LU-7148 osc: Remove remains of osc_ast_guard

osc_ast_guard has been removed by the clio simplification.
Remove the extern declaartion and lock class definition.

Change-Id: Ibcf14e7aebe1dab8b586d3cd8d81560f6d3dcc81
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Reviewed-on: http://review.whamcloud.com/16392
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>

LU-5951 ptlrpc: track unreplied requests

The request xid was used to make sure the ost object timestamps
being updated by the out of order setattr/punch/write requests
properly. However, this mechanism is broken by the multiple rcvd
slot feature, where we deferred the xid assignment from request
packing to request sending.

This patch moved back the xid assignment to request packing, and
the manner of finding lowest unreplied xid is changed from scan
sending & delay list to scan a unreplied requests list.

This patch also skipped packing the known replied XID in connect
and disconnect request, so that we can make sure the known replied
XID is increased only on both server & client side.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: Ic98e1599085871c0ac08d28609a044c79d5af75d
Reviewed-on: http://review.whamcloud.com/16759
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Grégoire Pichon <gregoire.pichon@bull.net>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7508 ldlm: Don't check opcode with NULL rq_reqmsg

When GSS is enabled it's possible to have a NULL rq_reqmsg
if a bad signature or no context is returned during the unwrap
of the request.  Don't check the opcode in this case.

Signed-off-by: Jeremy Filizetti <jeremy.filizetti@gmail.com>
Change-Id: I3a74dff7638b318190c5c4ad73acbe7ec299aa80
Reviewed-on: http://review.whamcloud.com/17414
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>

LU-7268 scrub: NOT assign LMA for EA inode

Originally, when OI scrub scans the device, if the target inode has
no FID-in-LMA EA, then it will generate an IGIF mode FID and store
it in the LMA EA. Such behavior is not suitable if the target inode
is used for large EA. The OI scrub should skip the EA inode that is
marked as "LDISKFS_EA_INODE_FL".

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I52b05b864ef8a2797a2f3dda0f80f95227809c34
Reviewed-on: http://review.whamcloud.com/17043
Tested-by: Jenkins
Reviewed-by: Kalpak Shah <kalpak.shah@seagate.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-6298 hsm: shutdown HSM CDTs in parallel

In sanity-hsm.sh rewrite copytool_cleanup() to shutdown and restart
the MDT HSM coordinators in parallel. This saves about 8 * (MDSCOUNT -
1) seconds per call.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Change-Id: I75445ad126dc73251a3d056611133e3ab6b83362
Reviewed-on: http://review.whamcloud.com/13901
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-5921 tests: enhance server target mount race testing

This patch is a follow on to LU-5299 to strengthen and enhance
concurrent server target mount race testing.
It uses OBD_RACE() feature to better set a concurrent/racy
situation, and also allow to handle all mount errors instead
of only EALREADY.

Signed-off-by: Bruno Faccini <bruno.faccini@intel.com>
Change-Id: I16a94e5aa046e15096d2e55d57e22899a93fa03f
Reviewed-on: http://review.whamcloud.com/17302
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7383 mdt: retry for busy lock during migration

In migration, if the lock of the migrating object
is being cached on other node, it should revoke
the lock and retry, instead of return -EBUSY.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: I1317681a892b9a21f2c78d7696ca6f94d43bd9bc
Reviewed-on: http://review.whamcloud.com/17048
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

New tag 2.7.64

Change-Id: I79fb95af8bc9e979edf3214315219f786eb12599
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7428 test: disable conf-sanity, test_84

Add failing test to ALWAYS_EXCEPT.
This is a temprorary workaround until a real
fix for the test failure is developed.

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I73e00d658a9e7728ce52b5dc90741e9a18ce15f9
Reviewed-on: http://review.whamcloud.com/17482
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-7437 lctl: list_param -R can't work correctly

We shouldn't call lprocfs_param_pattern() inside listparam_display(),
otherwise it will add the prefix "/proc/{fs,sys}/{lnet,lustre}" each
time, so that the parameters can be listed recursively. The similar
issue in {set/get}param is fixed as well.

Also, this patch adds sanity.sh test_401 to verify this function.

Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Change-Id: I21acec364cdbdfc025979153f66a87d44c9136e8
Reviewed-on: http://review.whamcloud.com/17223
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7297 osd-zfs: initialize oh_lock

The ZFS osd was not initializing od_brw_stats.hist[].oh_lock.
This rectifies that.

Change-Id: I3f637b73c77908c2297bfab97e33eca63b0d5986
Signed-off-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-on: http://review.whamcloud.com/16919
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>

LU-1026 ldiskfs: make bitmaps corruption not fatal

We still hit bitmaps problems for rhel6 series kernel,
corruptions happen because ext4_mb_check_ondisk_bitmap()
check failed and FS become RO again:

ldiskfs_mb_check_ondisk_bitmap: on-disk bitmap for group
294corrupted: 20180 blocks free in bitmap, 20181 - in gd
Aborting journal on device dm-6-8.
LDISKFS-fs (dm-6): Remounting filesystem read-only
ldiskfs_mb_new_blocks: Updating bitmap error: [err -30]
 [pa ffff880d9d6e4d68] [phy 14974976] [logic 8192] [len 3072]
 [free 3072] [error 1] [inode 278678]
ldiskfs_ext_new_extent_cb: Journal has aborted

this might be caused by some ext4 internal bugs, this patch
did the following things:

1.Inside ext4_read_block_bitmap() have gaven reasons
why it failed, so caller don't need call ext4_error() again.
2. mark block group corrupt and use ext4_warning() instead
of ext4_error().

There are still some bitmaps corruption places not handling,
let's keep it for now, and if it really hurt, let's add the
same handling codes logic later.

