Whamcloud - gitweb
Patrick Farrell [Thu, 13 Jun 2019 20:51:44 +0000 (16:51 -0400)]
LU-11963 osd: Add nonrotational flag to statfs
It is potentially useful for the MDS and userspace to
know whether or not an OST is using non-rotational media.
Add a flag to obd_statfs that reflects this.
Users can override this parameter in proc.
ZFS does not currently make this information available to
Lustre, so default to rotational and allow users to
override.
Lustre-Change: https://review.whamcloud.com/34235
Lustre-Commit:
68635c3d9b3113621b93fd989f1a3f8f064385b9
LU-12396 utils: lfs should not output 'nul' char
If lfs prints a nul char, it breaks parsing of the output.
Fixes:
68635c3d9b31 ("LU-11963 osd: Add nonrotational flag to statfs")
Lustre-Change: https://review.whamcloud.com/35137
Lustre-Commit:
fd3958b61c5f1c7ed520f07553b999af5522d8e0
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: Iac2b54c5d8cc1eb79cdace764e93578c7b058661
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35226
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Gu Zheng [Thu, 11 Jul 2019 05:52:38 +0000 (13:52 +0800)]
LU-11672 ldlm: awalys cancel aged locks regardless enabling or disabling lru resize
Currently cancelling aged locks is handled by of ldlm_pool_recalc routine,
and it only works when lru resize is enabled, means if we disabled lru
resize, old aged locks are still cached even though they reach the
ns_max_age.
But theoretically, even lru resize disabled, lru_max_age should behave
same as enabling lru resize. At the end, lru_size is like hard limit of
number of locks, but ns_max_age/lru_max_age is a elimination mechanism,
regardless enabling or disabling lru resize meaning once it gets
lru_max_age, locks need to be cancelled.
So fix it here with changing the lru flags when invoking ldlm_cancel_lru
to do the real cancel work, if lru resize is enabled, set flag to
LDLM_LRU_FLAG_LRUR, otherwise LDLM_LRU_FLAG_AGED.
Lustre-change: https://review.whamcloud.com/35467
Lustre-commit:
e4c490bac7701435cb08ce444d9b23b8fd1dd839
Change-Id: Ic2df2550af87fd7209fdb31ca3730683d727a74d
Signed-off-by: Gu Zheng <gzheng@ddn.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-on: https://review.whamcloud.com/35660
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Tue, 11 Jun 2019 15:39:17 +0000 (11:39 -0400)]
LU-8066 tests: use lod / osp tunables on servers
Before the lustre 2.4 OSD work the lov and osc code was used on
both servers and clients. With the OSD layer work we saw the new
lod and osp layers created that are server specific. To avoid
breakage symlinks were created that went from the lod / osp to
lov / osc directories in the proc tree on the server side.
It has been a very long time since that change so we can now
safely start to unwind that handling. The first step taken here
is to migrate the maloo test from using lov / osc for the server
tunables to using lod / osp instead.
Lustre-commit:
c2f43d4c7a609def4292c5b9bee63c9a33cb4598
Reviewed-on: https://review.whamcloud.com/35185
Change-Id: I9dd562cd74d68aaa0226d5ab93042b52193604a1
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35185
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35349
Tested-by: jenkins <devops@whamcloud.com>
Hongchao Zhang [Fri, 29 Mar 2019 13:28:06 +0000 (09:28 -0400)]
LU-11678 quota: make overquota flag for old req
For the old request with over quota flag, the over quota flag
should still be marked at OSC, because the old request could be
processed afther the new request at OST, then it won't break the
quota enforement at OST.
Lustre-change: https://review.whamcloud.com/34645
Lustre-commit:
c59cf862c3c06758c270564dd6e8948e167316b9
Test-Parameters: testlist=replay-single,replay-single,replay-single
Test-Parameters: testlist=replay-single,replay-single,replay-single
Change-Id: Ic34c438fe3f018c3b596b26ad6dc945547c8fada
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Shilong Wang <wshilong@ddn.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34916
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Mon, 29 Jul 2019 07:33:52 +0000 (00:33 -0700)]
LU-10100 llite: swab LOV EA user data
Many sub-tests failed with "Invalid argument" failures
on PPC client because of the endianness issue.
This patch fixes the issue by adding a common function
lustre_swab_lov_user_md() to swab the LOV EA user data.
Test-Parameters: clientarch=ppc64 \
envdefinitions=ONLY=27 testlist=sanity
This patch is back-ported from the following one:
Lustre-commit:
9d17996766e0fa93b1029d2422d45d087edde389
Lustre-change: https://review.whamcloud.com/35291
Change-Id: I46bab0788300cd79c4e66e1a4990c3e1f7192391
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35633
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andriy Skulysh [Wed, 27 Feb 2019 17:37:24 +0000 (19:37 +0200)]
LU-12095 ptlrpc: ocd_connect_flags are wrong during reconnect
Import connect flags are reset to original ones during
reconnect, so a request can be created with unsupported
features.
Use separate obd_connect_data to send connect request.
Lustre-change: https://review.whamcloud.com/34480
Lustre-commit:
1224084c6300d5b15ccb703dfe18209a0f1f12ab
Change-Id: I4cfc48bf7ef66c4f3832613e179030b0eb1d6fdf
Cray-bug-id: LUS-6397
Signed-off-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Alexander Boyko <c17825@cray.com>
Reviewed-by: Andrew Perepechko <c17827@cray.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35635
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Minh Diep [Tue, 23 Jul 2019 00:07:02 +0000 (17:07 -0700)]
LU-12575 build: add ibutils2 for MOFED build
MOFED 4.6 include ibutils2 instead of ibutils
Remove ofed rhel5 patch which we don't need
Lustre-change: https://review.whamcloud.com/35590
Lustre-commit:
26444d84c74e693e47c9785423e63402d32acb4f
Change-Id: I46c51eb8a194ea86bd8c951944e5c1427d0f37d0
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35631
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Alexey Zhuravlev [Mon, 15 Jul 2019 18:01:59 +0000 (21:01 +0300)]
LU-12510 osd: osd-zfs to release zrlock quickly
otherwise few threads trying to access same dnode can get stuck.
this patch is a quick workaround for the issue, it's supposed
to be replaced with a better patch using regular DMU API.
Lustre-commit:
88b329ac2ab568a25251f3f7c3a7e0c7367cb36f
Lustre-change: https://review.whamcloud.com/35524
Change-Id: I24d9ed7f8e68080c6a46409476a80799dbb45230
Signed-off-by: Alexey Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/35600
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shaun Tancheff [Mon, 22 Jul 2019 21:11:57 +0000 (14:11 -0700)]
LU-12479 utils: cleanup gcc8 string warnings
Cleanup some trivial buffer overflows
This patch is back-ported from the following one:
Lustre-commit:
164434eed2161337000257968e37d2714e9c9599
Lustre-change: https://review.whamcloud.com/35354
Test-Parameters: trivial
Cray-bug-id: LUS-6962
Signed-off-by: Shaun Tancheff <stancheff@cray.com>
Change-Id: I17dd1d042c4050e351aca856931ce1107fd5b08f
Reviewed-on: https://review.whamcloud.com/35586
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 14 Dec 2018 22:46:26 +0000 (15:46 -0700)]
LU-8066 lnet: properly isolate kernel_param_ops
Don't reference kernel_param_ops if not available. There is
already a HAVE_KERNEL_PARAM_OPS configure check for this, but
it is just misplaced.
Fixes:
7092309f32516cbfb95a964c87b8030129edeb27
Lustre-change: https://review.whamcloud.com/33874
Lustre-commit:
eb5a6f97af63585bb92ff65ad441a2281041bc47
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I56f86257c6bc8a9c53c7901bc2765e10587cab07
Reviewed-by: Sonia Sharma <sharmaso@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35542
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <stancheff@cray.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shaun Tancheff [Mon, 22 Jul 2019 21:22:16 +0000 (14:22 -0700)]
LU-12478 build: rhel8 missing module packaing tools
On RHEL8 kmodtool and kernel_module_package_buildreqs
are not installed with kernel-devel
This helps to bootstrap the developer into a working configuration.
This patch is back-ported from the following one:
Lustre-commit:
cd282e3e6f1a249546c284d4683bb5bfa5dfdf36
Lustre-change: https://review.whamcloud.com/35356
Test-Parameters: trivial
Cray-bug-id: LUS-7385
Signed-off-by: Shaun Tancheff <stancheff@cray.com>
Change-Id: I30cf5c11174b2aa94c663c72e34f8f88c5c90ed8
Reviewed-on: https://review.whamcloud.com/35587
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Fri, 7 Jun 2019 14:45:26 +0000 (23:45 +0900)]
LU-12401 gss: fix checksum for Kerberos and SSK
When computing checksum for Kerberos, krb5 wire token header is
appended to the plain text. Make sure the actual header is appended
in gss_digest_hash().
For interop with older clients, introduce new server side tunable
'sptlrpc.gss.krb5_allow_old_client_csum'. When not set, servers refuse
Kerberos connection from older clients.
In gss_crypt_generic(), protect against an undefined behavior by
switching from memcpy to memmove.
When computing checksum for SSK, make sure the actual token is used
to store the checksum.
Lustre-change: https://review.whamcloud.com/35099
Lustre-commit:
218fc688c11f081881b2cc1c1632ceaf9ec77a77
Fixes:
a21c13d4df ("LU-8602 gss: Properly port gss to newer crypto api.")
Test-Parameters: envdefinitions=SHARED_KEY=true testlist=sanity,recovery-small,sanity-sec
Test-Parameters: envdefinitions=SHARED_KEY=true clientbuildno=7033 clientjob=lustre-reviews-patchless testlist=sanity,recovery-small,sanity-sec
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I0233ada481f132af112bf88c065f5421902c942e
Reviewed-on: https://review.whamcloud.com/35536
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Wed, 3 Jul 2019 23:30:47 +0000 (16:30 -0700)]
LU-12498 kernel: kernel update [SLES12 SP3 4.4.180-94.97]
Update SLES12 SP3 kernel to 4.4.180-94.97.
