Whamcloud - gitweb
Oleg Drokin [Fri, 4 Mar 2022 22:10:25 +0000 (17:10 -0500)]
LU-15615 target: Free t10pi crypto state on error
Looks like when error happens we forgot to release crypto state that
not only leaks memory directly, but potentially can tie in-memory
pages too.
Change-Id: Ia0870ccbb194e4e9ca8701e1c01d519745c236df
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50539
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Gaurang Tapase [Tue, 4 Apr 2023 05:45:35 +0000 (11:15 +0530)]
LU-16659 build: Detect the mofed path based on running kernel
Test-Parameters: trivial
Change-Id: I519e93e8c26807da6143e2cf4d825ccf4a4180e4
Signed-off-by: Gaurang Tapase <gtapase@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50517
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Li Dongyang [Thu, 30 Mar 2023 12:45:04 +0000 (23:45 +1100)]
LU-14692 tests: wait for osp in conf-sanity/84
Wait for osp to change the first IDIF SEQ to a
normal SEQ, before using replay_barrier.
Otherwise the SEQ change could get lost and we
will trigger LASSERT during replay.
Change-Id: I32daa49d6329902b84eebb00090ae3cebe4a71b0
Test-Parameters: trivial testlist=conf-sanity env=ONLY=84,ONLY_REPEAT=10
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50477
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Arshad Hussain [Mon, 3 Apr 2023 21:48:15 +0000 (03:18 +0530)]
LU-16427 lfs: rmfid does not print anything on error
This patch:
01. Improve rmfid
Adds llapi_root_path_open() This function accepts the device
or path and returns the open fd for them. This was done so
that it is called only _once_ and not at every lookup.
Adds llapi_rmfid_at() This function makes the final IOCTL to
rmfid. Since llapi_root_path_open() we already had the valid
fd and populated 'fid_structure'. We could isolate this and
not call the former llapi_rmfid()
02. Fix rmfid silently accepting fid without fsname
or lustre root mount point. Make it correctly
fail if required arguments is not provided.
After Patch:
~~~~~~~~~~~~
$ lfs rmfid 0x200000402:0x1:0x0
lfs rmfid: missing <fsname|rootpath> or <fid>
Remove file(s) by FID(s)
usage: rmfid <fsname|rootpath> <fid> ...
Before Patch:
~~~~~~~~~~~~
$ lfs rmfid 0x200000402:0x1:0x0
$
03. Fix rmfid memory leak
After Patch:
~~~~~~~~~~~~
$ valgrind --leak-check=full lfs rmfid lustre 0x200000402:0x1:0x0
==33793== HEAP SUMMARY:
==33793== in use at exit: 0 bytes in 0 blocks
==33793== total heap usage: 4 allocs, 4 frees, 1,567 bytes allocated
==33793==
==33793== All heap blocks were freed -- no leaks are possible
Before Patch:
~~~~~~~~~~~~
$ valgrind --leak-check=full lfs rmfid lustre 0x200000402:0x1:0x0
==30812== LEAK SUMMARY:
==30812== definitely lost: 48 bytes in 1 blocks
==30812== indirectly lost: 0 bytes in 0 blocks
==30812== possibly lost: 0 bytes in 0 blocks
==30812== still reachable: 0 bytes in 0 blocks
==30812== suppressed: 0 bytes in 0 blocks
04. Update/Add Man pages
Update llapi_rmfid.3 and lustreapi.7 man pages
Add new llapi_rmfid_at.3 and llapi_root_path_open.3 man pages
Test-Parameters: trivial testlist=sanity
Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: Idda9313c97e48e9f7bf6486894b6ae3c74d71981
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50388
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Thomas Bertschinger <bertschinger@lanl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Fri, 10 Mar 2023 16:56:59 +0000 (16:56 +0000)]
LU-16631 llmount: improve usability
Add some simple help messages to both llmount.sh and
llmountcleanup.sh, similar to what auster has. This
will help unfamilar people understand the use of these
scripts.
Add option to disable client setup for llmount.sh. Add
options for llmount.sh environment variables.
Fix a couple small shellcheck warnings.
Update the file headers to have the SPDX license and
use the standard format.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I50dbb30bad8c8bc0479585293056d61a25aa001d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50260
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
James Simmons [Tue, 18 Apr 2023 01:07:47 +0000 (21:07 -0400)]
LU-16462 utils: handle lack of newer nla_attrs
Some platforms like SUSE12SP5 have an older version of libnl3.
Those versions lack proper support for newer nlattrs like NLA_S64.
Without proper support of these newer nlattrs this means
nla_validate() will see newer nlattrs as invalid. We need to
fill in this missing support on older platforms.
The "lctl ping" command will loop forever in jt_ptl_ping() if the
netlink yaml parser doesn't work, instead of falling back to the
"old_api" interface. However, because IOC_LIBCFS_PING was also
deleted, this doesn't work either. Use lustre_lnet_ping_nid() to
fallback to IOC_LIBCFS_PING_PEER (since v2_10_52_0-21-g7a36afd9df).
Also, jt_ptl_ping() was passing argv[1] directly to the kernel as
the NID to ping, without doing name resolution on it first, which
broke using "lctl ping HOSTNAME" instead of only numeric IP NIDs.
Ensure it returns a non-zero error code in case of failure.
Restore IOC_LIBCFS_GET_NI for compatibility until there have been
some releases with netlink support, so that "lctl list_nids" works.
Also, sanity test_217 that was testing "lctl ping" was always being
skipped, because "lctl list_nids" is never returning the hostname
with embedded '-', only numeric IP addresses. Change it to prefer
testing the hostname if it resolves to a NID, otherwise ping the
numeric NID anyway, to confirm that "lctl ping" is still working.
Test-Parameters: trivial clientdistro=sles12sp5 testlist=sanity
Test-Parameters: clientdistro=sles12sp5 testlist=conf-sanity env=ONLY=43+50+70+91+115+130
Fixes:
86ba46c244 ("LU-9680 obdclass: user netlink to collect devices info")
Fixes:
d137e9823c ("LU-10003 lnet: use Netlink to support LNet ping commands")
Fixes:
3e4061862e ("LU-864 test: Hostname name doesn't equal NID")
Change-Id: Ia2dfd84c2d1782578ceff1c2dc6f74d7aa9b458b
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49608
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Shaun Tancheff [Fri, 27 Jan 2023 06:54:42 +0000 (00:54 -0600)]
LU-16351 llite: Linux 6.1 prandom, folios_contig, vma_iterator
Linux commit v4.10-rc3-6-gc440408cf690
random: convert get_random_int/long into get_random_u32/u64
Linux commit v6.0-11338-gde492c83cae0
prandom: remove unused functions
prandom_u32 is a wrapper around get_random_u32, change users
of prandom_u32 to get_random_u32 and provide a fallback
to prandom_u32 when get_random_u32 is not available.