Tested by following scripts:

TEST_DEV="/dev/sdb"
TEST_MNT="/mnt/ext4"

mkdir -p $TEST_MNT
mkfs.ext4 -F $TEST_DEV >&/dev/null

mount -t ldiskfs $TEST_DEV $TEST_MNT
dd if=/dev/zero of=$TEST_MNT/largefile
oflag=direct bs=10485760 count=200
umount $TEST_MNT
dd if=/dev/zero of=$TEST_DEV bs=4096 seek=641
count=10 oflag=direct
mount -t ldiskfs $TEST_DEV $TEST_MNT
rm -f $TEST_MNT/largefile
dd if=/dev/zero of=$TEST_MNT/largefile oflag=direct
bs=10485760 count=200 && echo
  "FILESYSTEM still usable after bitmaps corrupts happen"
dmesg | tail
umount $TEST_MNT
e2fsck $TEST_DEV -y

Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: Iabb6ebf719d80d9ba4f41bee0b237e304212832b
Reviewed-on: http://review.whamcloud.com/16679
Tested-by: Jenkins
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7315 osd-ldiskfs: handle pdo lock properly

Inside the osd_dirent_check_repair(), if the logic comes to
"goto again", it only unlock the "hlock" but without seting
the variable @hlock as NULL. Althouth it will not cause any
logic failure, it may make the readers to be confused. This
patch will set "hlock = NULL;" explicitly to avoid trouble.

On the other hand, inside ldiskfs, the pdo lock users need
to check whether the lock handler is NULL or not properly.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I9db9dc758a2976849c299f76e06723e796da235d
Reviewed-on: http://review.whamcloud.com/16924
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Liang Zhen <liang.zhen@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6856 zfs: handle non existing file in osd_object_ref_del

Remove false assertion in zfs:osd_object_ref_del() because this
may be in the cleanning path of error handling.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Change-Id: Ib7b9d80816bdab7f68b36a33e95140ea7f3eae8c
Reviewed-on: http://review.whamcloud.com/15611
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>

LU-6802 ptlrpc: reset imp_replay_cursor

At client side, the replay cursor using to speed up the lookup
of committed open requests in its obd_import should be resetted
for normal connection (not reconnection) during recovery.

Change-Id: I68816780a5d79053d9109cb68ae1c3b8ea13ede8
Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com>
Reviewed-on: http://review.whamcloud.com/17351
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6693 out: not return NULL in object_update_param_get

Return ERR_PTR in object_update_param_get() for all cases to
avoid unnecessary confusion to callers.

Signed-off-by: wang di <di.wang@intel.com>
Change-Id: Idfcc19d99bbf308759481b3d60d95341745d19e8
Reviewed-on: http://review.whamcloud.com/16417
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7316 build: Update ZFS/SPL version to 0.6.5.3

Bug Fixes

* Fix CPU hotplug zfsonlinux/spl#482
* Disable dynamic taskqs by default to avoid deadlock
  zfsonlinux/spl#484
* Don't import all visible pools in zfs-import init script
  zfsonlinux/zfs#3777
* Fix use-after-free in vdev_disk_physio_completion
  zfsonlinux/zfs#3920
* Fix avl_is_empty(&dn->dn_dbufs) assertion zfsonlinux/zfs#3865

Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: I36347630be2506bee4ff0a05f1b236ba2ba7a0ae
Reviewed-on: http://review.whamcloud.com/16877
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>

LU-4423 lnet: don't use iovec instead of kvec

Replace struct iovec with struct kvec.

Linux commit: f351bad2b4b4bb74810ad4f127f6602e2d2ae403

Change-Id: Ib7bb49069e42ca82d66a149617361c73ee4d710d
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-on: http://review.whamcloud.com/17205
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7462 mdd: check object existence

Check object existence in mdd_is_parent()
and _mdd_lookup(), so the following retrieving
attributes will not panic if object does not
exist.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Icae243949e202f9d6a4d38b9823373101a093c74
Reviewed-on: http://review.whamcloud.com/17323
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Andreas Dilger <andreas.dilger@intel.com>

LU-7375 lbuild: add missing case to lbuild

Add the missing case of el6.7 to autodetect_target

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I6a8032eb03b2fb3002798aa49ee585879cf711c3
Reviewed-on: http://review.whamcloud.com/17024
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7103 test: avoid cat of /dev/urandom

Use a different command to generate a random number.
Using cat /dev/urandom doesn't work in all cases.

Test-Parameters: clientdistro=el7
testlist=ost-pools,ost-pools,ost-pools,ost-pools,ost-pools,ost-pools,ost-pools,
ost-pools,ost-pools,ost-pools,ost-pools,ost-pools,ost-pools,ost-pools

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I63cfad2654cf654460263529457538965e373f82
Reviewed-on: http://review.whamcloud.com/17350
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7447 lfsck: correct nlink attr for new created dir

For new created directory object, there are three main operations:
1) insert dot entry
2) insert dotdot entry
3) ref_add on the directory object.

Sometimes, the 3rd step maybe ahead of 1st, sometimes maybe between
1st and 2nd. Usually, the developers would think that such order is
not important. But in fact, such assumption isn't true, because the
ldiskfs will set the new created directory object's nlink attr as 2
when inserting dotdot entry. Then if ref_add() is called after ".."
inserted, the empty directory object's nlink attr is 3.

To avoid above trouble and make the OSD API to be order-independent,
we will make the osd-ldiskfs to backup the new created dir object's
nlink attr before calling ldiskfs_add_dot_dotdot(), then restore it
after ldiskfs_add_dot_dotdot() called.

On the other hand, this patch also adds some missed logic for lfsck
declaring to insert dot/dotdot entries, remove some redundant logic.

Add e2fsck check for on-disk consistency verification after LFSCK.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I065f683388417f3e23d47471c065419bd9bfcb19
Reviewed-on: http://review.whamcloud.com/17269
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>

LU-7371 test: wrong read length over isize

This patch adds tests to check read length is correct if reading
a file of size 4095.

Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I0c0f6b378fa4af053ed54f2f5dea2418191a7b69
Reviewed-on: http://review.whamcloud.com/17060
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-6851 lnet: Ignore hops if not explicitly set

Since the # of hops is not a mandatory parameter the LU-6060
patch will cause problems to already existing systems since it
changes the behavior by which a route is determined down.