Test-Parameters: trivial clientdistro=sles12sp3 serverdistro=sles12sp3
Change-Id: Ia545ff0a54d9cf483a38842f0f7bf42b1fda6875
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35416
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Thu, 28 Mar 2019 07:35:18 +0000 (08:35 +0100)]
LU-12131 tests: fix SSK handling in tests
SSK can be activated for Lustre tests by setting SHARED_KEY env
variable to true.
In setup_all() an additional env variable SK_MOUNTED is used to avoid
mounting an SSK file system twice. But this variable has to be set
back to false in stopall() for consistency.
Some tests are incompatible with SSK, so skip them in case SHARED_KEY
is true. Some other tests playing with nodemaps have to take SSK into
account.
Lustre-change: https://review.whamcloud.com/34521
Lustre-commit:
ee6904e312f7d7446f390cff0ec3c6e48b98e32b
Whamcloud-bug-id: ATM-1283
Test-Parameters: envdefinitions=SHARED_KEY=true testlist=sanity,recovery-small,sanity-sec
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I1016a459c42ffed1ab2b6f67d0a145ed2af9fa40
Reviewed-on: https://review.whamcloud.com/35556
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Andreas Dilger [Fri, 14 Dec 2018 22:49:36 +0000 (15:49 -0700)]
LU-11453 misc: add compat for INIT_LIST_HEAD_RCU
Add a compat version of INIT_LIST_HEAD_RCU() if unavailable.
Fixes:
68bc3984975bb72f730d8a8ab7aa2d836e50abe5
Lustre-change: https://review.whamcloud.com/33875
Lustre-commit:
773e60669b53b0ca2fb48723a21dcddba592af9a
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I9c1296206e002b895bb6bdf55ddd0d8ec70cab07
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35564
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Tue, 16 Jul 2019 07:01:01 +0000 (00:01 -0700)]
LU-12331 llite: create obd_device with usercopy whitelist
Since kernel 4.16 hardened usercopy has been added,
whitelist the struct obd_device to silence the warning.
Bad or missing usercopy whitelist? Kernel memory exposure attempt
detected from SLUB object 'll_obd_dev_cache' (offset 1256, size 40)!
WARNING: CPU: 1 PID: 17534 at mm/usercopy.c:83 usercopy_warn+0x7d/0xa0
Call Trace:
__check_object_size+0xfa/0x181
lmv_iocontrol+0x1146/0x1880 [lmv]
ll_obd_statfs+0x356/0x860 [lustre]
ll_dir_ioctl+0x1e37/0x6760 [lustre]
do_vfs_ioctl+0xa4/0x630
Linux-commit:
8eb8284b412906181357c2b0110d879d5af95e52
This patch is back-ported from the following one:
Lustre-commit:
7f77996b1c4ac3a874a1f9e016e8b0e3cfee6992
Lustre-change: https://review.whamcloud.com/34946
Test-Parameters: trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: Ie863e8a5e2cebd3fd716e7ccc4e0491f83f6fabc
Reviewed-on: https://review.whamcloud.com/35528
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Andreas Dilger [Sat, 29 Jun 2019 01:10:41 +0000 (19:10 -0600)]
LU-12368 obdclass: don't send multiple statfs RPCs
If multiple threads are racing to send a non-cached OST_STATFS or
MDS_STATFS RPC, this can cause a significant RPC storm for systems
with many-core clients and many OSTs due to amplification of the
requests, and the fact that STATFS RPCs are sent asynchronously.
Some logs have shown few 96-core clients have 20k+ OST_STATFS RPCs
in flight concurrently, which can overload the network if many OSTs
are on the same OSS nodes (osc.*.max_rpcs_in_flight is per OST).
This was not previously a significant issue when core counts were
smaller on the clients, or with fewer OSTs per OSS.
If a thread can't use the cached statfs values, limit statfs to one
thread at a time, since the thread(s) would be blocked waiting for
the RPC replies anyway, which can't finish faster if many are sent.
Also add a llite.*.statfs_max_age parameter that can be tuned on
to control the maximum age (in seconds) of the statfs cache. This
can avoid overhead for workloads that are statfs heavy, given that
the filesystem is _probably_ not running out of space this second,
and even so "statfs" does not guarantee space in parallel workloads.
Lustre-change: https://review.whamcloud.com/35380
Lustre-commit:
1c41a6ac390bf74a135861efcd576a3b433d3c49
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I95690e37aecbac08ac5768a5e5c6c70ca258a832
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-on: https://review.whamcloud.com/35485
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Mon, 8 Jul 2019 20:47:40 +0000 (16:47 -0400)]
LU-12491 obdclass: add comment for rcu handling in lu_env_remove
During the review it was pointed out why the RCU lock was dropped
in lu_env_remove() but the code itself doesn't explain why. Add
a comment giving the details why RCU locking is not needed.
Test-parameters: trivial
Lustre-change: https://review.whamcloud.com/35447
Lustre-commit:
709fbe6ee54aa2e601237a6981db3d42a8a719cd
Change-Id: I4fd761d2e1b4adad8e970904d56cdcd057dfe7d5
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Shaun Tancheff <stancheff@cray.com>
Reviewed-by: Neil Brown <neilb@suse.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35488
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Mon, 3 Jun 2019 02:52:42 +0000 (05:52 +0300)]
LU-12491 obdclass: use RCU to release lu_env_item
as rhashtable_lookup_fast() is lockless and can
find just released objects.
Fixes:
aa82cc8361 ("obdclass: put all service's env on the list")
Lustre-change: https://review.whamcloud.com/35038
Lustre-commit:
87306c22e4b977356f4857d5f750447639d89c26
Change-Id: I6ed8ccc5bb5b192eed90b55103d11b822ec90692
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.com>
Reviewed-by: Shaun Tancheff <stancheff@cray.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35487
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andriy Skulysh [Tue, 11 Jun 2019 14:44:32 +0000 (17:44 +0300)]
LU-11518 ptlrpc: don't reset lru_resize on idle reconnect
ptlrpc_disconnect_idle_interpret() clears imp_remote_handle,
so reconnect has pcaa_initial_connect set to 1.
Update only changed ns_connect_flags bits.
Fixes:
5a6ceb664f0 ("LU-7236 ptlrpc: idle connections can disconnect")
Lustre-change: https://review.whamcloud.com/35285
Lustre-commit:
acacc9d9b1d0a869f61d7940baa0700b63dcd8f7
Change-Id: I2368708b6381c1d772c47dc6e61c8fb39a14a2cc
Cray-bug-id: LUS-7471
Signed-off-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Alexander Boyko <c17825@cray.com>
Reviewed-by: Andrew Perepechko <c17827@cray.com>
Reviewed-by: Alexandr Boyko <c17825@cray.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35489
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Jian Yu [Wed, 3 Jul 2019 19:09:40 +0000 (12:09 -0700)]
LU-12456 kernel: kernel update RHEL 8.0 [4.18.0-80.4.2.el8_0]
Update RHEL 8.0 kernel to 4.18.0-80.4.2.el8_0 for Lustre client.
Change-Id: I70527c9317c48c61d0f515f35a1a53fc22ffd06b
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35412
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Nathaniel Clark [Wed, 9 Jan 2019 20:43:59 +0000 (15:43 -0500)]
LU-11845 osd-zfs: Support encrypted ZFS datasets
Call zfs::dmu_objset_own and zfs::dmu_objset_disown with
decrypt=B_TRUE
This is called the same way as in zfs modules.
Fixes:
0fedb017c1 ("LU-9890 osd-zfs: dmu_objset_own/disown changes")
Test-Parameters: envdefinitions=ZFS_MKFS_OPTS="encryption=on -o keylocation=file:///etc/adjtime -o keyformat=passphrase" testlist=sanity fstype=zfs
Lustre-change: https://review.whamcloud.com/33999
Lustre-commit:
3d43658fc36e5821d8b094c1c2365e9520dbe9fe
Signed-off-by: Nathaniel Clark <nclark@whamcloud.com>
Change-Id: I1d9bc1a579ac26706a9f6cc5a0d52649ce005228
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35476
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
James Simmons [Mon, 8 Jul 2019 17:42:47 +0000 (10:42 -0700)]
LU-11893 lnet: consoldate secondary IP address handling
The last piece of code with broken secondary IP address
support is lnet_parse_ip2nets(). We could fix it like
o2iblnd or socklnd was done but since the LND drivers
resolved those issues instead we can move the handling
out of the LND drivers into one place in the LNet core.
To do this we introduce struct lnet_inetdev which is
a collection of data that the current LNet layer requires.
The new function lnet_inet_enumerate() is used to collect
this information.
This patch is back-ported from the following one:
Lustre-commit:
d6d0194c1969db05a6a60718679750ecfd75739b
Lustre-change: https://review.whamcloud.com/34993
Change-Id: I0c532caa3cf6b2178eb1ab65e55e5883d408a185
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/35442
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <hornc@cray.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Wed, 9 Jan 2019 21:35:10 +0000 (14:35 -0700)]
LU-11845 zfs: put configure checks in version order
Put the ZFS feature checks in release version order, so that it is
easier to track when they apply and when they can be removed in
the future.
Make the configure checks use decrypt=B_TRUE just to illustrate
more correct usage.
Lustre-change: https://review.whamcloud.com/34000
Lustre-commit:
55c973ba5cb595d51b4990eea9ad6803e7c0645e
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I22053638d72b41b51b6f56dea5668e78535cab07
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35475
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Alexander Boyko [Tue, 6 Nov 2018 12:57:15 +0000 (07:57 -0500)]
LU-10717 tests: tests should not start mgs
The conf-sanity prolog do reformat_and_config which leaves
mgs service started, if it is not combined.