Linux commit v6.0-rc1-2-g25885a35a720
Change calling conventions for filldir_t
Add a test for the new filldir_t signature
Provide wrappers for transition from int (error code) to bool
Linux commit v6.0-rc3-94-g35b471467f88
filemap: add filemap_get_folios_contig()
Provide a wrapper and fallback to find_get_pages_contig
Linux commit v6.0-rc3-225-gf39af05949a4
mm: add VMA iterator
Use vma_iterator and for_each_vma when available.
Test-Parameters: trivial
HPE-bug-id: LUS-11377
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: I23dc23d0252e1995555b6685f5cf7c207edf642b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49232
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Qian Yingjin [Thu, 17 Feb 2022 03:42:14 +0000 (22:42 -0500)]
LU-15550 ptlrpc: retry mechanism for overflowed batched RPCs
Before send the batched RPC, the client has no idea about the
actual reply buffer size. The reply buffer size prepared by a
client may be smalller than the reply buffer buffer size in need.
We already have the patch to grow the reply buffer properly in
most cases.
However, when the reply buffer size is growing larger than
BUT_MAXREPSIZE (1000 * 1024), the server will return -EOVERFLOW
error code. At this time, the server only executed the partial
sub requests in the batched RPC. The overflowed sub requests are
not handled.
In this patch, it adds a retry mechanism for overflowed batched
RPC. When found that the reply buffer overflowed, the client will
rebuild the batched RPC for the unhandled sub requests, and use
work queue mechanism to resend the new batched RPC to the server
to re-execute then again.
Add the test case sanity test_123f to verify it for large LOV
stripes with overstriping.
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: If84fad32f2026bd34ffb47b3e163f84a9d950dbb
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/46540
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Sat, 21 Aug 2021 15:45:53 +0000 (18:45 +0300)]
LU-8704 osp: update local fldb cache
update local fldb cache during precreate. this is to avoid
a situation when LOD is generating LOVEA and has to lookup
in FLDB. in turn this may lead to an RPC with a local
transaction running. not good.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Iadf89eddcef88750d234b0139b67e04715e68855
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/44720
Reviewed-by: Sergey Cheremencev <scherementsev@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Andreas Dilger [Mon, 20 Mar 2023 02:22:10 +0000 (20:22 -0600)]
LU-16634 misc: standardize iocontrol param handling
Validate uarg and karg early in iocontrol processing where needed.
This needs kernel interop for 4.20+ kernels for access_ok(), but
this can be checked by #ifdef and does not need an autoconf test.
Fix incorrect definition of OBD_IOC_BARRIER to match reality.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Reported-by: Tao Lyu <tao.lyu@epfl.ch>
Change-Id: I1a0d2f839949debf346aa15c65b0f407e9ce7057
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50314
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Vitaliy Kuznetsov <vkuznetsov@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Tue, 11 Apr 2023 21:10:24 +0000 (17:10 -0400)]
LU-15515 contrib: Some style fixes for epython scripts
tighten formatting, remove spurious if() constructs,
add int conversion for shifts
Test-Parameters: trivial
Change-Id: Id89bc8ca898d0d5870ca6955b185571060eafd58
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50607
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Thu, 20 Apr 2023 05:12:27 +0000 (05:12 +0000)]
LU-16751 misc: remove unused small files
There are a number of small files which don't seem to serve any
purpose anymore. This patches removes them.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: Iddada509bfe1aaefbffdd0ec623a4efd4fc289e6
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50699
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Wed, 19 Apr 2023 07:20:33 +0000 (10:20 +0300)]
LU-15123 tests: check quota reintegration after recovery
4th step of quota reintegration (reconciliation) waits for recovery
completion. So the tests (like sanity-quota/7a) should wait for
recovery completion before checking reintegration results.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Id0aa5db01658621103d94ad6dafe91b2960b3a33
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50688
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Sergey Cheremencev <scherementsev@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Timothy Day [Tue, 18 Apr 2023 03:13:57 +0000 (03:13 +0000)]
LU-12610 contrib: spelling warning for OBD_ macros
Add a spelling warning for OBD_ macros, which are
going to removed in follow-up patches.
Now, checkpatch.pl will suggest the CFS_ alternatives:
WARNING: 'OBD_FAILED' may be misspelled - perhaps 'CFS_FAILED'?
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I97180894d30c24acd03eb2784270879a7ec80c0d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50665
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>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Hongchao Zhang [Fri, 14 Apr 2023 11:12:21 +0000 (19:12 +0800)]
LU-16735 test: cancel MDC locks and wait recovery
During test_35, the MDC LDLM locks should also be cancelled to
flush the pending operations and the recovery should be waited
to complete before checking the quota.
Test-Parameters: trivial testlist=sanity-quota mdscount=2 mdtcount=4
Change-Id: I6508644976be77ad2895107abf90144b51790cfe
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50638
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Mr NeilBrown [Wed, 12 Apr 2023 12:58:01 +0000 (08:58 -0400)]
LU-10391 lustre: obd_connect and reconnect now use large nid
The 'localdata' argument for o_connect and o_reconnect when called
server-side is now a 'struct lnet_nid *' rather than an
'lnet_nid_t *'.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I1ce72ec11a5d2463fb90ab2686410e2dd96118e2
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50097
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Patrick Farrell [Mon, 20 Mar 2023 21:43:43 +0000 (17:43 -0400)]
LU-14541 llite: Check for page deletion after fault
Before completing a page fault and returning to the kernel,
we lock the page and verify it has not been truncated. But
we must also verify the page has not been deleted from
Lustre, or we can return a disconnected (ie, not tracked by
Lustre) page to the kernel.
We mark deleted pages !uptodate, but this doesn't matter
for faulted pages, because the kernel assumes they are
returned uptodate, and maps them in to the process address
space. Once mapped, the page state is not checked until
the page is unmapped.
But because the page is referenced by the mapping, it stays
in the page cache even though it's been disconnected from
Lustre.
Because the page is disconnected from Lustre, it will not
be found and cancelled on lock cancellation. This can
result in stale data reads.