To fix this case the # of hops now defaults to LNET_UNDEFINED_HOPS
if no hop count is specified.

LNET_UNDEFINED_HOPS is defined to ((__u32)-1). When it's printed as
%d, it displays as -1.

__u32 is used through out the call stack for hop count to explicitly
define the size of the hop count and to avoid any sizing issues when
passing data to and from the kernel.

To keep existing behavior both lnet_compare_routes() and LNetDist()
will treat undefined hop count as hop count 1.

When executing the logic in lnet_parse_rc_info() there is no
longer an assumption that the default hop count is 1. If
the hop count is 1 then it must've been explicitly set by
the user.

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Change-Id: I1a28a35a4edc2437cf95cb9d455e59c8102736fa
Reviewed-on: http://review.whamcloud.com/15719
Tested-by: Jenkins
Reviewed-by: Olaf Weber <olaf@sgi.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7400 lod: register stop callbacks at create

In some cases we stop just created transaction (i.e. it's
empty, not started), but then top_trans_stop() waits
indefinitely for stop callbacks which were supposed to be
register at top_trans_start(). instead register them at
top_trans_create().

Instead of register the commit callback in top_trans_start(),
we should register the commit callback once the top thandle
is added into commit list, because in some error cases,
top_trans_start() might not be called, then the top thandle
will stay in the commit list forever, and also blocking
llog cancellation.

Change-Id: I1ca528e2f8f5e4d9cf6b5dd484653a055e17cc6c
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Signed-off-by: wang di <di.wang@intel.com>
Reviewed-on: http://review.whamcloud.com/17059
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7199 nodemap: assign nodemap to export before connecting

This patch moves the nodemap assignment to be before the connection
is active, and the nodemap deassignemnt to be after the connection
is made inactive. It also checks for null nodemaps and returns
-EACCES if the nodemap is not set in order to avoid a kernel panic.

Signed-off-by: Kit Westneat <kit.westneat@gmail.com>
Change-Id: Id4dd02cf5b4208bda6d6f17aed6adb13b18c7731
Reviewed-on: http://review.whamcloud.com/16802
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7428 tests: write superblock in conf-sanity test_84

In conf-sanity.sh test_84() a newly-formatted MDS filesystem's block
device is set read-only immediately after mount via replay_barrier(),
which may result in initial formatting or configs to be discarded,
resulting in a wide variety of different failure modes for this test.

Ensure that the superblock and other configuration logs are flushed
to disk before replay_barrier() is called, so that the MDS can mount
properly again later in the test.

Test-Parameters: testlist=conf-sanity,conf-sanity,conf-sanity,conf-sanity
Test-Parameters: testlist=conf-sanity,conf-sanity,conf-sanity,conf-sanity
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I95894b600ed7596c2014cba4f35fef4b443ebbe5
Reviewed-on: http://review.whamcloud.com/17371
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7329 obdclass: sync device to flush journal callbacks

After llog_test_10() has been added to check correct Catalog
wrap-around as part of LU-6556, frequent auto-tests failures
have been encountered due to OOM situation on VMs caused by
a lot of journal commit callbacks backlog induced by the huge
LLOG activity generated by new test.
This patch forces frequent device sync to flush these journal
commit callbacks.

Signed-off-by: Bruno Faccini <bruno.faccini@intel.com>
Change-Id: If06914a395f5bacbbb55fcdc229ffcd47d742843
Reviewed-on: http://review.whamcloud.com/17052
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Mike Pershin <mike.pershin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7098 osd-ldiskfs: don't alloc inode directly

We should alloc ldiskfs_inode_info instead alloc inode
directly. Else will overflow in follow ldiskfs_add_entry.

Signed-off-by: Yang Sheng <yang.sheng@intel.com>
Change-Id: Ib88f847ea8f160f6f0d0c7e7e680247d7ec96fb6
Reviewed-on: http://review.whamcloud.com/16804
Reviewed-by: Fan Yong <fan.yong@intel.com>
Tested-by: Jenkins
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>

LU-1095 mdc: remove console spew from mdc_ioc_fid2path

In some cases with a very long pathname, such as with sanity.sh
test_154c, mdc_ioc_fid2path() would spew long debug messages to
the log, because libcfs_debug_vmsg2() refuses to log messages over
one page in size.

Truncate the debug message to only log the last 512 characters
of the pathname, which is sufficient for most debugging, saves a
bit of space in the debug log, and will prevent the debug logging
from printing to the console in the first place.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I3cad19d02e8065574b8baf5694e9894e43112318
Reviewed-on: http://review.whamcloud.com/17078
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7396 llite: check request != NULL in ll_migrate

Check if the request is NULL, before retrieve reply body
from the request.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Ifec9caf270b938b7583de0315610f930fa52649d
Reviewed-on: http://review.whamcloud.com/17079
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-7276 utils: make llog_reader consistent with kernel

Handle the CM_SKIP records in the same manner as the kernel,
by printing "SKIP" only until the first CM_END record, rather
until the first CM_END|CM_SKIP record.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: Ie0c35ecd9d74a1e0b378bcf6642bca32e4dfa35e
Reviewed-on: http://review.whamcloud.com/16788
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Reviewed-by: Mike Pershin <mike.pershin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7068 mdd: object leak in mdd_migrate_entries

mdd_migrate_entries should release the object
if mdd_trans_start() fails.

Signed-off-by: wang di <di.wang@intel.com>
Change-Id: If10dce8b450e8c19bb93465beb08118e98c4ed96
Reviewed-on: http://review.whamcloud.com/16384
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>

LU-6666 osc: Do not merge extents with partial pages

After range lock is introduced to Lustre, it's possible for
multiple threads to submit osc_extents with partial pages, and
finally I/O engine may try to merge these extents, which will
end up with assert in osc_build_rpc().