So, in general, test should not start mgs service, if it don't
stop mgs. And test should start mgs after reformat.
The client mount requires start of all MDTs, because of
MDT0000-osp-MDT000X synchronization.
Lustre-change: https://review.whamcloud.com/33589
Lustre-commit:
aa9f9344fc741523ee8693569effb5b77204f90b
Test-Parameters: trivial testlist=conf-sanity
Test-Parameters: mdscount=2 mdtcount=4 testlist=conf-sanity
Test-Parameters: standalonemgs=true testlist=conf-sanity
Test-Parameters: standalonemgs=true mdscount=2 mdtcount=4 testlist=conf-sanity
Cray-bug-id: LUS-2524
Signed-off-by: Alexander Boyko <c17825@cray.com>
Change-Id: I226eab5683afc36efe908b200f46b710f6235374
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Elena Gryaznova <c17455@cray.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35368
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Minh Diep [Fri, 28 Jun 2019 21:28:39 +0000 (14:28 -0700)]
LU-12395 build: build mpitests for el8
RHEL8 has rpm-mpi-hooks which requires binaries
to be in specific mpi bin to generate the correct
requires
See https://fedoraproject.org//wiki/Changes/RpmMPIReqProv
and https://fedoraproject.org/wiki/Packaging:MPI
Test-Parameters: trivial clientdistro=el8 serverdistro=el7.6 testgroup=regression-mpi
Lustre-change: https://review.whamcloud.com/35374
Lustre-commit:
3c7aca74729edb5339d2b84259ba042bf83c214a
Change-Id: Id9fa50e15b48b9da846083b9e9cd894ad1eac967
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35473
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Wed, 3 Jul 2019 19:28:51 +0000 (12:28 -0700)]
LU-12494 kernel: kernel update SLES12 SP4 [4.12.14-95.19.1]
Update SLES12 SP4 kernel to 4.12.14-95.19.1 for Lustre client.
Test-Parameters: trivial clientdistro=sles12sp4 \
envdefinitions=LNET_SELFTEST_EXCEPT=smoke,SANITY_EXCEPT=103a
Change-Id: I6a101dc2637945192cf8aca661e23c3bccb47609
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35413
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Thu, 6 Jun 2019 02:36:39 +0000 (10:36 +0800)]
LU-12383 utils: only check project inherit bit for dir
Currently, ZFS won't set inherit bit on regular files, but
ext4 always set it, it doesn't make sense for regular files
have this bit, but own it won't do any harm as well.
To make test happy and give a consistent view on users,
let's fix project check only complain erros for Direcotry.
Lustre-change: https://review.whamcloud.com/35076
Lustre-commit:
e4ad5c17c99e7ede5deabffe0bacdd851240eb86
Test-Parameters: trivial testlist=sanity-quota
Change-Id: I194f3ed9d6ded69313a683995295ab8c07b4fb3a
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35393
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Wang Shilong [Fri, 25 Jan 2019 08:54:50 +0000 (16:54 +0800)]
LU-11872 utils: don't follow link files in default
We actually don't support operation on link files itself for now.
As a first step, let's skip link files for now in default,
otherwise, it cause unexpected behavior.
Lustre-change: https://review.whamcloud.com/34111
Lustre-commit:
004b80da5c4b2a7cf4f4885b43c9edec76cd2493
Test-Parameters: trivial testlist=sanity-quota
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: Ib0069ed1982e26984c6cf093f0803bf4a2208fe1
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-on: https://review.whamcloud.com/35392
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Chris Horn [Wed, 5 Jun 2019 00:14:47 +0000 (19:14 -0500)]
LU-12387 utils: Avoid passing symlink to tune_block_dev
In tune_block_dev_slaves we iterate over the directories inside the
slaves subdirectory for the multipath device that is being tuned. For
example:
$ /usr/sbin/l_tunedisk /dev/mapper/mpathc
Suppose mpathc maps to /dev/dm-2. tune_block_dev will initially set
the value of
/sys/devices/virtual/block/dm-2/queue/max_sectors_kb
equal to the value of
/sys/devices/virtual/block/dm-2/queue/max_hw_sectors_kb
Then it looks at the entries in /sys/devices/virtual/block/dm-2/slaves
Suppose the slave devices are as follows:
$ ls /sys/devices/virtual/block/dm-2/slaves
sdc sdh sdm sdr
$
It then calls tune_block_dev recursively, passing
/sys/devices/virtual/block/dm-2/slaves/sdc,
/sys/devices/virtual/block/dm-2/slaves/sdh, etc. However, these are
symlinks that point to directories and as such tune_block_dev will not
tune them because stat does not identify them as block devices.
Instead we should contruct the path argument for these recursive calls
as /dev/<d_name>. In this example, /dev/sdc, /dev/sdh, etc.
Lustre-change: https://review.whamcloud.com/35065
Lustre-commit:
c632a238e6d0c4a3240959a894d36a8a409d64f8
Cray-bug-id: LUS-7358
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I63bc073a82384d68648ff23a56b7d43d6656159b
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35371
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Tue, 4 Jun 2019 19:34:01 +0000 (14:34 -0500)]
LU-12387 utils: Read existing ldd data in l_tunedisk
Read the lustre_disk_data from the device passed to l_tunedisk, so
we can determine whether the device is an MGT or MDT and thus skip
the tuning of the device.
Lustre-change: https://review.whamcloud.com/35066
Lustre-commit:
9cb4f810164efa5f058dc7605fb1835ea51b0a92
Fixes:
892280742a2b ("LU-9551 utils: add l_tunedisk to fix disk tunings")
Fixes:
2f8d7b4679de ("LU-11736 utils: don't set max_sectors_kb on MDT/MGT")
Cray-bug-id: LUS-7358
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I193fe008d5777b0e83f2be9a500eaffb1d3ca615
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35370
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
James Simmons [Mon, 17 Jun 2019 19:25:48 +0000 (12:25 -0700)]
LU-12381 ko2iblnd: ignore down interfaces
The for_each_netdev() loop in kiblnd_create_dev() scans for all
network devices on a system. Currently the code exit when an
network device is down but the device could be something besides
an IB device. Instead of exiting just ignore any device that is
down.
This patch is back-ported from the following one:
Lustre-commit:
1dea5aac9d9be99c4b317a491f308872b97bf0e6
Lustre-change: https://review.whamcloud.com/35098
Test-Parameters: trivial
Fixes:
c4b39bf56bbc ("LU-11893 o2iblnd: add secondary IP address handling")
Change-Id: I0a3bf808d849cd00711b6ef2e4e5bbd876b64903
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Chris Horn <hornc@cray.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35249
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Mon, 17 Jun 2019 19:01:36 +0000 (12:01 -0700)]
LU-11838 osd-ldiskfs: inode times switched to timespec64
Since kernel 4.18 inode times swtich from struct timespec
to timespec64 to make it y2038 safe.
Linux-commit:
95582b00838837fc07e042979320caf917ce3fe6
This patch is back-ported from the following one:
Lustre-commit:
3af55b3159ac2133dc35eeb2f02825848fb65548
Lustre-change: https://review.whamcloud.com/34675
Test-Parameters:trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: Iaddb2f2be27ec348fb97e13371aa3d7e6f6e5c9f
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35247
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Tue, 2 Oct 2018 21:52:28 +0000 (15:52 -0600)]
LU-10885 llite: enable flock mount option by default
The "flock" mount option has been optional for many years, initially
because of potential stability issues, and also to provide a choice
for administrators to select between "flock" and "localflock" options.
However, from the large number of problems that users report when
trying to use applications that depend on this feature (typically
databases and other cloud stacks) that disabling flock by default
causes more problems than it solves.
Enable the "flock" (distributed coherent userspace locking) feature
by default. If applications do not need this functionality, then it
will not affect them. If applications *do* need this functionality,
they will get it. If administrators really know what they are doing,
then they can use the "localflock" feature to enable client-local
flock functionality, possibly only on select nodes that need this.
Users wanting to disable this functionality should mount with the
existing "-o noflock" mount option, or build the client with the
"configure --disable-flock" option.
If clients are already using "-o {flock|localflock|noflock}" then
their existing options will be handled appropriately.
Lustre-change: https://review.whamcloud.com/32091
Lustre-commit:
3613af3e15cbc6091e3a16c8caeb1307be2d91f6
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I182637604fa22573b1da6b6b86d8915e3c3ebbe5
Reviewed-by: Patrick Farrell <paf@cray.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-on: https://review.whamcloud.com/34987
Tested-by: Jenkins
Reviewed-by: Li Xi <lixi@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Lokesh Nagappa Jaliminche [Mon, 4 Jul 2016 09:04:20 +0000 (14:34 +0530)]
LU-8365 ldiskfs: procfs entries for mballoc
Export mballoc streaming block allocator variables
mb_last_group and mb_last_start through procfs.
Lustre-change: https://review.whamcloud.com/21142
Lustre-commit:
75703118588f2b23afd8c8815e5ebb768fc7a8ff
Test-Parameters: testgroup=review-ldiskfs
Change-Id: I5dd00503a81c6819751c9f99b64615b497ef4e28
Cray-bug-id: LUS-3176
Signed-off-by: Lokesh Nagappa Jaliminche <lokesh.jaliminche@seagate.com>
Signed-off-by: Alexander Zarochentsev <c17826@cray.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34842
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Fri, 28 Jun 2019 18:07:19 +0000 (11:07 -0700)]
LU-12458 kernel: kernel update RHEL7.6 [3.10.0-957.21.3.el7]
Update RHEL7.6 kernel to 3.10.0-957.21.3.el7.
Test-Parameters: clientdistro=el7.6 serverdistro=el7.6
Change-Id: I78133d5bc7567d8ea56c4b1aebc3e97096495fad
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35269
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Sun, 7 Apr 2019 03:44:51 +0000 (11:44 +0800)]
LU-12166 test: fix broken detection on ZFS
We intent to run the command on mds, otherwise
project quota will never be tested.