This is particularly an issue with releasepage (called from
drop_caches or under memory pressure), which can delete
pages separately from cancelling covering locks.
If releasepage is disabled, which is effectively what
"LU-14541 llite: Check vmpage in releasepage"
does, this is not an issue. But disabling releasepage
causes other problems and is incorrect anyway.
Fixes:
c524079f4f ("LU-14541 llite: Check vmpage in releasepage")
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: If1164db8f8e92a1cf811431d56d15f30d8eb3faa
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49653
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Zhenyu Xu <bobijam@hotmail.com>
Andreas Dilger [Fri, 24 Mar 2023 02:31:59 +0000 (20:31 -0600)]
LU-16661 build: improve lustre.spec.in Requires
Add Suggests: bash-completion for lustre-client and lustre for
lctl and lfs sub-command completion.
Move perl from Requires to Recommends, since there are only some
uncommonly used tools (llstat, llobdstat) that are using perl.
Remove a couple of ancient obsolete test scripts that used perl.
lustre-iokit incorrectly Required perl instead of python3.
Set minimum kernel version for client to be 3.10 or later.
Change "netstat" to "ss" in tests to avoid dependency issues.
Fix sanity.sh and conf-sanity.sh tests for sles12sp5 issues.
Test-Parameters: trivial testlist=runtests clientdistro=sles12sp5
Test-Parameters: trivial testlist=runtests clientdistro=el9.1
Test-Parameters: trivial testlist=runtests clientdistro=sles15sp3
Fixes:
7521473bdd ("LU-16382 spec: add more dependencies for lustre-tests")
Fixes:
fd734cffb3 ("b=18443 tests: remove obsolete tests scripts")
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I48c6819596c81cb044e983bc64f1edf1ee3ebbe5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50397
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Alexey Zhuravlev [Mon, 7 Dec 2020 14:25:48 +0000 (17:25 +0300)]
LU-14189 contrib: docker example to build Lustre
can be easily modified for any target like specific RHEL, CentOS, etc
Test-Parameters: trivial
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I9f1abf653e0a89c5d115c53abc21df8d87b86e79
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40900
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Qian Yingjin [Thu, 14 Jan 2021 08:07:24 +0000 (16:07 +0800)]
LU-14139 statahead: add test for batch statahead processing
This patch adds sanity test_123ad() for batched statahead
processing, verify it work as expected.
Test-Parameters: trivial testlist=sanity env=ONLY=123ad,ONLY_REPEAT=50
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I742dcac7140ad842fff2afadeb2d948661365b95
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/41220
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Li Dongyang [Thu, 30 Mar 2023 13:02:33 +0000 (00:02 +1100)]
LU-11912 tests: SEQ rollover fixes
To avoid changeing SEQ after replay_barrier, we
use force_new_seq when starting the test suites heavily
using replay_barrier, e.g. replay-single.
However when there are fewer OSTs, the default 16384
SEQ width could not last the entire test suite, SEQ
rollover could still happen randomly after replay_barrier.
To overcome this, change the default OSTSEQWIDTH to
65536, and divide by number of OSTs, so the SEQ width is
larger with fewer OSTs. For 8 OSTs, the SEQ width is 16384,
and make sure we don't go under it.
Use force_new_seq_all for the test suites using replay_barrier
on MDTs other than mds1.
Add force_new_seq_all for replay-ost-single, which is using
replay_barrier on OST. If SEQ rollover happens after that,
the SEQ range update on ofd is lost due to replay_barrier,
the next time when we try to allocate a new SEQ will end up
with an old one.
Use force_new_seq_all for the test cases(namely sanity-pfl/0b
0c 1c 16b sanity/27Cd) checking for number of stripes created
with overstriping, to make sure we have enough objects
in the precreate pool.
Test-Parameters: trivial ostcount=4 testlist=replay-single
Test-Parameters: ostcount=2 testlist=replay-single
Test-Parameters: mdtcount=2 testlist=conf-sanity env=ONLY=122a,ONLY_REPEAT=10
Test-Parameters: testlist=sanity,sanity-pfl
Test-Parameters: testlist=sanity-scrub,replay-single,obdfilter-survey,replay-ost-single,large-scale
Fixes:
0ecb2a167c ("LU-11912 ofd: reduce LUSTRE_DATA_SEQ_MAX_WIDTH")
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Change-Id: Ic65111199f042405d6db8acb729b2cddf91138af
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50478
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Mon, 17 Apr 2023 08:34:12 +0000 (02:34 -0600)]
LU-11170 tests: don't fail sanity/415 in VM
Don't fail sanity test_415 when running in a VM due to variable
runtimes for the tests.
A proper solution would be to examine the logs to determine if
the renames are blocked or just all slow due to VM contention.
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I1a5d0f601705c9ec8559e760c4ec27c7f83ebbe5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50654
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Vikentsi Lapa <vlapa@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
hxing [Thu, 20 Apr 2023 07:46:29 +0000 (15:46 +0800)]
LU-15690 tests: skip replay-ost-single/12a on old server
Skip 12a of replay-ost-single for older server version.
Test-Parameters: trivial testlist=replay-ost-single env=ONLY=12a
Fixes:
28769c65987c ("LU-15195 ofd: missing OST object")
Signed-off-by: Xing Huang <hxing@ddn.com>
Change-Id: I473452a5326691f4394c9e3ab2ab5dfecbc6ec58
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50701
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Qian Yingjin [Fri, 14 May 2021 14:53:52 +0000 (22:53 +0800)]
LU-14139 ptlrpc: grow PtlRPC properly when prepare sub request
In this patch, it prepares and grows PtlRPC reply buffer
properly for SUB batch request in @req_capsule_server_pack().
At the same time, it adds a limit of reply buffer size with
BUT_MAXREPSIZE = (1000 * 1024).
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I4277974b3b0e9cd19fd0d18ae7c029cccaa9c558
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/43707
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Tue, 11 Apr 2023 20:22:38 +0000 (16:22 -0400)]
LU-16465 tests: update sanity test 806 to use save/restore_opencache
There are existing primitives to do this, so no need to opencode them
Fixes:
dfb08bbf77 ("LU-16465 llite: fix LSOM blocks for ftruncate and close")
Test-Parameters: trivial testlist=sanity env=ONLY=806
Change-Id: Ibc0d34050999bfda8f56384752ede09cacc4df91
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50606
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Etienne AUJAMES <eaujames@ddn.com>
Aurelien Degremont [Fri, 31 Mar 2023 09:30:37 +0000 (11:30 +0200)]
LU-16734 gss: fix lookup_user_key() bug
With more recent kernels, like on Ubuntu 22.04, trying to
delete some keyring resources trigger a kernel warning message
and cleaning is not successful, leading to stuck resources
and warning messages being regularly printed.