In this patch, osc_extent::oe_no_merge is introduced, and this flag
is set if osc_extent submitted via osc_io_submit() includes partial
pages. This flag is used by I/O engine to stop merging this kind
of extents.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Change-Id: I0c851912beb0370553bc2e2b05f80dee175a2f1f
Reviewed-on: http://review.whamcloud.com/15468
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Patrick Farrell <paf@cray.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7463 osd: Change existence assert into error

In osd_declare_xx(), some objects might not existent,
especially when calling out_handler(). Let's change
these assert into -ENOENT error to avoid panic on
MDS.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: If17a013a6939a1ebe0519406d39e405fd915110f
Reviewed-on: http://review.whamcloud.com/17324
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>

LU-7461 lod: retry to get remote update log

If the remote MDT is also in recovery status,
then retrieving update logs in lod_sub_recovery_thread()
might return -EAGAIN or -EIO or -EBUSY, let's
retry in this case until the recovery is aborted or
the local MDT is umounted.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Iee945942bd01925cdcfe75c4e59dccbd63b34498
Reviewed-on: http://review.whamcloud.com/17322
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7416 osp: check rq_repmsg in osp_request_commit_cb

Check if rq_repmsg is NULL before retrieving
last committed transno from reply message.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Ibf1e110e33df333934c65dfcf52870954e936180
Reviewed-on: http://review.whamcloud.com/17130
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7343 osd-ldiskfs: handle ldiskfs_append failure

In new linux kernel (linux-3.1x, x>=0), the ldiskfs exported
function ldiskfs_append() return error# via the return value,
instead of via the output parameter @err as it does on other
kernels (linux-2.6). Under such case, the caller should not
assume non-NULL returned value is valid buffer head, it can
stands for error#. So check that properly.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I4dca43bcfd31aafd999f54934a51d258071dab22
Reviewed-on: http://review.whamcloud.com/17148
Tested-by: Jenkins
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-7376 tests: sanity-hsm/59 should skip old servers.

there is no poin to crash vulnerable versions.

Change-Id: Iacafd10d2a3d04ba1bb9ca70d8e343809490a349
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/17029
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7436 tests: skip conf-sanity/91 with old servers

due to missing functionality.

Change-Id: I2be72820413aee9a9d7082c2d8c6f308eeb6e141
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/17222
Tested-by: Jenkins
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7415 kernel: kernel update RHEL 6.7 [2.6.32-573.8.1.el6]

Update RHEL6.7 kernel to 2.6.32-573.8.1.el6

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I1b3c4b144b06e5e96f818e35c08f490e574ed798
Reviewed-on: http://review.whamcloud.com/17119
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7164 osc: osc_extent should hold refcount to osc_object

To avoid a race that osc_extent and osc_object destroy happens on the
same time, which causes kernel crash.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Change-Id: I3e3237f0d1cff4bd992bef4e4c01355a1d5c8d9f
Reviewed-on: http://review.whamcloud.com/16433
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Patrick Farrell <paf@cray.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7384 lfsck: check transaction stop status

The LFSCK modification will be sent to remote server when the
transaction stop, for sync transaction case, we can check the
dt_trans_stop() result.

If the lfsck_namespace_create_orphan_dir() failed, but we may
ignored that before because of ignoring dt_trans_stop result.
Then it may cause subsequent lfsck_namespace_insert_normal()
failed at LASSERT(lu_object_exists(o) != 0);

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: If897b7bd479ecdb61e6435f3177211f865a4e303
Reviewed-on: http://review.whamcloud.com/17042
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-3536 lfsck: reuse parameter name for re-locating object

Usually, LFSCK engine will locate the object against the bottom
device (OSD), then make related check/repair directly. Sometimes,
such as lfsck_namespace_repair_dirent(), we need to modify based
on LOD device. Under such case, the LFSCK will re-locate related
object with the same FID.

Originally, there is no special rules about the parameter's name,
that is confused which one should be used. For example, the input
parameter is named as "parent" that is against OSD, we need to
re-locate the obj based on the LOD, named as "pobj", then in the
subsequent logic, "pobj" should be used, but unfortunately, the
"parent" may be used by wrong. It is difficult to find out such
invalid usage.

To avoid such trouble, we prefer to reuse the (input) parameter
name after re-locating the object, name "pobj" as "parent".

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I5b6d7c5c10e1817ef2bade4931485228b26c511d
Reviewed-on: http://review.whamcloud.com/16932
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>

LU-3322 lnet: make connect parameters persistent

Store map-on-demand and peertx credits in the peer, since the peer
is persistent. Also made sure that when assigning the parameters
received on the connection to the peer structure through create,
that if another peer is added before grabbing the lock we assign
these parameters to it as well.

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Change-Id: Ie68f1ba1349d15b0a31eff9a2ca454df8e408ea9
Reviewed-on: http://review.whamcloud.com/17074
Tested-by: Jenkins
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7324 lnet: recv could access freed message

When lnet_parse_put calls lnet_ptl_match_md, this function can attach
current message on the delayed list if there is no match. It means
this message can be taken over and freed by another thread who is
posting new MD, then it is not safe for caller of lnet_parse_put to
check this message again.

This patch fixes this issue by adding a local variable "ready_delay"
to store corresponding status of lnet_msg, so lnet doesn't need to
check the message again if lnet_ptl_match_md returned MATCH_NONE for
it.

Signed-off-by: Liang Zhen <liang.zhen@intel.com>
Change-Id: I0f8827103dd637648112e936ce6e685266e5ca40
Reviewed-on: http://review.whamcloud.com/17065
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7221 ldlm: do not take a reference on target if stopping

In the set of changes of patch for LU-5569
(http://review.whamcloud.com/11750/,
commit 892078e3b566c04471e7dcf2c28e66f2f3584f93) one is to take a
reference on target even if it is stopping (umount'ed). Then, upon
connections attempts, this can lead to unwanted cleanup actions to
occur on [obd_self_]export from class_decref(), finally causing a
LBUG in class_export_put() because export's exp_refcount has already
reached 0.