Lustre-change: https://review.whamcloud.com/34609
Lustre-commit:
0f2cd5948b870c0f82a70bdf32f0c5f6d845144d
Test-Parameters:trivial fstype=zfs
Fixes: a046e87 ("LU-7991 quota: project quota against ZFS backend")
Change-Id: I8650a0e1065f0bb465da01556472d3d23b22a530
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34924
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Fri, 21 Jun 2019 06:28:10 +0000 (23:28 -0700)]
LU-12447 utils: specify correct size for lfs project buffer
Enviorment:
Fedora release 28 (Twenty Eight)
gcc (GCC) 8.0.1
20180324 (Red Hat 8.0.1-0.20)
Copyright (C) 2018 Free Software Foundation, Inc.
Hit build failure:
lfs_project.c: In function ‘lfs_project_item_alloc’:
lfs_project.c:72:2: error: ‘strncpy’ specified bound 4096
equals destination size [-Werror=stringop-truncation]
strncpy(lpi->lpi_pathname, pathname, sizeof(lpi->lpi_pathname));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This patch is back-ported from the following one:
Lustre-commit:
ffef6e3271ad1136d3ab1c2ee229b4690a6722a0
Lustre-change: https://review.whamcloud.com/35257
Test-Parameters: trivial testlist=sanity-quota
Change-Id: Ia6429c47391bf503546609ec6a262fe24664bdd4
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-on: https://review.whamcloud.com/35284
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Mon, 17 Jun 2019 19:30:26 +0000 (12:30 -0700)]
LU-12399 tests: avoid 'pdsh localhost' in sanity test_420
sanity test_420 needs a clean env to execute openfile, ie not
inherited from root user.
Replace 'pdsh localhost' with simpler 'su - $uname -c' alternative
to achieve this.
This patch is back-ported from the following one:
Lustre-commit:
1476ac047b449886a0c382b840a7b09dc0cec7eb
Lustre-change: https://review.whamcloud.com/35176
Test-Parameters: trivial envdefinitions=ONLY=420 testlist=sanity
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ifeba7fc1eba86d74a64cca187e286adb23147e2e
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35250
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Mon, 17 Jun 2019 19:21:53 +0000 (12:21 -0700)]
LU-11893 o2iblnd: add secondary IP address handling
Using dev_get_by_name() in kiblnd_create_dev() means we can only
discover primary IP addresses. This breaks using network
aliasing which some people use. Move away from dev_get_by_name()
to using for_ifa() so we can detect any secondary IP addresses.
This patch is back-ported from the following one:
Lustre-commit:
c4b39bf56bbcacd49d7f888a0745cd4b5580b36b
Lustre-change: https://review.whamcloud.com/34476
Change-Id: I03f2f8d18118b716a5eb5fb87694000ac06fe242
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Petros Koutoupis <pkoutoupis@cray.com>
Reviewed-by: Neil Brown <neilb@suse.com>
Reviewed-on: https://review.whamcloud.com/35248
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Wed, 12 Dec 2018 16:19:45 +0000 (11:19 -0500)]
LU-8066 sysfs: make ping sysfs file read and writable
Starting with 4.15 kernels any sysfs read only is limited to
root access only. To retain the ability for non root users
to detect if a remote server is alive using the 'ping' sysfs
file we need to change it to writable. Retain the read ability
so older tools will work.
Lustre-change: https://review.whamcloud.com/33776
Lustre-commit:
6bbae72c6900dbd2b853d716bc4d456dc7fd586e
Change-Id: I6560c119328d723a20a2b32e1fa8c68dce5d407a
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/33776
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-on: https://review.whamcloud.com/35313
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Wang Shilong [Tue, 4 Jun 2019 12:54:01 +0000 (20:54 +0800)]
LU-12382 llite: fix deadloop with tiny write
For a small write(<4K), we will use tiny write and
__generic_file_write_iter() will be called to handle it.
On newer kernel(4.14 etc), the function is exported and will
do something like following:
|->__generic_file_write_iter
|->generic_perform_write()
If iov_iter_count() passed in is 0, generic_write_perform() will
try go to forever loop as bytes copied is always calculated as 0.
The problem is VFS doesn't always skip IO count zero before it comes
to lower layer read/write hook, and we should do it by ourselves.
To fix this problem, always return 0 early if there is no
real any IO needed.
Lustre-change: https://review.whamcloud.com/35058
Lustre-commit:
e9a543b0d3039027423cb469525015f97caa3a3f
Change-Id: I765a723da79eb5fd09317c3fad47fe479b1dd4fb
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35312
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
James Simmons [Tue, 25 Jun 2019 13:29:07 +0000 (09:29 -0400)]
LU-8066 utils: have llapi_target_iterate use sysfs tree
Update llapi_target_iterate() to not use 'devices' but collect the
data from the lustre sysfs tree itself.
Lustre-change: https://review.whamcloud.com/33799
Lustre-commit:
b24d69492b818457d9da0d6dce3adc0f91f18ec6
Change-Id: If100b4918bdcc8b24e72f37127048a32a808310f
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34781
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Ben Menadue [Tue, 11 Jun 2019 03:38:18 +0000 (20:38 -0700)]
LU-12269 build: fix hardened builds in rpm spec file
The hardened build configure on RHEL8 has a quoted string
with spaces in it, and this breaks the construction of
%eval_configure on lustre.spec.in - the quotes end up in
the wrong place.
Moreover, the hardened build flags are only for user-space
code, and breaks kernel code compilation on RHEL 8.0 (it
adds -fPIE, which isn't valid for kernel code.
This patch stores the %build_cflags and %build_ldflags from
rpmbuild as environment variables before turning hardened
build off to allow the kernel code to build. These
environment variables are used in the lnet/utils and
lustre/utils Makefiles so that the user-space code there
gets the benefit of any system-specific RPM build flag
(such as hardened builds).
For RHEL7 on PPC64 we then also need to define the C macro
__SANE_USERSPACE_TYPES__ so that __s64 and __u64 are long
long instead of the default long - otherwise the build will
fail with a format string error on this platform because
Lustre uses %ll when printing/scanning __s64/__u64.
The environment variables (UTILS_CFLAGS and UTILS_LDFLAGS)
could also be used for a standalone, non-RPM build to pass
flags to the user-space code, with the usual CFLAGS and
LDFLAGS still used for kernel code.
This patch is back-ported from the following one:
Lustre-commit:
5270583ae6e436e9e7ae0199312e7f50365744af
Lustre-change: https://review.whamcloud.com/34882
Signed-off-by: Ben Menadue <ben.menadue@anu.edu.au>
Change-Id: I9b4ba830bf63838fd88ef1bae5dd10dff2109a1d
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/35161
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Fri, 14 Jun 2019 04:52:06 +0000 (21:52 -0700)]
LU-11742 test: have libtool execute the test binaries
With the move to libtools the ability to run all the lustre
utilities form the source tree was lost. To work around this
the libtool -no-install flag was used to prevent the creation
of the libtool wrappers. While this worked to restore the
source tree sand box development new package breakage is showing.
This is due to the rpath being hard coded into the utilies when
-no-install is used and some platforms disable fixed rpaths.
A very similar problem exist for people who want to use gdb to
debug their projects application. gdb does not work on libtool
wrappers as well so the recommended approach to this type of
problem is to use the libtool execute command. This command
allows the execution of an external non project binary, like
gdb, with the projects real binary application. Apply this
approach to the lustre test suite so commands like kill can
be used to shutdown lustre utilies that are not installed into
the testing environment.
Lustre-change: https://review.whamcloud.com/33947
Lustre-commit:
f9e5224fbb60bb8b44753b7be10cb06108627f89
Change-Id: I74112f7250f1c43313d868c0edc7c8815d373002
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Chris Horn <hornc@cray.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34583
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Fri, 14 Jun 2019 04:43:47 +0000 (21:43 -0700)]
LU-12195 tests: use sleep instead of wrapped multiop
in sanity/43* and sanity/14* tests as multiop is not a binary,
but libtool-wrapped script. the tests fail when started from a
build tree.
Lustre-commit:
9a1f327a76f72c7713e53d8b354ff7f0e32be870
Lustre-change: https://review.whamcloud.com/34721
LU-12261 tests: Race between exec and truncate
Execing '$tdir/sleep' with & doesn't guarantee the file is
actually open before returning, so it is sometimes losing
the race with truncate, resulting in errors like this:
/usr/lib64/lustre/tests/sanity.sh: line 4172:
/mnt/lustre/d43b.sanity/sleep: Text file busy
Where $tdir/sleep gets ETXTBSY, instead of truncate as
expected.
A 1 second delay should be enough to guarantee exec wins
the race vs truncate.
Test-Parameters: trivial
Test-Parameters: testgroup=review-ldiskfs-arm
Test-Parameters: testgroup=review-ldiskfs
Test-Parameters: testgroup=review-ldiskfs-arm
Lustre-commit:
c64855fca1504bddcb0fc7ad7316d8d6b20a9c6f
Lustre-change: https://review.whamcloud.com/34791
Change-Id: Iaec3433f03aab23583052373e5f0252d9eac7f04
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34955
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Tue, 11 Jun 2019 07:49:50 +0000 (00:49 -0700)]
LU-11893 ksocklnd: add secondary IP address handling
With ksocknal_enumerate_interfaces() use of for_primary_ifa() only
primary IP addresses are returned. This disables using network
aliasing which some people use. Change for_primary_ifa() to
for_ifa() so we can detect any secondary IP addresses. Update the
string handling since ifa_device names can be different than the
net_device name. Discard the 'j' counter and instead keep
ksnn_ninterfaces up to date. This measn that we return 0 on
sucess, rather than a count of added interfaces. Update the
too many interfaces test in ksocknal_enumerate_interfaces()
with a better test using ARRAY_SIZE.