This is because Linux 5.8, in commit 8c0637e, introduced an API
change for lookup_user_key() that was not taken in account.
Update the lookup_user_key() call from _user_key() to fix it.
Change-Id: I34ef4dac3f56cbb4aac6bc5a3bad36feb66b8675
Signed-off-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50623
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Jonathan Calmels <jcalmels@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Chris Horn [Fri, 7 Apr 2023 16:46:29 +0000 (10:46 -0600)]
LU-16573 lnet: Check empty list in cfs_match_nid_net
cfs_match_nid_net() needs to check whether the list of range
expressions describing the address is empty. Otherwise, for numeric
based addresses, we may hit the assert in libcfs_num_match().
Test-Parameters: trivial testlist=sanity-lnet
HPE-bug-id: LUS-11480
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Id2460137607f5564751e729ee7b716d9151b5d37
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50576
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Arshad Hussain [Thu, 6 Apr 2023 07:00:45 +0000 (03:00 -0400)]
LU-16721 lfs: Fix path2fid segfault
This patch fixes segfault when calling path2fid
under interactive mode.
Before Patch
============
lfs > path2fid /mnt/lustre/a
[0x200000401:0x1:0x0]
Segmentation fault
After Patch
============
$ lfs
lfs > path2fid /mnt/lustre/a
[0x200000401:0x1:0x0]
lfs > path2fid --parent /mnt/lustre/a
[0x200000007:0x1:0x0]/a
lfs >
Testcase sanity/154h added.
Test-Parameters: trivial testlist=sanity
Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: I4693e0d476e9e7f570f45fd7a31d275d549245aa
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50557
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Patrick Farrell [Wed, 5 Apr 2023 16:27:18 +0000 (12:27 -0400)]
LU-16714 utils: Clarify fd/fdv naming
The migrate code uses the deeply opaque 'fd' and 'fdv'
(file-descriptor-volatile) to refer to the source and
destination file descriptors.
Replace these with something that tells a casual reader
what's going on.
Test-parameters: trivial
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: I56f2dfe81bcd4eff1c168f093e8580b373dc1984
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50546
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Mon, 3 Apr 2023 22:43:46 +0000 (15:43 -0700)]
LU-16706 kernel: update RHEL 9.1 [5.14.0-162.22.2.el9_1]
Update RHEL 9.1 kernel to 5.14.0-162.22.2.el9_1.
Test-Parameters: trivial fstype=ldiskfs \
clientdistro=el9.1 serverdistro=el9.1 testlist=sanity
Change-Id: Ib5186e6f0dcd89660b7000db7f37c0c5a29f944f
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50512
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Fri, 31 Mar 2023 04:40:38 +0000 (04:40 +0000)]
LU-16694 tests: replace resolveip script
The resolveip script can be replaced with a bash one-liner,
using getent and awk.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I207ea011e43b7b236d5082994ffb51654d8d782c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50491
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Vitaly Fertman [Fri, 2 Dec 2022 19:41:21 +0000 (22:41 +0300)]
LU-16693 lod: ENODEV on setstripe with wrong OST#
ENODEV should not be returned as it is a recoverable error and
the RPC will be just resent with the same set of parameters
Also, make getstripe to print out the object list for dirs
if it is MAGIC_SPECIFIC, to see the obdidx.
HPE-bug-id: LUS-11395
Signed-off-by: Vitaly Fertman <vitaly.fertman@hpe.com>
Change-Id: Idffbf3c2b525c4e00c4b662c948460e3735445fc
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Vladimir Saveliev <vladimir.saveliev@hpe.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50487
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Sergey Cheremencev <scherementsev@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Thu, 30 Mar 2023 10:10:23 +0000 (13:10 +0300)]
LU-16684 utils: reset buf_len in print_out_devices()
so that print_out_devices() doesn't hit false buffer overflow
leading to missing devices in lctl device_list output.
Fixes:
ba0d5ffc1c ("LU-9680 utils: new llapi_param_display_value().")
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Ia0ebbd0150651d7c631348201c26ea8b3d1ff704
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50475
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Charlie Olmstead [Mon, 27 Mar 2023 20:08:57 +0000 (14:08 -0600)]
LU-16672 tests: auster node.yml labels Alma and Rocky as CentOS
release() assumes a node with /etc/centos-release is CentOS. This patch
removes that assumption and uses the name in the centos-release file.
Corrected the os-release code to strip off the last word if present.
Test-Parameters: trivial
Change-Id: Ia5acbce3351ca23f4d9265d1aaf8d952a2c8b502
Signed-off-by: Charlie Olmstead <charlie@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50442
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Fri, 24 Mar 2023 17:24:52 +0000 (17:24 +0000)]
LU-16666 doc: remove unmaintained doxyfiles
These Doxygen related files had old urls, some using http rather
than https. Some code scanning tools will flag this. Rather than
update these, remove all of the old doxyfiles. They are very out
of date, Doxygen throws many errors when you try to use them, and
they do not seem to generate usable documentation.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I5a24d4754582ecee558c4e87385b8835d2675adc
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50415
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Wed, 12 Apr 2023 13:27:16 +0000 (09:27 -0400)]
LU-13340 mgs: convert class_parse_nid4 to class_parse_nid
All callers of class_parse_nid4() now use class_parse_nid()
and so much handle a large nid.
do_lcfg_nid() is introduced to help with this.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I502fa16871d689a8248e4243679918d58464efcd
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50094
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Wed, 12 Apr 2023 13:24:09 +0000 (09:24 -0400)]
LU-13340 mgs: change record_add_uuid() to take large nids
This just changes the function signature and does not really
change any functionality, because it always just converts the
struct lnet_nid back to lnet_nid_t.
Future patches will make more sense of this.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib3427ba7cfe54fc00427232835e7378922d8f616
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50093
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Mr NeilBrown [Mon, 12 Dec 2022 04:31:16 +0000 (15:31 +1100)]
LU-16382 spec: SUSE OBS requires kernel-source requirement.
The SUSE OBS creates a virtual environment containing
ONLY the stated BuildRequires requirements, and some defaults.