Signed-off-by: Bruno Faccini <bruno.faccini@intel.com>
Change-Id: If960437934fb694d173a4fd1fbfb9e43d496fea6
Reviewed-on: http://review.whamcloud.com/16940
Tested-by: Jenkins
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7318 out: dynamic reply size

every update on the initiator side can declare how many bytes
it expects back. OUT packing library put these numbers on the
wire and prepary an appropriate buffer for the reply. then OUT
target do few checks to ensure individual replies fit their
buffers.

Change-Id: I443b5c879bc321c33efb70af665ecd2b2f7baa18
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/16889
Tested-by: Jenkins
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7077 target: avoid using possible error return NULL pointer

previous fix http://review.whamcloud.com/15576 added a call
to cfs_hash_getref().  add LASSERT() to ensure the can
never happen here return value of NULL is in fact never seen.

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Ic6132b5450534db0bb9b89c3dd6f55517450c42a
Reviewed-on: http://review.whamcloud.com/16473
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>

LU-7174 build: make git ignore dkms generated file

While testing patches other non-patch the related build
by product dkms.mkconf show up with git status. To avoid
adding this by accident place thes by product files in the
proper .gitignore files.

Change-Id: I49a5411f8c1159a75d1cd28067dfbf3c2a677d6c
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: http://review.whamcloud.com/17136
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7169 tests: check disk corruption during failover

It is a debug patch for conf-sanity test_84. It is suspected
that there is some disk corruption during the MDT0 failover.

Test-Parameters: mdsfilesystemtype=ldiskfs mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs testlist=conf-sanity,conf-sanity,conf-sanity
Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I7e20f26e1ecee483474ace44c8284b5776f3c602
Reviewed-on: http://review.whamcloud.com/16664
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>

LU-7377 utils: Don't fail on plugin load for mount/tunefs

While loading mount_utils_zfs, if zfs modules aren't loaded, but
zfs plugin is present, it will return an error, this shouldn't cause
all module loading to fail.

Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: Idad52745cdfa9d673ab9bd4afe38de4d51ae9a49
Reviewed-on: http://review.whamcloud.com/17128
Tested-by: Jenkins
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7414 target: do not share update and rdbuf

Redefine lu_rdbuf structure to simplify the rdbuf
allocation in out_read().

And also move tti_u.rdbuf out of tgt_thread_info
union, otherwise rdbuf and update will share
the same memory and cause corruption, see out_read().

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Idb0f5af1b00fd5fd15ebc8742aa60d9a43df0a8a
Reviewed-on: http://review.whamcloud.com/17129
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>

LU-7200 kernel: kernel update [SLES11 SP3 3.0.101-0.47.67]

Update SLES11 SP3 kernel to 3.0.101-0.47.67

Test-Parameters: mdsdistro=sles11sp3 ossdistro=sles11sp3 \
  clientdistro=sles11sp3 mdsfilesystemtype=ldiskfs \
  mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs \
  testgroup=review-ldiskfs

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: Icd7bac6bea6866f82e2e03f5dbbb1bda1a4ecacf
Reviewed-on: http://review.whamcloud.com/16617
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

New tag 2.7.63

Change-Id: I79f285380612f61679c1cf8e51446c9018e8225c
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6204 build: clean up kernel module metadata

Update static MODULE_VERSION() lines - this should be automated.

Improve MODULE_DESCRIPTION() descriptions.

Make the name of the module_init()/_exit() functions consistently
{module_name}_init and {module_name}_exit.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I1c3fe5698c7f41d971a38225650597c913500c1e
Reviewed-on: http://review.whamcloud.com/16787
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7345 obdclass: annotate locks in __local_file_create

dt_write_lock() is called for both the child and the parent dt_objects
when a directory is created.  This triggers a false positive in
lockdep when running with CONFIG_LOCKDEP=y, as the structure
containing the lock and the name of the lock is the same, and so it
appears to be a recursive lock attempt based on lock class.

This gives the two locks different subclasses so lockdep can
differentiate between them.

Also, osd-zfs osd_object_{read,write}_lock() functions currently
ignore the subclass (role) provided by the caller, calling down_read()
instead of down_read_nested() for example.

Make osd_zfs use the _nested variants so the role takes effect.

Signed-off-by: Olaf Faaland <faaland1@llnl.gov>
Change-Id: Iab79feadfbd7d1a5a06749ecb9f6888b55a78d73
Reviewed-on: http://review.whamcloud.com/16957
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7269 ptlrpc: remove ptlrpc_prep_req

Remove unused functions ptlrpc_prep_req, ptlrpc_prep_req_pool
Combine __ptlrpc_request_bufs_pack and ptlrpc_request_bufs_pack

Signed-off-by: Ben Evans <bevans@cray.com>
Change-Id: I4e4e64aa1f7fa4c85daf311906f6417a513dcddc
Reviewed-on: http://review.whamcloud.com/16765
Tested-by: Jenkins
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Chris Horn <hornc@cray.com>

LU-7362 lnet: Remove LASSERTS from router checker

In lnet_router_checker(), there are two LASSERTS.  Neither protects
us from anything and one of them triggered for a customer crashing
the system unecessarily.  This patch removes them.

Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Change-Id: If2732632e47103fb8fa63a263c4c5ef4a44142a3
Reviewed-on: http://review.whamcloud.com/17003
Tested-by: Jenkins
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Chris Horn <hornc@cray.com>
Reviewed-by: Matt Ezell <ezellma@ornl.gov>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7296 ldlm: improve lock timeout messages

In ldlm_expired_completion_wait() remove the useless LCONSOLE_WARN()
message and upgrade the LDLM_DEBUG() statement to LDLM_ERROR().

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Change-Id: I6293720bf8e038057a2c84a715359cdbb8cebe91
Reviewed-on: http://review.whamcloud.com/16824
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7120 scrub: handle osd_scrub_post return value

To avoid missing some failure cases during write scrub status
to disk.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I7f77bdba184f634b4f9dd748c3f1b97609b81960
Reviewed-on: http://review.whamcloud.com/16368
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Ashish Purkar <ashish.purkar@seagate.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6013 utils: don't initialize OSD code for client mount

Don't even try to initialize the server OSD handling code if this
is a client mountpoint.  That avoids potential problems if the OSD
code isn't working or available when it isn't needed on a client.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I104b70b9d27811d306879fc047a83f85ea3ebbe5
Reviewed-on: http://review.whamcloud.com/13019
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Tested-by: Jenkins
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

Revert "LU-4865 zfs: grow block size by write pattern"

This reverts commit 3e4369135127b350dbc26a4a5dc94cfa46e394cf.