This patch is back-ported from the following one:
Lustre-commit:
9a2013af0668737dc56424c5c6eaac01621f6c17
Lustre-change: https://review.whamcloud.com/34392
Change-Id: I832df89148def5088502ac92df27b8b3872f3792
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Petros Koutoupis <pkoutoupis@cray.com>
Reviewed-by: Neil Brown <neilb@suse.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35159
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
NeilBrown [Tue, 11 Jun 2019 05:37:58 +0000 (22:37 -0700)]
LU-11838 socklnd: use for_each_netdev() instead of lnet_ipif_enumerate()
for_each_netdev() is a more direct interface and doesn't require
library support.
Also get the ip address directly from the net_device, rather than
using lnet_ipif_query().
Linux-commit:
f703f71afd98e6e7ec70f92ffc52ef3ffffcd849
Linux-commit:
9eb957b98aa6322abde33240bf50dd483c5d1190
This patch is back-ported from the following one:
Lustre-commit:
e9d9cbb072956f2582c97263184aecd196bba14a
Lustre-change: https://review.whamcloud.com/33966
Change-Id: I82894991b9a4a250d0560af31325b6c765cc0620
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Sonia Sharma <sharmaso@whamcloud.com>
Reviewed-by: Doug Oucharek <dougso@me.com>
Reviewed-on: https://review.whamcloud.com/35158
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Tue, 11 Jun 2019 05:26:42 +0000 (22:26 -0700)]
LU-11838 kernel: harden current_time autoconf test
In newer kernels CURRENT_TIME was replaced by current_time(). The
return value of current_time() was struct timespec but to support
time after 2038 the return value was changed to struct timespec64.
This change broke the autoconf test. The solution is to use one
of the struct iattr field in the autoconf test since it hides
the return value type.
Test-Parameters: trivial
This patch is back-ported from the following one:
Lustre-commit:
74b3726f42b1f72e289e3c3252030a62646afa7b
Lustre-change: https://review.whamcloud.com/33963
Change-Id: I95abd2cd2b777f99cbf6ab78370ee2171e5fca67
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-on: https://review.whamcloud.com/35170
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mikhail Pershin [Wed, 31 Oct 2018 13:28:29 +0000 (16:28 +0300)]
LU-11359 mdt: fix mdt_dom_discard_data() timeouts
The mdt_dom_discard_data() issues new lock to cause data
discard for all conflicting client locks. This was done in
context of unlink RPC processing and may cause it to be stuck
waiting for client to cancel their locks leading to cascading
timeouts for any other locks waiting on the same resource and
parent directory.
Patch skips discard lock waiting in the current context by
using own CP callback for that which doesn't wait for blocking
locks. They will be finished later by LDLM and cleaned up in
that completion callback. So current thread just makes sure
discard locks are taken and BL ASTs are sent but doesnt't wait
for lock granting and that fixes the original problem.
At the same time that opens window for race with data being
flushed on client, so it is possible that new IO from client
will happen on just unlinked object causing error message and
it is not possible to distinguish that case from other
possibly critical situations. To solve that the unlinked object
is pinned in memory while until discard lock is granted.
Therefore, such objects can be easily distinguished as stale one
and any IO against it can be just silently ignored.
Older clients are not fully compatible with async DoM discard so
patch adds also new connection flag ASYNC_DISCARD to distinguish
old clients and use old blocking discard for then.
Lustre-change: https://review.whamcloud.com/34071
Lustre-commit:
9c028e74c2202a8a481557c4cb22225734aaf19f
Test-Parameters: testlist=racer,racer,racer
Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: I419677af43c33e365a246fe12205b506209deace
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35197
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Stephan Thiell <sthiell@stanford.edu>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
NeilBrown [Tue, 11 Jun 2019 07:52:04 +0000 (00:52 -0700)]
LU-11838: lnet: remove lnet_ipif_enumerate()
Also remove lnet_ipif_query() and related functions.
There are no longer any users of these functions, so remove them.
Linux-commit:
6e659fcfab0cdd876a555a752acf9997f98acbcd
This patch is back-ported from the following one:
Lustre-commit:
dedd3706945ef759d7d645cde30fa488c8ced4a1
Lustre-change: https://review.whamcloud.com/34234
Change-Id: I8183e505e3dbe12ff71ddf38f5b18a945d8a4a6c
Signed-off-by: NeilBrown <neilb@suse.com>
Reviewed-by: Olaf Weber <olaf.weber@hpe.com>
Reviewed-by: Petros Koutoupis <pkoutoupis@cray.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-by: Aurelien Degremont <degremoa@amazon.com>
Reviewed-on: https://review.whamcloud.com/35160
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Tue, 11 Jun 2019 07:47:14 +0000 (00:47 -0700)]
LU-12270 o2iblnd: pci_unmap_addr() removed in 4.19
Since kernel 4.19 the pci_unmap_addr() wrappers have
been removed, along with linux/pci-dma.h
We can use the good old DEFINE_DMA_UNMAP_ADDR instead
of DECLARE_PCI_UNMAP_ADDR.
Linux-commit:
18b01b16e8bae9cd227909f6e6d2783d74855f65
This patch is back-ported from the following one:
Lustre-commit:
0cae491cc6d3cc949972366a3fdfdf32dfea5912
Lustre-change: https://review.whamcloud.com/34827
Test-Parameters:trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: I387bd3d1c4e8c3bc75400ce1be05132fb25f8a50
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Chris Horn <hornc@cray.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35157
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Tue, 11 Jun 2019 05:55:30 +0000 (22:55 -0700)]
LU-11838 llite: address_space ->page_tree renamed ->i_pages
kernel 4.17 renamed address_space renamed ->page_tree to ->i_pages,
and switched to xa_lock on the radix_tree_root.
Linux-commit:
b93b016313b3ba8003c3b8bb71f569af91f19fc7
This patch is back-ported from the following one:
Lustre-commit:
2d0c621d21be4e67b6075b76017af6e6fcd18c64
Lustre-change: https://review.whamcloud.com/34673
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: Iadbc5eda884dbe8ad0d694e0f88255bc496dea5b
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35156
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Li Dongyang [Tue, 11 Jun 2019 05:52:12 +0000 (22:52 -0700)]
LU-11838 ldlm: struct timespec64.tv_sec type change
Since kernel 4.18 struct timespec64 is no longer defined
as struct timespec on 64bit systems, this means tv_sec
is no longer __kernel_time_t but now time64_t.
Use %llu as the format specifier and explicitly cast it
to unsigned long long.
This patch is back-ported from the following one:
Lustre-commit:
f2bf0379a773c8c1659bfe018a22861784a0b9a6
Lustre-change: https://review.whamcloud.com/34677
Test-Parameters:trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: Ib4c80c9b20854d45b1b3c04057c45ee20d5413d9
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35175
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Tue, 11 Jun 2019 05:46:42 +0000 (22:46 -0700)]
LU-11838 osp: atomic64_read() returns s64
Since kernel 4.17 atomic64_read on x86_64 returns s64
instead of long.
Use %llu as the format specifier and explicitly cast it
to unsigned long long.
This patch is back-ported from the following one:
Lustre-commit:
dc46952ecd1aa09e738b2de6b1a3076ecbaa740e
Lustre-change: https://review.whamcloud.com/34676
Test-Parameters:trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: I805d43251f24417e6405f5d087927c15cf531619
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35174
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Tue, 11 Jun 2019 05:43:55 +0000 (22:43 -0700)]
LU-11838 lnet: getname dropping addrlen argument
Since kernel 4.17 ->getname() does not take int *addrlen
argument anymore, instead it's returning the length to
the caller.
Linux-commit:
9b2c45d479d0fb8647c9e83359df69162b5fbe5f
This patch is back-ported from the following one:
Lustre-commit:
dbb81e826290b2db27e24a85869c9d0736726caa
Lustre-change: https://review.whamcloud.com/34672
Test-Parameters:trivial
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: I4ad5de4a22f3fb23c07a356650ea7925acf07eed
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/35173
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Tue, 11 Jun 2019 05:40:39 +0000 (22:40 -0700)]
LU-11838 llite: remove assert for acl refcount
The purpose of this asssert to was to ensure lustre
was properly managing its posix_acl access. This test
is invalid due to the VFS layer also taking references
on the posix_acl. In reality their is no simple way to
detect this class of mistakes.
* lastest kernels remove this refcount *
Linux-commit:
6a42e615a28bad49f2e04829486e94190c066390
This patch is back-ported from the following one:
Lustre-commit:
df7bfbb1c7890deed15fd85e75da70d88be2ef7f
Lustre-change: https://review.whamcloud.com/34236
Change-Id: I167f2de449a2e8357517f33c2e81a25b25104d57
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35172
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
NeilBrown [Tue, 11 Jun 2019 05:33:43 +0000 (22:33 -0700)]
LU-11838 o2iblnd: get IP address more directly.
Use dev_get_by_name() and for_primary_ifa() to
get IP address for a named device. This is more
direct.
Linux-commit:
10e138e41a4343fd1a88e4543990205d134e562a
Linux-commit:
9eb957b98aa6322abde33240bf50dd483c5d1190
This patch is back-ported from the following one:
Lustre-commit:
7a40cd2c83d174ae0bb7e22d62fad9fbd247a654
Lustre-change: https://review.whamcloud.com/33970
Change-Id: Ic4562c3948934bacb8613e9f6f57f609ecc04de7
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Sonia Sharma <sharmaso@whamcloud.com>
Reviewed-by: Doug Oucharek <dougso@me.com>
Reviewed-on: https://review.whamcloud.com/35166
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
NeilBrown [Tue, 11 Jun 2019 05:29:31 +0000 (22:29 -0700)]
LU-11838 lnet: change lnet_ipaddr_enumerate() to use for_each_netdev()
for_each_netdev() is a more direct interface than
lnet_ipif_enumerate(), so use it instead. Also get
address and 'up' status directly from the device.
This means we need to possible re-allocate the storage
space if there are lots of IP addresses.