To be able to build ldiskfs we need the kernel source, so we
need "BuildRequires: kernel-source".
However when contrib/lbuild it is used to build lustre, it finds the
source by other means and fails on that BuildRequires. So it must be
conditional on running under abuild. (the OBS build tool).
When abuild extracts these BuildRequires, it cannot parse %() so
conditions using that all appear to be "false". But %() is the only
way to detect abuild - but looking in environment.
So a dance is needed to fit with all these odd requirements.
Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I2fe9ecaf857ecbd5fda7e857b661b5b756501190
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49365
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Thu, 13 Apr 2023 06:16:51 +0000 (23:16 -0700)]
LU-15374 tests: check FULL and IDLE for client import state
The client-to-OST import state can be FULL or IDLE.
Test-Parameters: trivial testgroup=review-dne-part-3
Test-Parameters: trivial env=SLOW=no,FAILURE_MODE=HARD \
clientcount=4 mdtcount=1 mdscount=2 osscount=2 \
austeroptions=-R failover=true iscsi=1 \
testlist=recovery-mds-scale
Fixes:
25606a2ce1 ("LU-15342 tests: escape "|"")
Fixes:
3da8f014fd ("LU-12857 tests: allow clients to be IDLE after recovery")
Fixes:
5a6ceb664f ("LU-7236 ptlrpc: idle connections can disconnect")
Change-Id: I3582ceb273d241ee71fe907f6d1423746e453faa
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49298
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Qian Yingjin [Fri, 11 Dec 2020 10:38:03 +0000 (18:38 +0800)]
LU-14139 ptlrpc: grow reply buffer properly for batch request
This patch adds the support to grow the reply buffer for batch
PtlRPC request.
With this support, statahead sanity-pfl test for test_16b will
pass for large LOV stripes with overstriping.
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: Iaa7eb88b49d6ee068ec1fd9666a8bac2839b5041
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40945
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Qian Yingjin [Fri, 11 Dec 2020 03:20:38 +0000 (11:20 +0800)]
LU-14139 statahead: add stats for batch RPC requests
This patch adds stats for batch PtlRPC request. It can show the
statistical information such as how many subreqs in a batch RPC.
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I2f71ff5d01ab1070bd8d771a72edd786ad27f03c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40943
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 18 Oct 2019 13:39:50 +0000 (22:39 +0900)]
LU-12885 mdd: only bottom of clf_flags in changelog
Make it clear at the callers that only the bottom 12 bits of
clf_flags are stored in the Changelog record.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I23c0844e7ee59dd007b83f37d591df6fbe3ebbe5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/36522
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Mr NeilBrown [Thu, 26 May 2022 04:42:44 +0000 (14:42 +1000)]
LU-10391 ptlrpc: change tc_nid in nrs to be struct lnet_nid
switch to struct lnet_nid and adjust accordingly.
Test-Parameters:trivial testlist=sanityn envdefinitions=ONLY="77"
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Iaa73bc5b95d78f1d69500ea8856ae0d5cd442f8b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50101
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Nikitas Angelinas <nikitas.angelinas@hpe.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Thu, 19 May 2022 01:45:19 +0000 (11:45 +1000)]
LU-10391 lustre: change cfs_match_nid to take large nid.
large nid now used more places.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I181ab0345a4bf2f9bb5c4b27eafb794968e8ef7e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50098
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Wed, 12 Apr 2023 13:29:58 +0000 (09:29 -0400)]
LU-10391 obdclass: remove class_parse_nid4()
class_parse_nid4() not used any more.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Iefff39a5eaeed8a8ba6fd016e3392db9e92d5422
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50095
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Mon, 20 Feb 2023 01:06:41 +0000 (12:06 +1100)]
LU-10391 libcfs: add large-nid string conversion functions.
The user-space libcfs now has functions to convert between strings and
large nids.
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I554e2da0c3d56397ebe60fc84fad28fec6704a18
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50092
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Etienne AUJAMES [Mon, 12 Sep 2022 11:13:13 +0000 (13:13 +0200)]
LU-14976 nrs: change nrs policies at run time
This patch take extra references on policy to avoid stop a NRS policy
with pending/queued request in it.
It uses a new refcount_t "pol_start_ref" for this purpose to keep
track of policy usage in started state. It enables to safely stop a
policy without "nrs_lock" and avoids to sleep in the spinlock.
It adds a wait queue field "pol_wq" in "struct ptlrpc_nrs_policy" to
wait all queued request in a stopping policy to be drained when
restarting policy with a different argument.
Add test sanityn 77r for this use case.
Test-Parameters: testlist=sanityn env=ONLY=77r,ONLY_REPEAT=20
Signed-off-by: Etienne AUJAMES <eaujames@ddn.com>
Change-Id: I1425f52324f755f1b76ea8210de52647c072a592
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48523
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Alex Zhuravlev [Tue, 1 Feb 2022 20:50:02 +0000 (23:50 +0300)]
LU-15300 mdt: refresh LOVEA with LL granted
this change tries to fix two problems:
1) mdt_reint_open() fetches LOVEA before layout lock is taken.
this may race with another process changing the layout and
may result in a stale layout returned with a granted layout
lock - re-fetch LOVEA once layout lock is granted
2) lov_layout_change() should not apply old layouts which
can get through when MDS doesn't take layout lock
3) LFSCK shouldn't ignore layout version stored on MDS to avoid
a situation when version degrades compared to client's copy.
This patch misses an optimization and can result in a number of
useless calls to OSD to fetch LOVEA. To be fixed in a followup
patch.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Idee1101d152ab09947faf6d75574a8761a7690a5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/46413
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Zhenyu Xu <bobijam@hotmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sergey Cheremencev [Wed, 5 Apr 2023 10:24:51 +0000 (13:24 +0300)]
LU-8367 osp: remove unused fail_locs from sanity/27S,822
OBD_FAIL_OST_GET_LAST_FID and OBD_FAIL_OSP_GET_LAST_FID
are not used anymore since sanity test_27S has been removed.
This may lead to failures in interoperability testing.
For example, 27S caused osp_precreate_cleanup_orphans to
stuck due to not cleared opd_pre_recovering flag(normally,
it should be set to 0 before calling osp_precreate_reserve).
It was the reason of failure tests 27U, 27R, 39r and 51d
in sanity.sh.
The patch also removes OBD_FAIL_NET_ERROR_RPC and
OBD_FAIL_OSP_PRECREATE_PAUSE that were used in sanity-822
"test precreate failure". This test also has been removed
in
63e17799a3.