This has shown problems in testing and may be the cause of LU-7392.

Change-Id: I664f7f8c943d8a90f2d2a9845aea2636535d6b1e
Reviewed-on: http://review.whamcloud.com/17053
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7330 ldlm: fix race of starting bl threads

There is race in the code of starting bl threads which leads to
thread number exceeds the maximum number when race happened, it
can also lead to duplicated thread name.

This patch fixes the race and cleanup the code a bit.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: I9c9be125d1d76890b8c52476684976dad3cb3d87
Reviewed-on: http://review.whamcloud.com/17026
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-2222 mdt: restore evict-by-nid functionality

Writing a NID or UUID to mdt.*.evict_tgt_nids will evict clients
with NID or UUID specified all the targets (OSTs and MDTs).

Change-Id: I66a60a6c81fbac1571f5685111df7b00a306be36
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/16867
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7367 tests: fix fail_loc code in 110g

test 110g used wrong code for OBD_FAIL_MIGRATE_NET_REP

Change-Id: I5dc2e18fb99e35422a9ae227e2e16ba7d39600a3
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/17009
Tested-by: Jenkins
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7366 build: resolve compile issues on Ubuntu 15.04

Ubuntu 15.04 has been released which uses a 4.2 kernel
and gcc 5.2. The gcc version fails to build lustre due
to missing headers in the GSS userland code and a
variable not being uninitialized in the llite layer.

Change-Id: I3615414ac039277a6ef6c6af1a541590b9d79566
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: http://review.whamcloud.com/17008
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7353 utils: fix lctl usage messages

Fix the lctl usage message for sub-commands that need the LNet network
to be specified.  lctl::g_net_is_set() incorrectly recommended using
the "network" command to specify the LNet network, which is incorrect
when using lctl in command-line mode:

  # lctl peer_list
  You must run the 'network' command before 'peer_list'.
  # lctl network tcp0 peer_list
  # lctl --net tcp0 peer_list
  12345-192.168.20.1@tcp [1]192.168.40.147->mookie-gig:988 #15

Fix that to correctly recommend using the "--net" command when using
non-interactive mode, and to return an error if "network" is used in
non-interactive mode with extra arguments.

Replace mention of "portals" in help messages with "LNet".
Remove mention of obsolete elan, qsw, ra network types.

Improve the help message content for related subcommands.
Fix whitespace and command descriptions for related subcommands.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: Idf9bf663b16012ebc9f38566ecba9859a54cab07
Reviewed-on: http://review.whamcloud.com/16980
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7336 ofd: cleanup proc when ofd_info_init fails

In ofd_init0(), if ofd_info_init() fails it should cleanup
procs.

Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I3ff278526f09ef7e36631712ce21a498a6644907
Reviewed-on: http://review.whamcloud.com/16934
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7012 osp: don't use OSP when import is deactivated

Unset opd_imp_connected flag upon IMP_EVENT_INACTIVE event,
it will stop any llog processing by that device until import
will be activated again.

Signed-off-by: Mikhail Pershin <mike.pershin@intel.com>
Change-Id: Ie219e536c216130f428ba933d11842511692c95b
Reviewed-on: http://review.whamcloud.com/16937
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Jenkins
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7371 osd-ldiskfs: fix wrong read length over isize

If the isize is 4095, a read length of 4096 will be
returned because a wrong calculation of EOF. This patch fixes the
problem.

Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I73b18641f000a2d96067243c08c26e51d0d53244
Reviewed-on: http://review.whamcloud.com/17020
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Mike Pershin <mike.pershin@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7261 ldiskfs: clean up code style for large_xattr

Clean up the code style for the large_xattr patches to match the
upstream kernel style (use ! instead of == 0, and similar), and
the original style of the code before the earlier versions of the
patch was applied.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: Icffe339d1b002a55984856829afde9e3eae98bd9
Reviewed-on: http://review.whamcloud.com/16778
Tested-by: Jenkins
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7379 kernel: kernel update RHEL7.1 [3.10.0-229.20.1.el7]

Test-Parameters: mdsdistro=el7 ossdistro=el7 \
  clientdistro=el7 mdsfilesystemtype=ldiskfs \
  mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs \
  testgroup=review-ldiskfs

update RHEL 7.1 kernel to 3.10.0-229.20.1.el7

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I690daa6493232353703f5392cfe0a979b824f3f1
Reviewed-on: http://review.whamcloud.com/17044
Tested-by: Jenkins
Reviewed-by: Minh Diep <minh.diep@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7304 ldiskfs: fix bug when bigalloc is enabled

See following error when enabled bigalloc feature
for ldiskfs rhel7:

LDISKFS-fs error (device sdb):
ldiskfs_mb_check_ondisk_bitmap:3611: comm mkdir:
 on-disk bitmap for group 8corrupted: 0 blocks free in
 bitmap, 32768 - in gd

Fixed to use EXT4_CLUSTERS_PER_GROUP, otherwise,
we will get wrong value and fail to check, which
make FS become RO..

Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I7f61918918e6f4e2f372929181b704b0648dcbca
Reviewed-on: http://review.whamcloud.com/16832
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7354 osd: avoid NULL pointer in osd_obj_update_entry

In osd_obj_update_entry(), the variable @oi_fid may be NULL.
We need to check such case before further using it to avoid
accessing invalid RAM.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: Ibf47e949d69f0b9e5657a6dce2007fe4f6f1a9f6
Reviewed-on: http://review.whamcloud.com/17010
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7243 misc: update Intel copyright messages 2015

Update copyright messages in files modified by Intel employees
in 2015 by non-trivial patches.  Exclude patches that are only
deleting code, renaming functions, or adding or removing whitespace.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I70fe6a346790e15d23606a3f380e7ef8fb8b84a0
Reviewed-on: http://review.whamcloud.com/16758
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7230 llite: clear dir stripe md in ll_iget

If ll_iget fails during inode initialization, especially
during striped directory lookup after creation failed,
then it should clear stripe MD before make_bad_inode(),
because make_bad_inode() will reset the i_mode, which
can cause ll_clear_inode() skip freeing those stripe MD.