However there is no need to resize the allocation down if we
over-allocated. This is only used once, and is freed soon
after it is allocated, so that is a false optimization.
Linux-commit:
0400cf406c32ac3968241cd528747d922b6c55c3
This patch is back-ported from the following one:
Lustre-commit:
f5991afd8779fe747778e28e998277a10242a57d
Lustre-change: https://review.whamcloud.com/33969
Change-Id: I1c1e7722c7b2b267dcb8134ae295a54f976d96ad
Signed-off-by: NeilBrown <neilb@suse.com>
Reviewed-by: Doug Oucharek <dougso@me.com>
Reviewed-by: Sonia Sharma <sharmaso@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35171
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
NeilBrown [Tue, 11 Jun 2019 05:23:20 +0000 (22:23 -0700)]
LU-11838 lustre: discard LTIME_S macro
Rather than using a macro, just access the required
field directly.
Linux-commit:
5b7cc4e4ce3cc15f67462ae75c55eecc7edc3a40
This patch is back-ported from the following one:
Lustre-commit:
65a8ff5fbe8ca014bd01150ab102d8aa43f78cff
Lustre-change: https://review.whamcloud.com/33984
Change-Id: I325cac7458265d1cf6ad9f195a513f2612865906
Signed-off-by: NeilBrown <neilb@suse.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-on: https://review.whamcloud.com/35169
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Ben Menadue [Tue, 11 Jun 2019 03:40:15 +0000 (20:40 -0700)]
LU-12269 build: remove %{fullrelease} from Provides
Commit 7532409 adds a version number to lustre-osd-mount
Provides lines in lustre.spec.in, but include the
%{fullrelease} macro that was previously removed by
28c17d4. This causes an "unexpanded macro" warning when
building the RPM, and the result contains a bogus string
for that name, e.g.
2.12.53_45_g43fc4db-%{fullrelease}
This patch simply removes the "-%{fullrelease}" suffix from
these lines in lustre.spec.in.
This patch is back-ported from the following one:
Lustre-commit:
48e16113731a1a9fff06370a0a11ea083ad290b8
Lustre-change: https://review.whamcloud.com/34883
Test-Parameters: trivial
Signed-off-by: Ben Menadue <ben.menadue@anu.edu.au>
Change-Id: Ia13f339f57b89c02443ebc2d68f0aa3b0802319a
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/35162
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Ben Menadue [Tue, 11 Jun 2019 03:42:50 +0000 (20:42 -0700)]
LU-12269 build: add value to definition of with_gss in spec
rpmbuild currently fails when gss_keyring is enabled (which
happens automatically if the right packages are installed).
This is due to an ill-formed %define in lustre.spec.in that
doesn't include the value to set the macro do.
This patch updates this line to set the value to 1.
This patch is back-ported from the following one:
Lustre-commit:
5bdf89f0c13eab1513d11b2e1950fba31479535d
Lustre-change: https://review.whamcloud.com/34892
Signed-off-by: Ben Menadue <ben.menadue@anu.edu.au>
Change-Id: I2f52b19795091702622eb3b4c110f09eb80654db
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/35163
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Tue, 11 Jun 2019 03:48:26 +0000 (20:48 -0700)]
LU-12269 kernel: new kernel [RHEL 8.0 4.18.0-80.el8]
This patch makes changes to support new RHEL 8.0 release
for Lustre client.
Test-Parameters: trivial
This patch is back-ported from the following one:
Lustre-commit:
d37b0ab99eaeeac391088848c275d2757b6ff17d
Lustre-change: https://review.whamcloud.com/34862
Change-Id: I89b4f1e59f8b25bf9d37d3564e2d05d6e87d9b38
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/35164
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Tue, 22 Jan 2019 08:39:21 +0000 (16:39 +0800)]
LU-11678 quota: protect quota flags at OSC
There is no protection in OSC quota hash tracking the quota flags of
different qid, which could cause the previous request to modify the
quota flags which was set by the current request because the replies
could be out of order.
This patch also adds a lock to protect the operations on the quota
hash from different requests.
Test-Parameters: testlist=sanity-quota,sanity-quota,sanity-quota,sanity-quota
Lustre-change: https://review.whamcloud.com/33747
Lustre-commit:
77d9f4e05a5c366ad0f7c2e97a338c6958676f73
Change-Id: Ia6e5141265beacb9401dd533081fa0b85fd5ea6a
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34915
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Sat, 22 Dec 2018 22:21:22 +0000 (17:21 -0500)]
LU-11768 test: reset qsd_time before test
In test_6 of sanity-quota, if the qsd_timeout is larger than
TIMEOUT*2, it will trigger the watchdog and cause the test fail.
Lustre-change: https://review.whamcloud.com/33931
Lustre-commit:
6cb284c2ed92284454580e6e54e00ddc33530c6e
Change-Id: I3f2993ce2b88e1520b6907ae134557abcd30aa0c
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35212
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Nathaniel Clark [Sun, 2 Jun 2019 13:50:53 +0000 (09:50 -0400)]
LU-12375 scripts: Start lnet after opa
Ensure ordering of lnet after opa for startup and lnet before opa on
shutdown.
Lustre-change: https://review.whamcloud.com/35032
Lustre-commit:
54adf45f2c73e542a2fcf3125781de791fac9c56
Signed-off-by: Nathaniel Clark <nclark@whamcloud.com>
Change-Id: I4c2cad2381349f866bdc08e2a81e3d8990d8752e
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Artur Novik <anovik@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35260
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Li Dongyang [Mon, 17 Jun 2019 19:36:11 +0000 (12:36 -0700)]
LU-12395 build: require python2 for lustre-iokit
RHEL8 has splitting python2 and python3 rpms,
and none of them provdes python anymore.
We can just require python2 in the spec, other
distros all have python rpm providing python2.
This patch is back-ported from the following one:
Lustre-commit:
6f0fcc289887737f45687b5f8a15835aeab32ef4
Lustre-change: https://review.whamcloud.com/35094
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: I881c90a4e66d1a431d11d16b9e89781de2f87a7d
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35251
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Thu, 11 Apr 2019 00:40:23 +0000 (08:40 +0800)]
LU-11355 lustre: enable fstrim on lustre device
pass the FITRIM ioctl through the OST/MDT
mountpoint to the underlying filesystem, which
allows us to run fstrim on server mount point directly.
Lustre-change: https://review.whamcloud.com/33131
Lustre-commit:
d5be104cc9b0c7a71b30aa5feb16873aa30803a9
Change-Id: Ia6f9b43e48245ee7907a47f05c3924b3640bc734
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Tested-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35196
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Wed, 3 Apr 2019 08:29:06 +0000 (11:29 +0300)]
LU-12034 obdclass: put all service's env on the list
to be able to lookup by current thread where it's too
complicated to pass env by argument.
this version has stats to see slow/fast lookups. so, in sanity-benchmark
there were 172850 fast lookups (from per-cpu cache) and 27228 slow lookups
(from rhashtable). going to see the ration in autotest's reports.
Fixes:
2339e1b3b690 ("LU-11483 ldlm ofd_lvbo_init() and mdt_lvbo_fill() create env")
Fixes:
e02cb40761ff ("LU-11164 ldlm: pass env to lvbo methods")
Lustre-change: https://review.whamcloud.com/34566
Lustre-commit:
aa82cc83612dbd4c7d05fc101b98e8660f1373db
Change-Id: Ia760e10fa5c68e7a18284e4726d215b330fc0eed
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andrew Perepechko <c17827@cray.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35110
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Arshad Hussain [Fri, 10 May 2019 00:54:03 +0000 (06:24 +0530)]
LU-12314 tests: Add Missing Description to sanity test 258a
This patch adds missing test description to sanity test 258a.
This patch is a back port from master:
Lustre-change: https://review.whamcloud.com/34902
Lustre-commit:
ef2a05a61b4201af612356f78197a57b427260b2
Test-Parameters: trivial
Signed-off-by: Arshad Hussain <arshad.super@gmail.com>
Change-Id: I972549cd049b965c9e6da9b43aa245bab875a77a
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-on: https://review.whamcloud.com/35245
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Andrew Perepechko [Thu, 7 Mar 2019 20:18:45 +0000 (12:18 -0800)]
LU-8047 llite: optimizations for not granted lock processing
This patch removes ll_md_blocking_ast() processing for
not granted locks. The reason is ll_invalidate_negative_children()
can slow down I/O significantly without a reason if there
are thousands or millions of files in the directory
cache.
Lustre-change: https://review.whamcloud.com/19665
Lustre-commit:
2c126c5a73edea434456c6c335772daaac717f2f
Change-Id: Ic69c5f02f71c14db4b9609677d102dd2993f4feb
Seagate-bug-id: MRP-3409
Signed-off-by: Andrew Perepechko <c17827@cray.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35001
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Olaf Faaland [Mon, 6 May 2019 18:31:21 +0000 (11:31 -0700)]
LU-11946 build: no zlib check during configure --enable-dist
If the zlib libraries are not found, the error is fatal, and prevents
the sources from being packaged.
This check is unnecessary when sources are being packaged, so this patch
disables the test when configure is run with --enable-dist.
Lustre-change: https://review.whamcloud.com/34811
Lustre-commit:
e9d9a0eeb0ff71b207e9d0095fce16af74f842b6
Change-Id: Ie262b17b63c0edc8e8bfbd0c1a466ec37d05622c
Signed-off-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35194
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Mon, 11 Feb 2019 16:20:04 +0000 (11:20 -0500)]
LU-8130 libcfs: support latest rhashtable API
With the broad support range of the OpenSFS lustre version pieces
are missing in some distributions to properly support using the
rhashtable API as required by Lustre.