Fixes:
63e17799a3 (LU-8367 osp: enable replay for precreation request)
Signed-off-by: Sergey Cheremencev <scherementsev@ddn.com>
Change-Id: Ia78ce39adb9d59e476eb36e5e69954cc26353b27
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50543
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Thu, 7 Oct 2021 16:07:14 +0000 (19:07 +0300)]
LU-15072 lod: extra status for spilling
to debug possible issues
Test-Parameters: trivial
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Ibec0173c5fd70ecf835e96a372352989df1e1f92
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/45153
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alexandre Ioffe <aioffe@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 31 Mar 2023 23:43:54 +0000 (17:43 -0600)]
LU-16655 build: add wirecheck/wiretest for disk structs
Move OI scrub_file and related constants to lustre/lustre_disk.h
to ensure these on-disk structures do not change in the future.
Add lr_server_data and lsd_client_data structs from last_rcvd file.
Add replay_data_v1 and replay_data_v2 structs from recovery.
Move struct ost_layout together with lustre_ost_attrs where it is
used, and add PFID_STRIPE_IDX_BITS/PFID_STRIPE_COUNT_MASK checks
for the usage of this struct.
Change uuid.h header includes to allow uuid_t in lustre_disk.h.
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ie992ca383f24c29d2449ee22849a3c476096551c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50482
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Fri, 31 Mar 2023 23:38:38 +0000 (17:38 -0600)]
LU-16655 scrub: change sf_uuid to guid_t
Change the type of sf_uuid from uuid_t to guid_t. The sizes are
identical, but the benefit is that guid_t is usable in userspace
with the <linux/uuid.h> header, unlike uuid_t.
Change the accessors to use the corresponding guid_*() functions,
but no functional changes are needed.
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I3048cfee20a5e4ea2c0c2203d22eb76c1437577b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50496
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>
Alex Deiter [Fri, 7 Apr 2023 19:49:23 +0000 (23:49 +0400)]
LU-16163 tests: skip racer_on_nfs for NFSv3
Export ALWAYS_EXCEPT env for child NFS test
Fixes:
513eb670b0 ("LU-16163 tests: skip racer_on_nfs for NFSv3")
Test-Parameters: trivial testlist=parallel-scale-nfsv3
Signed-off-by: Alex Deiter <adeiter@tintri.com>
Change-Id: Ibb4a9916166f13ab9bd2374b33d4313453972276
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50579
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Oleg Drokin [Tue, 11 Apr 2023 20:17:03 +0000 (16:17 -0400)]
LU-16728 misc: fix typos in comments
destroied -> destroyed
preapre -> prepare
Change-Id: Id00be2bb4a219d70bb4a69b90f624ef2cc0d6712
Test-Parameters: trivial
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50605
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Patrick Farrell [Fri, 7 Apr 2023 18:13:03 +0000 (14:13 -0400)]
LU-16510 build: include unsafe_memcpy definition
The original LU-16510 missed a key part of the
unsafe_memcpy code from the upstream kernel, and so we
weren't actually defining unsafe_memcpy() as intended.
Thanks to Aurelien Degremont <adegremont@nvidia.com> for
pointing this out.
Fixes:
919b93b9 ("LU-16510 build: fortified memcpy from linux 6.1")
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: Ib9e2d56ed0b3691f1ab9fcd25403fa86ac784b6d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50573
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Tue, 4 Apr 2023 14:53:22 +0000 (16:53 +0200)]
LU-16708 tests: fix zconf_mount_clients for SSK
When SHARED_KEY is in use, zconf_mount_clients can add 'skpath' mount
option to load nodemap-specific keys. But this must not overwrite
already specified mount options.
Fixes:
05e5cb0b0c ("LU-16683 tests: fix sanity-sec test_61 for SSK")
Test-Parameters: trivial
Test-Parameters: testlist=sanity-sec env=ONLY=61
Test-Parameters: testlist=sanity-sec env=SHARED_KEY=true,ONLY=61
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I3dbc6ffe72722659ead77fb1c2c2675873c7aff2
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50523
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Timothy Day [Fri, 31 Mar 2023 04:12:46 +0000 (04:12 +0000)]
LU-16694 tests: remove deprecated h2tcp, h2o2ib
These functions were deprecated about 6 years ago. It's
probably safe to remove them now.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I6b16d0ad1dc1ea6f00fc730cb95672ef3c07f8a5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50513
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Qian Yingjin [Mon, 3 Apr 2023 09:38:03 +0000 (05:38 -0400)]
LU-16700 pcc: reserve flags for PCC-RO
This patch reserves flags for PCC-RO. It also adds wire check /
test for these new flags.
Test-Parameters: trivial
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I7086843770d16a6a7d2fd6b6ad3c77d43e04bf3b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50504
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Etienne AUJAMES [Tue, 28 Mar 2023 19:46:24 +0000 (21:46 +0200)]
LU-16674 obdclass: optimize job_stats reads
This patch has 2 objectives:
1/ limit the lock time on ojs_list (list of job stats)
"lctl get_param mdt.*.job_stats" can not dump job_stats in a single
read (seq_file buffer is limited to 4k). So, several reads are needed
to dump the full job list.
For each read, we have to find the job entry corresponding to the file
offset. For now, we walk ojs_list from the beginning to get this
entry.
This patch saved the last known entry and the corresponding offset to
start the next read from here.
2/ avoid the lock contention when reading job_stats
This patch replaces the read lock on ojs_lock by RCU locking, this
enables userspace processes reading the job_stats not to interfere
with the kernel target threads.
Add the stress test sanity 205g to check for possible races.
Add stack_trap in sanity test 205a and 205e to restore jobid_name and
jobid_var.
* Performance *
The following command is used to capture records:
$ time grep -c job_id /proc/fs/lustre/mdt/lustrefs-MDT0000/job_stats
- job_stats dump with no fs activity
Here are results after ending sanity test 205g with slow mode and
job_cleanup_interval=300s.
___________________________________
| nbr of job | time | rate |
_____________|____________|______|_______________|
|without patch| 14749 | 1.3s | 11345 jobid/s |
|_____________|____________|______|_______________|
|with patch | 22209 | 0.6s | 37015 jobid/s |
|_____________|____________|______|_______________|
|diff % | +43% | -54% | +226% |
|_____________|____________|______|_______________|
- job_stats dump with fs activity
Here are results before ending sanity test 205g with slow mode and
job_cleanup_interval=300s.