Remove the name entry from the directory, once creation
failed. Note: this will not rollback all of local
operation, and LFSCK will take care of the orphan object.

Add sanity.sh 300p to verify the case.

And also enable lfs rm_entry for local object as well,
because sometimes it is quite possible to create the
local corrupted striped directory, and we might need
use "lfs rm_entry" to delete the corrupted striped dir.

Signed-off-by: wang di <di.wang@intel.com>
Change-Id: I892c52117b83c8348aa0ceb888e73c84e79ffe46
Reviewed-on: http://review.whamcloud.com/16677
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-6634 llog: destroy plain llog if init fails

llog_cat_add_rec() should destroy the plain llog
in the same transaction if initialization of that
failed. also, llog_osd_write_rec() should check
the object still exists as it's possible that
another thread failed to initialize and destroyed
the llog.

Change-Id: I7b823d34b32b5caaf0cc17b4cfe278a07a78ec15
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/16427
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7277 lod: keep trying to get remote update log

Because the remote MDT might be in recovery at the same
time, let's Keep trying to get remote update log until
the recovery is abort.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Id9543201ce543be730e73f9f51f3f7a0d10d3dfc
Reviewed-on: http://review.whamcloud.com/16786
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6992 test: wait device to be registered

in mount_facet, the device label can only be used after the device
registered to MGS and it was rewritten.

Change-Id: I9ed65631391f2be84e484e409fbfe59020d982be
Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com>
Reviewed-on: http://review.whamcloud.com/16781
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6627 mdc: quiet console message for known -EINTR

If a user process is waiting for MDS recovery during close, but the
process is interrupted, the file is still closed but it prints a
message on the console.  Quiet the console message for -EINTR, since
this is expected behaviour.

Fix code style issues in this function.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I3bc7284bb7014dbaeb9bfb27a4e98a8abb4a54b6
Reviewed-on: http://review.whamcloud.com/14911
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-5187 ofd: Fix precreate console warning

Alter console warning to be more explanitory.  Also add debug log
message to save developer centric info.

Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: Ib1192562f75b1e2a2c8c9f4ee476da54ebeb1a9a
Reviewed-on: http://review.whamcloud.com/10767
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-5472 tests: export UMOUNT to avoid SLES12 issue

sles12 umount command has issue with '-d' option.
it will report error while a absolute pathname
of mountpoint present with '-d'. So we export
UMOUNT to avoid such problem. In fact, loopdev
can be free automaticly. So '-d' needn't given
explicity.  But consider compatibility we still
keep it.
Also include http://review.whamcloud.com/#/c/14799/
port patch to master.
In SLES12, umount command will run statfs() on the
filesystem, which will cause unmounting Lustre client
hang when OST is unavailable. This patch adds "-f"
option to zconf_umount in conf-sanity.sh to avoid the
issue.

Test-Parameters: alwaysuploadlogs envdefinitions=ONLY=32 clientdistro=sles12 mdtcount=1 testlist=sanity
Test-Parameters: alwaysuploadlogs envdefinitions=SLOW=yes clientdistro=sles12 mdtcount=1 testlist=conf-sanity
Signed-off-by: Yang Sheng <yang.sheng@intel.com>
Change-Id: If466c2101e0db52b5ec1f7273a846dc2497cfb84
Reviewed-on: http://review.whamcloud.com/16445
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Tested-by: Jenkins
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-5770 osd: find bufsize in declare_xattr_set

mdd_link/mdd_unlink need to know LINKEA buffer size before
transaction start for correct estimation of tx credits.
In case if xattr size not known, make a call to osd_xattr_set()
to find correct buffer size.

Xyratex-bug-id: MRP-2093
Signed-off-by: Alexander Zarochentsev <alexander.zarochentsev@seagate.com>
Change-Id: I890e8182bd2d191a0322d25f0684f2a220873546
Reviewed-on: http://review.whamcloud.com/12412
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7039 llog: skip to next chunk for corrupt record

Skip to next chunk if current record is corrupted.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: I7729ca1b10646fa796a3f94aabe39d8d36cf613a
Reviewed-on: http://review.whamcloud.com/16740
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7325 ldiskfs: use correct types for inode num

using signed integer for inode numbers in large EA code
resulting incorrect inode numbers when casting to
unsigned long in ext4_iget() call.

Seagate-bug-id: MRP-3025
Signed-off-by: Alexey Lyashkov <alexey.lyashkov@seagate.com>
Signed-off-by: Alexander Zarochentsev <alexander.zarochentsev@seagate.com>
Change-Id: I49e578a87c4d0f0274a9a42151675822f57c1c5f
Reviewed-on: http://review.whamcloud.com/16913
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6385 tests: sync I/O in obdfilter-survey

This ensures I/O is synced before each test.  Previous patch
http://review.whamcloud.com/14143 works only with the Lustre test
framework.  This will ensure I/O is always sysnced.

Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: Ib19a9e4afd8ca83deceb78fe8fdb4d231da0bc40
Reviewed-on: http://review.whamcloud.com/16942
Tested-by: Jenkins
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6215 kernel: report 4.2.1 kernel support

Add to the Lustre ChangeLog file support for 4.2.1 linux
kernels for clients and servers in the case of using the
latest ZFS backend.