Lustre-change: https://review.whamcloud.com/34036
Lustre-commit:
8de7221201c0707245c9ee2ef7cdd1d54207b3ee
Change-Id: I7ce2949ca2f1d497dcb60a8b17b964e47cdff223
Test-Parameters: trivial
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35143
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Wang Shilong [Thu, 30 May 2019 14:46:09 +0000 (22:46 +0800)]
LU-12361 lov: fix wrong calculated length for fiemap
lov_stripe_intersects() will return a closed interval
[@obd_start, @obd_end], so to calcuate length of interval we need
@obd_end - @obd_start + 1
rather than
@obd_end - @obd_start
Wrong extent length will make us return wrong fiemap information.
Lustre-change: https://review.whamcloud.com/34998
Lustre-commit:
225e7b8c70fb68bc3aa3a6d88c5e9bda322c9cc9
Change-Id: I30deb17cf5fa80a6d3046098fbac0d3faa01ad1c
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35083
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Nikitas Angelinas [Tue, 11 Dec 2018 04:49:52 +0000 (20:49 -0800)]
LU-11755 ldlm: remove LASSERT() in ns_is_{client,server}()
(ns->ns_client == LDLM_NAMESPACE_CLIENT ||
ns->ns_client == LDLM_NAMESPACE_SERVER) implies that
(!(ns->ns_client & ~(LDLM_NAMESPACE_CLIENT |LDLM_NAMESPACE_SERVER)))
is also true, so the latter LASSERT() can be removed from
ns_is_{client,server}().
Lustre-change: https://review.whamcloud.com/33822
Lustre-commit:
0af247b17619f586702b0ad37f0796bdf783b11c
Test-Parameters: trivial
Signed-off-by: Nikitas Angelinas <nangelinas@cray.com>
Cray-bug-id: LUS-6762
Change-Id: Ib165d97ca397d5109c046415d2ec513357b07cbd
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alexandr Boyko <c17825@cray.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35193
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Chris Horn [Wed, 22 May 2019 16:21:14 +0000 (11:21 -0500)]
LU-12323 libcfs: check if save_stack_trace_tsk is exported
Lustre 2.12 commit
afedf9343686504c89f2e28cf6133540166f2347 introduced
the use of save_stack_trace_tsk, but this symbol is not exported for
all architectures. When it's possible we can use save_stack_trace
instead. Otherwise skip printing stack trace.
Lustre-change: https://review.whamcloud.com/34937
Lustre-commit:
ffb2b46ed7eda42530596df3d52f76250d53e506
Cray-bug-id: LUS-7352
Test-Parameters: clientarch=aarch64
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I142b542f5c5672abbad461a621aedd1e49db1bdd
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Shaun Tancheff <stancheff@cray.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35085
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Olaf Faaland [Mon, 6 May 2019 18:38:37 +0000 (11:38 -0700)]
LU-11946 build: no yaml check during configure --enable-dist
If the yaml libraries are not found, the error is fatal, and prevents
the sources from being packaged.
This check is unnecessary when sources are being packaged, so this patch
disables the test when configure is run with --enable-dist.
Lustre-change: https://review.whamcloud.com/34812
Lustre-commit:
748ede308d0eb2bb2762cae9c45978808e20a735
Change-Id: I160e0d54efc59480d2f830607467dbc9f34c9de3
Signed-off-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35088
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andriy Skulysh [Wed, 30 Jan 2019 09:33:49 +0000 (11:33 +0200)]
LU-11925 hsm: attributes aren't updated after RESTORE
MDS returns file size to a client with UPDATE lock
while file is RELEASED. It isn't cancelled after RESTORE
and the client has old file size after appending data.
Flush update lock after RESTORE completed.
Lustre-change: https://review.whamcloud.com/34180
Lustre-commit:
63493381b3a1a1ebd5293df69403ee2ff8b0440e
Change-Id: Ib956dbd075691ce5fac1ce552df9519f9fa768e4
Cray-bug-id: LUS-6945
Signed-off-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-by: Sergey Cheremencev <c17829@cray.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35082
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Sat, 1 Jun 2019 11:22:11 +0000 (19:22 +0800)]
LU-12374 lustre: push rcu_barrier() before destroying slab
From rcubarrier.txt:
"
We could try placing a synchronize_rcu() in the module-exit code path,
but this is not sufficient. Although synchronize_rcu() does wait for a
grace period to elapse, it does not wait for the callbacks to complete.
One might be tempted to try several back-to-back synchronize_rcu()
calls, but this is still not guaranteed to work. If there is a very
heavy RCU-callback load, then some of the callbacks might be deferred
in order to allow other processing to proceed. Such deferral is required
in realtime kernels in order to avoid excessive scheduling latencies.
We instead need the rcu_barrier() primitive. This primitive is similar
to synchronize_rcu(), but instead of waiting solely for a grace
period to elapse, it also waits for all outstanding RCU callbacks to
complete. Pseudo-code using rcu_barrier() is as follows:
1. Prevent any new RCU callbacks from being posted.
2. Execute rcu_barrier().
3. Allow the module to be unloaded.
"
So use synchronize_rcu() in ldlm_exit() is not safe enough, and we might
still hit use-after-free problem, also we missed rcu_barrier() when destory
inode cache, this is simiar idea what current local filesystem does.
Lustre-change: https://review.whamcloud.com/35030
Lustre-commit:
1f7613968c800f99ed074f17cd7ba1086847d2db
Change-Id: I76c7dfe7b6472d377fe1b60b0891c61ac8a0fbfc
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-on: https://review.whamcloud.com/35097
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Patrick Farrell [Tue, 28 May 2019 21:02:49 +0000 (17:02 -0400)]
LU-12350 tests: Do not use background failover
For some reason, test 33 chooses at one point to take an
OST offline by starting failover in the background. It
seems to assume the OST will be offline during the
subsequent read, without doing anything to verify it is
offline - In fact, it could either be not offline yet or
back online with failover complete.
Just use stop like the rest of the test does.
Lustre-change: https://review.whamcloud.com/34985
Lustre-commit:
4ac0324fb9d824915b3dd11b75e81e609d9e8e84
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: I9c074ff1412793b8f0d8f15dc1e2ee21bb6d9fd6
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35086
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mikhail Pershin [Thu, 30 May 2019 09:30:43 +0000 (12:30 +0300)]
LU-12120 grants: prevent negative ted_grant value
Add check in tgt_grant_shrink() to protect ted_grant
against negative value.
Lustre-change: https://review.whamcloud.com/34996
Lustre-commit:
7e08317ef5cbed5cd587017cbe343eb4cc52822c
Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: Iddea86f052124413ac60f5d0f26bcb68e376ede5
Reviewed-by: Quentin Bouget <quentin.bouget@cea.fr>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35084
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Götz Waschk [Tue, 28 May 2019 06:48:02 +0000 (08:48 +0200)]
LU-12342 spec: mark lsvcgss as a config file in the rpm
The file /etc/sysconfig/lsvcgss shouldn't be overwritten on package
upgrades.
Lustre-change: https://review.whamcloud.com/34978
Lustre-commit:
cfe9e1a56c696bcbba24dd4041845ead12aba291
Signed-off-by: Götz Waschk <goetz.waschk@desy.de>
Change-Id: I3fa0a3a5a06d9e59699d23e652329365f38fd028
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35087
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Hongchao Zhang [Tue, 2 Apr 2019 02:49:53 +0000 (22:49 -0400)]
LU-12232 test: commit before df
In sub_test6 of replay_ost_single, the transactions at OSTs should
be committed to cleanup the test environment.
Lustre-change: https://review.whamcloud.com/34808
Lustre-commit:
f1cbfb96c820aa7e1e5a84176619679d696a117a
Test-Parameters: trivial
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Change-Id: Icbb06789855ab02252b7f1b0b9aff6bbb0f5f2e1
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34927
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Mon, 29 Apr 2019 13:13:59 +0000 (21:13 +0800)]
LU-12225 obdclass: improve jobid memory reclaim policy
jobid_should_free_item() will be called in following three
cases to decide whether @pidmap should be deleted from hash list:
1) expire normal timeout and memory reclaimer called to
try free some items.
2) admin echo sys interface to free some jobid.
3) Umount client to free all memory.
For case 2 && 3, it makes sense we always return 1,
add a warn_on in case3 to make sure there isn't any
bug in the codes.
For the case1, we could change policy a bit to not
return 1 if reference count of @pidmap is larger than 1,
a common case is a newly added @pidmap is easily freed
from hash list with current policy.
Actually, even we delete @pidmap from hash list, memory
will be eventually freed with its references count reached
1, and it is very likely we deleted and inserted @pidmap
again since this could be a hot and long runtime job.
Lustre-change: https://review.whamcloud.com/34775
Lustre-commit:
3e9fedfa7ea52a03a1975572ab37cc1ae9344a8a
Change-Id: I61b894a900319953d5a3369bee69bda050102129
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Ben Evans <bevans@cray.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35008
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Wed, 22 May 2019 05:22:49 +0000 (01:22 -0400)]
LU-12324 mdd: Do not record xattr size get in changelogs
It looks like if the xattr itself was not fetched there's no
need to create a changelog entry for it. The real get will come
and we'd do it there
Lustre-change: https://review.whamcloud.com/34936
Lustre-commit:
8f2599c599011a50549cdc79f0b3379239cf4f0c
Change-Id: I5b19f9309f65da0a4c58cb79a95787dab862eb94
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35010
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Wang Shilong [Mon, 29 Apr 2019 12:46:47 +0000 (20:46 +0800)]
LU-12225 obdclass: fix race access vs removal of jobid_hash
We added @pidmap into hash and reference count will be 1.
However, another thread might reclaim this newely added
@pidmap from hash list, we try to access this @pidmap
will become a user-after-free operation.
Fix this problem by init reference count as 1 before
adding hash list, which gurantee memory could be not
freed during our access.
Check other places where memory reclaim used did similar
idea like this.