___________________________________
| nbr of job | time | rate |
_____________|____________|______|_______________|
|without patch| 14849 | 2.3s | 6428 jobid/s |
|_____________|____________|______|_______________|
|with patch | 22776 | 1.2s | 18823 jobid/s |
|_____________|____________|______|_______________|
|diff % | +53% | -47% | +192% |
|_____________|____________|______|_______________|
Test-Parameters: testlist=sanity env=SLOW=yes,ONLY=205g,ONLY_REPEAT=10
Test-Parameters: testlist=sanity env=ONLY=205g serverversion=2.15.2
Test-Parameters: testlist=sanity env=SLOW=yes,ONLY=205
Signed-off-by: Etienne AUJAMES <eaujames@ddn.com>
Change-Id: Ic4cd90965720af76eff0ed4e00ca897518bfbc66
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50459
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Mon, 27 Mar 2023 08:46:07 +0000 (10:46 +0200)]
LU-16670 enc: make sure DoM files are correctly decrypted
Make sure DoM files are decrypted upon read by loading their
associated encryption context, via llcrypt_prepare_readdir()/
llcrypt_get_encryption_info().
Fix sanity-sec test_50 accordingly.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ie9ef3cbb08d2295a2fd10b9e9ab0862119c7723e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50429
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andreas Dilger [Tue, 7 Feb 2023 23:55:00 +0000 (16:55 -0700)]
LU-16096 tgt: improve messages for reply_data
Due to the format change in reply_data for WBC, it is not possible
to downgrade the MDT filesystem to an older release after mounting
it with 2.16.0 and later.
On the path to fixing that, improve the error messages printed by
tgt_reply_data_init() so that it is possible to see why the mount
is failing. The next step is to avoid upgrading the replay_data
file if it is not necessary (no clients mounting with WBC enabled).
Test-Parameters: trivial testlist=conf-sanity env=ONLY=32
Fixes:
bbf0017fdea5 ("LU-16096 recovery: upgrade reply data after recovery finish")
Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Change-Id: Idd176031fef99b16de9278f05516dc45e817a819
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49939
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Vitaliy Kuznetsov <vkuznetsov@ddn.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Fri, 27 Jan 2023 20:47:25 +0000 (14:47 -0600)]
LU-16483 ptlrpc: Track highest reply XID
Keep track of the highest XID that we've received a reply for.
When an OBD_PING expires, do not disconnect the import if the failed
XID is less than or equal to the last reply XID. This avoids situation
where a lost OBD_PING rpc causes a reconnect even though we've
completed other RPCs in the meantime.
HPE-bug-id: LUS-11474
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I7e66bcc1368fa41ec86ffd843abac676f8d29254
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49807
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Wed, 29 Mar 2023 20:16:19 +0000 (20:16 +0000)]
LU-930 doc: add lctl-lcfg man pages
Add "lctl lcfg_clear", "lctl lcfg_erase", and "lctl lcfg_fork"
as aliases for "clear_conf", "erase_lcfg", and "fork_lcfg"
respectively, for more consitent naming.
Add separate man pages for lctl-lcfg_clear, lctl-lcfg_erase,
lcfg-lcfg_fork, lctl-erase_lcfg, lctl-fork_lcfg, lctl-clear_conf.
Remove the combined man page for lctl-lcfg.
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I9daf2efc0cf0f9a2b549d8513098729151178207
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48887
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>
Etienne AUJAMES [Fri, 12 Aug 2022 19:41:08 +0000 (21:41 +0200)]
LU-16077 tbf: pb_uid/pb_gid ptlrpc_body fields for TBF rules
The file UID/GID are packed inside bulk IO because the requests are
sent asynchronously (cannot use the current thread UID/GID).
This is an issue for TBF rules if the file/inode UID/GID doesn't
match the process ones (e.g: reading common libraries): we can't limit
the user RPCs doing the IOs in that case.
This patch pack UID/GID for TBF rules inside ptlrpc_body (
pb_padding64_2 -> (pb_uid, pb_gid)) to be independent of quota
interactions: it stores the client process UID/GID instead of the
values of the file attrs.
Moreover, it enables to track requests naturally without UID/GID like
ldlm_flock_enqueue.
This patch saves the process UID/GID inside the ll_inode_info struct.
Then it restores these values when sending a bulk IO from a ptlrpc
thread (like for jobids).
Add sanityn test_77jb to verify.
Fixes: e0cdde1 ("LU-9658 ptlrpc: Add QoS for uid and gid in NRS-TBF")
Test-Parameters: testlist=sanityn env=ONLY=77,ONLY_REPEAT=20
Test-Parameters: serverversion=2.12.9 testlist=sanityn env=ONLY=77
Test-Parameters: clientversion=2.12.9 testlist=sanityn env=ONLY=77
Signed-off-by: Etienne AUJAMES <etienne.aujames@cea.fr>
Change-Id: I61e42267ae568f9a1eb6fe57937a5e96f1824010
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48235
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Wed, 18 Aug 2021 08:09:49 +0000 (11:09 +0300)]
LU-13132 osd: osd-zfs to cache dbufs for llog objects
working set for llog objects is tiny and very predictable. osd-zfs
can cache couple dbufs (first block storing the header and last
block for new records).
for sanity/60a (llog test) it gives 5939307 hits and 5776 misses
while average osd_write() goes down from 1.09 usec to 0.27 usec,
total time for sanity/60a: before - 153s, after - 101s.
this approach can be used in few other cases like last_rcvd.
Change-Id: Icc0126658894085d33ef79ae41ac6c1ed4140f4c
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/37222
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Andreas Dilger [Mon, 3 Apr 2023 20:31:26 +0000 (14:31 -0600)]
LU-980 mount: improve mount/unmount messages
In some cases, unmount errors are printed in multiple places, or
are expected so printing them on the console log is not necessary.
Conversely, some status messages such as mounting or unmounting the
whole target should not be rate limited.
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Id55f6ee3af5ad3cbe44a380314aa4b31f6b4bad3
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50511
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Andrew Perepechko [Tue, 10 Jan 2023 21:53:38 +0000 (16:53 -0500)]
LU-16609 target: top_trans_create cannot alloc memory
top_trans_create() requests __GFP_IO memory allocation,
which does not allow direct reclaim. However, if the
memory shortage is temporary, direct reclaim is reasonable.