Change-Id: If126c6c1dc8e46e0485f04090590759dc2eb1587
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: http://review.whamcloud.com/16769
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7040 test: skip sanity-hsm 12q for old MDSs

Before 2.7.58 MDSs do not set OBD_MD_TSTATE in some of the cases that
they should. So skip sanity-hsm test_12q when running against these
servers.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Change-Id: I2ebe1b24e28bd4a0f5c03eace1c74678c7442c5f
Reviewed-on: http://review.whamcloud.com/16683
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-1032 build: DKMS RPM for Lustre Client modules

Permit Lustre Client (only) modules DKMS RPM creation.

This patch is a follow on to the first set of patches for LU-1032
that only allowed for the creation of Lustre Server (zfs only)
modules DKMS RPM.

It also changes original behavior by allowing to dynamically
modify dkms.conf on-target. This particularly helps to change
configure and list of modules to be built/installed list, like
to configure with gss to build ptlrpc_gss.ko module when
krb5_devel is present instead to have it been a mandatory
required dependency.

Also implements feature of DKMS RPM creation from Makefile
(thanks to mjmac), now in 2 separate SRPM/RPM steps and for both
Client and Server versions.

Also use an auto-increment (Array[${#Array[@]}]=) operator in
dkms.conf modules declarations to help for future changes when
there will be a need to add/delete modules.

Change in lustre/utils Makefile has been required to allow
building of ptlrpc_gss module with --enable-gss and without the
need to specify --enable-utils which was causing an unexpected
zfs user-land dependency for DKMS Server build.

To satisfy lustre rpm requirement of a package providing
lustre-osd, provides has been added to DKMS Server RPM since
it does actually generate osd-zfs module.

Signed-off-by: Michael MacDonald <michael.macdonald@intel.com>
Signed-off-by: Bruno Faccini <bruno.faccini@intel.com>
Change-Id: I278d50307a17fe49a06392351890946b7dd3557a
Reviewed-on: http://review.whamcloud.com/12347
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6889 kernel: new kerrnel [SLES11 SP4 3.0.101-65]

add target and config files for SLES11 SP4

Test-Parameters: mdsdistro=sles11sp4 ossdistro=sles11sp4 \
  clientdistro=sles11sp4 mdsfilesystemtype=ldiskfs \
  mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs \
  testgroup=review-ldiskfs

Test-Parameters: envdefinitions=ONLY=205 \
  testlist=sanity,sanity,sanity,sanity,sanity,sanity \
  clientdistro=sles11sp4 mdsdistro=sles11sp4 ossdistro=sles11sp4

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: Iba96fbbc834df76fbc1af019c5e67c4ca0282272
Reviewed-on: http://review.whamcloud.com/15832
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7299 utils: allow mkfs.lustre --index to specify in hex/dec

The mkfs.lustre --index argument should be able to handle hex
index values as well as decimal values, especially since the
OSTxxxx identifiers are printed in hexadecimal as well.

Signed-off-by: Thomas Stibor <t.stibor@gsi.de>
Change-Id: I9f7564e3d674353fbebef18bde1598c01bb5bb2c
Reviewed-on: http://review.whamcloud.com/16831
Tested-by: Jenkins
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>

LU-7186 lod: do not propagate size if stripeless

if a file has no stripes, but the size isn't zero,
then do not try to propagate this size to stripes.

Change-Id: I25401bcc41e3ea84a2b9158120f1e907af47fafa
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: http://review.whamcloud.com/16743
Reviewed-by: Frank Zago <fzago@cray.com>
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7086 tests: resolve /sbin symlink in test-framework.sh

In rhel7 and other new distros, /sbin is linked to /usr/sbin, so
the mount check in load_modules_local() is not matched anymore.
To avoid multiple bind mounts, this patch checks /sbin symlink
before using it directly.

Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Change-Id: I287ccd81ae4187a381a7f94dee30338d20dd6155
Reviewed-on: http://review.whamcloud.com/16606
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7109 lfsck: update OST-index in IDIF inside OSD

Old IDIF used "0" as the OST index, that may cause compatibility
issues when update. There is switch inside osd-ldiskfs for controlling
whether convert old IDIF to new one. Once the real OST index became
part of the IDIF and stored on disk (in LMA EA), then the OST cannot
be downgraded.

Because the conversion switch is inside osd-ldiskfs, the LFSCK should
not update the IDIF-in-LMA, instead, leave it to be handled by OSD to
avoid downgrading trouble.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I9fee60ffabf732c0ab0734b1184e4e49638c9e88
Reviewed-on: http://review.whamcloud.com/16282
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6920 test: add some slack to jobstats expiry in test_205

Add a tiny fudge factor to the time allowed for jobstats
to expire before checking on them.

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: Iff2c70e2522b0da0e9c2080cadde49b96a9af60a
Reviewed-on: http://review.whamcloud.com/16753
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6443 tests: add debugging to mmp.sh test

Print out information about MMP intervals to help debug problems
in the mmp.sh tests.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I6865f1c4a59c83f0dc9e46953b9d7ac7cc3ebbe5
Reviewed-on: http://review.whamcloud.com/14442
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7245 socklnd: Bind peers to a specific CPT

Currently the socklnd driver doesn't support
CPT affinity for its peers. Binding peers to
a specific CPT and memory allocated to the
NUMA node belonging to the CPT should give a
performance boost.

Change-Id: I1cc418dc4ba6269e346a4aa1454de79e580e1fba
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: http://review.whamcloud.com/16710
Tested-by: Jenkins
Reviewed-by: Olaf Weber <olaf@sgi.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-7232 statahead: lock leaks if statahead file recreated

During statahead file may be recreated, though this is rare case,
current code will leak the lock, this patch will release lock in
this case.

Signed-off-by: Lai Siyao <lai.siyao@intel.com>
Change-Id: Ifc1fee0fe5a0f377badc3fdf6dc2a6950a26cff6
Reviewed-on: http://review.whamcloud.com/16841
Tested-by: Jenkins
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

LU-6895 lfsck: conflict lu_dirent_attrs members

It is by wrong that the LUDA_UNKNOWN is defined the same value as
LUDA_UPGRADE.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I4067d7852166fab77df69736b3a160e9f67a6abc
Reviewed-on: http://review.whamcloud.com/16821
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>