Lustre-change: https://review.whamcloud.com/34763
Lustre-commit:
b664182e0361731fa409ac6a0a0f19637a7e5288
Change-Id: Idd5f429b97e064e29b6883243f8a012c2b4b4ae7
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35007
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Mon, 15 Apr 2019 12:58:59 +0000 (15:58 +0300)]
LU-11233 build: support for gcc8
this patch covers kernel portion of Lustre
Lustre-change: https://review.whamcloud.com/34660
Lustre-commit:
6601661f96325b4971d0d1cb0be0fa01cc2ddc97
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I3fac8b89eef2291b5cb91ea05ee0b6ff32d11741
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35002
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Wed, 22 May 2019 20:28:55 +0000 (13:28 -0700)]
LU-11233 tests: fix gcc8 build warnings
this patch covers Lustre tests
Lustre-change: https://review.whamcloud.com/34661
Lustre-commit:
6733fbff9a682bcec5fdca6f7062c24f0fe27cfe
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I6345d603772fb32bbc4b38a758a3e97f0361d116
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35005
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Nathaniel Clark [Thu, 16 May 2019 17:18:04 +0000 (13:18 -0400)]
LU-12309 osd-zfs: Support disabled project quotas
Allow project quotas to be compiled in but disabled in the zpool.
This would be the case for zpools created by pre-0.8.0 ZFS, but then
used with newer ZFS.
Lustre-change: https://review.whamcloud.com/34888
Lustre-commit:
291e7196d39365739f9daa02efd25535b5415174
Signed-off-by: Nathaniel Clark <nclark@whamcloud.com>
Change-Id: I79c2c4ee3b191dad4150c218b25ced2508062d51
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35009
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Yang Sheng [Fri, 1 Feb 2019 05:04:10 +0000 (13:04 +0800)]
LU-11851 ldiskfs: reschedule for htree thread.
Thread may be waken inproperly in htree code. This patch
reschedule thread to keep locking correct.
Lustre-change: https://review.whamcloud.com/34160
Lustre-commit:
892f62c7a20d87954744c6c8937960379a870992
Change-Id: I6a8d1bbc0470b2577ca80faa304eb06f7913c218
Signed-off-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35006
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Mon, 15 Apr 2019 13:25:50 +0000 (16:25 +0300)]
LU-11233 utils: fix build warnings for gcc8
Quiet new build warnings that appear with GCC8, mainly related
to the length of string buffers not being long enough (in theory)
for the maximum possible string sizes, even if this never actually
is possible in practice.
Lustre-change: https://review.whamcloud.com/34662
Lustre-commit:
5e0ad2afa62e9eb7cf4f48c394c6a84c74a02f2f
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I83a955fc68f3e03fe84622ddf1cedfb30d5916ac
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35004
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Andreas Dilger [Thu, 18 Apr 2019 23:29:44 +0000 (17:29 -0600)]
LU-11233 utils: fix double-free of params fields
Call find_param_fini() on error so that the params are not leaked
during initialization if there is an intermediate error.
Zero out the parameters as they are freed, so if find_param_fini()
is called multiple times (as it is in some error paths) it does
not corrupt the heap by double freeing pointers. This can be hit
by calling "lfs getstripe -m" on multiple pathnames, some of which
do not exist.
Lustre-change: https://review.whamcloud.com/34711
Lustre-commit:
7c7c39d84c98df3c6fe33c04c9e391529a86db53
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ie0d7e9ee134deb0633af2f8052b8a458333ebbe5
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/35003
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Sebastien Buisson [Tue, 5 Feb 2019 15:06:39 +0000 (16:06 +0100)]
LU-930 doc: man page for l_getsepol
Man page for l_getsepol.
Lustre-change: https://review.whamcloud.com/34184
Lustre-commit:
e82adfcbd00fecaf588ee8ddd3d5432a5d92a51d
Test-Parameters: trivial
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I338492cebca9a088657ff8bd5122274e7e49a5c7
Reviewed-on: https://review.whamcloud.com/34833
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Joseph Gmitter <jgmitter@whamcloud.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Tue, 23 Apr 2019 14:51:28 +0000 (17:51 +0300)]
LU-12178 osd: do not rebalance quota under memory pressure
this will happen eventually.
Lustre-change: https://review.whamcloud.com/34741
Lustre-commit:
c5e5b7cd872eb2fa0028cef8b1a5e5c51b085b44
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Ibe4ef9e45deed5ea19169f3affed322351785357
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34926
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Mon, 8 Apr 2019 13:22:45 +0000 (21:22 +0800)]
LU-12169 llite: fill copied dentry name's ending char properly
Dentry name expect an extra '\0'. and dentry_len won't calcualte
extra '\0' for it, but we should allocate memory and fill it
when copying dentry name by ourselves.
Otherwise, lu_name_is_valid_2() will try to access @name[len]
and check whether it is '\0'. this is invalid memory access.
We will possibly hit a crash if the first access that bit is '\0'.
and the bit overwritten by someone else, and finally we failed
sanity check in mdc_name_pack().
LustreError: 157839:0:(mdc_lib.c:137:mdc_pack_name()) LBUG
Fixes: f575b65("LU-12020 llite: make sure name pack atomic")
Lustre-change: https://review.whamcloud.com/34611
Lustre-commit:
bc9cc327983c45e6255e0d6475b8bdbdcd82c938
Change-Id: I533e19a0e6efb0fca5a46bcdbdb0006d1b1bedab
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34925
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Mikhail Pershin [Mon, 22 Apr 2019 18:18:01 +0000 (21:18 +0300)]
LU-10777 dom: disable read-on-open with resend
The read-on-open can fill more data on reply buffer than
client allocated, this causes buffer re-allocation followed
by resend. Meanwhile FIO read test shows that such resends
perform worse than separate READ RPC. For example:
FIO 8k read is ~50% better without buffer re-allocation
with resend. Considering that there is parameter on MDC
'mdc_dom_min_repsize' to control read-on-open inline buffer
size, there is no sense to keep 'reallocation+resend'
option on MDT. Patch removes it.
Lustre-change: https://review.whamcloud.com/34700
Lustre-commit:
e0adb618a4b0d0182419a5731fe046e9157b9f51
Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: I7eb9d64f5551789e93b1f7676f61c0e7a5149f76
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34912
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Stephan Thiell <sthiell@stanford.edu>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Thu, 21 Mar 2019 03:05:14 +0000 (14:05 +1100)]
LU-12093 osc: don't check capability for every page
We check CFS_CAP_SYS_RESOURCE for every page during the io.
This is expensive on apparmor enabled systems, we can only
do that once for the entire io and use the result when
submitting the pages.
Don't init the oap_brw_flags during osc_page_init(), the flag
will be set in either osc_queue_async_io() or osc_page_submit().
Lustre-change: https://review.whamcloud.com/34478
Lustre-commit:
c1cab789aaa25bbb4062208aeb2822fde3007cd4
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: I0e664f43ce31c276b33476fdff11794185ab0a3b
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34920
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Thu, 4 Apr 2019 02:40:58 +0000 (21:40 -0500)]
LU-12152 lnet: Cleanup lnet_get_rtr_pool_cfg
The cfs_percpt_for_each loop contains an off-by-one error that causes
memory corruption. In addition, the way these loops are nested results
in unnecessary iterations. We only need to iterate through the cpts
until we match the cpt number passed as an argument. At that point we
want to copy the router buffer pools for that cpt.
Lustre-change: https://review.whamcloud.com/34591
Lustre-commit:
187117fd94e4904c168de02fc439b41a1fcc3e48
Cray-bug-id: LUS-7240
Test-Parameters: trivial
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I8c0dc7bab7ca42dbce04a9e6efa4343da4139239
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34922
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Tue, 16 Apr 2019 13:32:43 +0000 (22:32 +0900)]
LU-12098 mdd: explicitly clear changelogs on deregister
In case of MDS crash in the middle of changelog_deregister, the system
can end up with the changelogs user deregistered, but the changelog
entries not actually cleared. Then the only way to get rid of the
remaining changelogs not used anymore by any user is to register a new
changelogs user and then deregister it.
To protect from this scenario, explicitly clear changelogs used by the
user, before actually deregistering it.
Also add recovery-small test_136 for non-regression purpose.
Lustre-change: https://review.whamcloud.com/34688
Lustre-commit:
83ffa859bc629e246de9fcdfc82838b14c6d0ea3
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I14576180c9351337fc4d9ed0e1b176d352584750
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34921
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Wang Shilong [Sat, 6 Apr 2019 13:23:23 +0000 (21:23 +0800)]
LU-12165 quota: fix to use correct fsname array size
Max fsname is allowed to be LUSTRE_MAXFSNAME, plus '\0',
we expected arrary size should be LUSTRE_MAXFSNAME + 1.
Otherwise, we will hit following crash easily.
[864870.292204] [<
ffffffff9230e84e>] dump_stack+0x19/0x1b
[864870.293186] [<
ffffffff92308b50>] panic+0xe8/0x21f
[864870.294104] [<
ffffffffc0f3f805>] ? qsd_enabled_seq_write+0x205/0x210 [lquota]
[864870.295418] [<
ffffffff91c91b8b>] __stack_chk_fail+0x1b/0x20
[864870.296437] [<
ffffffffc0f3f805>] qsd_enabled_seq_write+0x205/0x210 [lquota]
[864870.297760] [<
ffffffff91e1e418>] ? __sb_start_write+0x58/0x110
[864870.298894] [<
ffffffff91e91050>] proc_reg_write+0x40/0x80
[864870.299883] [<
ffffffff91e1b490>] vfs_write+0xc0/0x1f0
[864870.300765] [<
ffffffff91e1c2bf>] SyS_write+0x7f/0xf0
[864870.301711] [<
ffffffff92320795>] system_call_fastpath+0x1c/0x21
Lustre-change: https://review.whamcloud.com/34608
Lustre-commit:
4414fd5365612a5fc1243e0e83f91d01949ba7d2
Test-parameters: trivial
Change-Id: I33dd331a83ddac0e0c36a82480e7e90ad0ed2c2a
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34923
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
git://git.whamcloud.com / fs / lustre-release.git / log