:10 +0000 (19:59 -0800)]
LU-919 obdclass: remove hard coded 0x5a5a5a
We assert atomic_t value with hard coded 0x5a5a5a in several places,
which could result in false assertion failure when the reference count
getting very large in some extreme case.
The hard coded 0x5a5a5a should be replaced by LI_POISON.
Signed-off-by: Bruno Faccini <bruno.faccini@bull.net>
Signed-off-by: Niu Yawei <niu@whamcloud.com>
Change-Id: Idc271621017d071b3e2dce5d0ec6fb854127a955
Reviewed-on: http://review.whamcloud.com/1953
Reviewed-by: Lai Siyao <laisiyao@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Thu, 5 Jan 2012 16:55:48 +0000 (08:55 -0800)]
LU-814 test: automate NFS over lustre testing
Provide setup nfs within auster framework
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Change-Id: Icfd61bf6772807a344576b92b5268a83a7b79e4b
Reviewed-on: http://review.whamcloud.com/1664
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
wangdi [Sat, 14 Jan 2012 07:15:46 +0000 (23:15 -0800)]
LU-879 tests: Add test for checking rename_stats
Add 133d in sanity for checking rename_stats.
Signed-off-by: Di Wang <di.wang@whamcloud.com>
Change-Id: If9a57b9ac458fdf729c19f597d6197f410966e91
Reviewed-on: http://review.whamcloud.com/1970
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Yu Jian <yujian@whamcloud.com>
Johann Lombardi [Tue, 10 Jan 2012 16:41:26 +0000 (17:41 +0100)]
LU-169 lov: add basic infrastructure for layout lock
This patch adds some basic infrastructure to support the layout lock
in a near future. This includes defining a new inode lock bit to lock
the file layout (namely MDS_INODELOCK_LAYOUT) as well as a new lookup
intent (IT_LAYOUT).
Signed-off-by: Jacques-Charles Lafoucriere <jacques-charles.lafoucriere@cea.fr>
Change-Id: Ibf1c3c166b5def4654684febbcf3a99ea7e482eb
Signed-off-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1854
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
yangsheng [Thu, 5 Jan 2012 10:30:42 +0000 (18:30 +0800)]
LU-781 kernel: kernel update [RHEL6.2 2.6.32-220]
Add support for RHEL6.2. The version is 2.6.32-220.el6.
Change-Id: Icc03a2f5d8b377aa1b1180ae09056989bbc84a9d
Signed-off-by: Yang Sheng <ys@whamcloud.com>
Reviewed-on: http://review.whamcloud.com/1892
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Johann Lombardi <johann@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Jinshan Xiong [Thu, 12 Jan 2012 00:03:41 +0000 (16:03 -0800)]
LU-948 clio: add a callback to cl_page_gang_lookup()
Add a callback to cl_page_gang_lookup() so that it will be easier to
fix this issue and be helpful for new IO engine.
If a read lock is being canceled, we used to grab page lock and then
check if they are covered by another lock, otherwise they will be
discarded. This is unnecessary because we can do this w/o grabbing
page lock.
With the above fix, when a read-ahead page is in IO during recovery,
and one of covering locks is being canceled by early cancel for
recovery, it will detect that this page is being covered by another
one, and then this page will be skipped w/o trying to grab page lock.
Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
Change-Id: I22a3ea0790f5c0e01c12c29208b6d60c38058f12
Reviewed-on: http://review.whamcloud.com/1955
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
git://git.whamcloud.com / fs / lustre-release.git / log