GFP_NOFS is __GFP_IO with additional reclaim bits.
Change-Id: I2c84d9d74188660063c948573780745a2b59a688
Signed-off-by: Andrew Perepechko <andrew.perepechko@hpe.com>
HPE-bug-id: LUS-11293
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50176
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alexey Lyashkov <alexey.lyashkov@hpe.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Shaun Tancheff [Wed, 12 Apr 2023 12:54:15 +0000 (07:54 -0500)]
LU-16732 ldiskfs: update for ext4-delayed-iput for RHEL9.1
ext4-delayed-iput patch does not apply cleanly to RHEL9.1
kernel.
Adjust the minor conflict in ext4_put_super()
Test-Parameters: trivial
Fixes:
616fa9b581 ("LU-15404 ldiskfs: use per-filesystem workqueues to avoid deadlocks")
HPE-bug-id: LUS-11570
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Id286904177cb444aa12e3c16e134d5acc17030f3
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50615
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Andreas Dilger <adilger@whamcloud.com>
Alex Deiter [Tue, 28 Mar 2023 17:05:42 +0000 (21:05 +0400)]
LU-930 doc: update Alex Deiter contact info
Update my email address to the one corporate email
Test-Parameters: trivial
Change-Id: Ic540c5a289e2862212cc0164155b5f85cbc0d96c
Signed-off-by: Alex Deiter <adeiter@tintri.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50458
Reviewed-by: Peter Jones <pjones@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Wed, 22 Mar 2023 16:09:58 +0000 (17:09 +0100)]
LU-16646 krb: improve lookup of user's credentials
Rather than only looking up for user's credentials in hard-coded
FILE:/tmp/krb5cc_<uid>, try first the default ccache on the system,
and then fallback to files matching /tmp/*krb5cc* and
/run/user/<uid>/*krb5cc*.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ic2bedb4cc12e9adad0ce63bd0617b2e0ec13907e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50377
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Jonathan Calmels <jcalmels@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Fri, 17 Mar 2023 16:30:07 +0000 (17:30 +0100)]
LU-16646 krb: use system ccache for Lustre services
Instead of hard-coding a FILE credentials cache for Lustre services,
comply with the system configuration in place and use the default
ccache.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ib89fe117e9f1d937925a02c7ed786a81cd8954cb
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50342
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Jonathan Calmels <jcalmels@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Sebastien Buisson [Fri, 17 Mar 2023 12:39:31 +0000 (13:39 +0100)]
LU-16646 krb: get rid of MEMORY private cache for krb creds
On client side, Kerberos credentials for root are cached in MEMORY,
but this is just in addition to the original credentials cache.
As there is no need to cache credentials twice, get rid of this
MEMORY private cache.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ifd02ee6cfafc27347b3c31e0dbbaab15190cf883
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50341
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Jonathan Calmels <jcalmels@nvidia.com>
Sebastien Buisson [Fri, 10 Mar 2023 17:02:31 +0000 (18:02 +0100)]
LU-16630 sec: improve Kerberos cross-realm trust remapping
Improve Kerberos cross-realm trust remapping by leveraging existing
Kerberos mechanisms. gss_localname() can be used to resolve usernames:
it goes through the auth_to_local translation rules in krb5.conf and
thus can easily be configured by security administrators.
This new mechanism does not replace the existing and rudimentary
mapping based on /etc/lustre/idmap.conf. If /etc/lustre/idmap.conf
exists, it is used for user mapping. If not, the new mechanism based
on gss_localname() gets involved.
But we now print a warning that idmap.conf is deprecated if we detect
it is in use.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Iaaf15a757dc246673e2f412181219cc978079fab
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50259
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Jonathan Calmels <jcalmels@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Oleg Drokin [Sat, 4 Dec 2021 07:30:56 +0000 (02:30 -0500)]
LU-15515 contrib: epython scripts to python3 conversion
This is to allow them to work with newer pykdump versions
that are actually current and therefore desirable.
Test-Parameters: trivial
Change-Id: I84a1be5f24c04b61d2ab1f8d1744e3ef752ce8c6
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/46442
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Alex Zhuravlev [Fri, 10 Apr 2020 08:06:38 +0000 (11:06 +0300)]
LU-13444 tests: sanity to set PTLDEBUG
otherwise any server remount reset full debug to the default
value different from full debug we want in sanity.
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I2b5c5a4d2a62d11f0327686b11f9ab8f16b47b4f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/38194
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Andreas Dilger [Fri, 31 Mar 2023 02:37:08 +0000 (20:37 -0600)]
LU-16677 utils: add bitfields and ifdefs to wiretest
Add CHECK_BITFIELD() for "checking" bitfields in data structures
(currently only adds a comment to wiretest.c, maybe improve later).
Add CHECK_COND_START/FINISH() for adding #ifdef/#endif conditions
into wiretest, mainly for structs not used by (upstream) client.
Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I8ecc4bdd5b5d651faa42f65ce8ea46da003ebbe5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50479
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Wed, 29 Mar 2023 21:39:33 +0000 (16:39 -0500)]
LU-16652 tests: sanity-lnet test 253/254 race with discovery
The delay rule used for test 253/254 can race with incoming discovery
PUSH (PUT + ACK). This results in an extra GET being queued on the
peer NI tx queue and subsequently dropped.
Fix this issue by restricting the delay rule to GET message type.
Test-Parameters: trivial testlist=sanity-lnet env=ONLY=253,254
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Ie66cc5e00cc3718e74221b080e099958cef7a401
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50469
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Alexander Zarochentsev [Tue, 28 Mar 2023 16:45:41 +0000 (19:45 +0300)]
LU-16677 utils: synchronize wirecheck.c and wiretest.c
wirecheck.c is not compilable and out of sync with
wiretest.c.
The patch adds forgotten changes to wirecheck.c and
replaces wiretest.c by a version generated by
"make newwiretest".
Test-Parameters: trivial
Signed-off-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Change-Id: I22cc75b02e163ce3ca743be608ec334c85129b4c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50456
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andrew Perepechko <andrew.perepechko@hpe.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shaun Tancheff [Fri, 24 Mar 2023 10:34:48 +0000 (05:34 -0500)]
LU-16664 build: Debian server fails building crypto.c
When building deb files against a server built without
CONFIG_FS_ENCRYPTION it still attempts to build crypto.c
when it should not.
Test-Parameters: trivial
Fixes:
068e5f13fb ("LU-13743 build: Explicitly require encryption support")
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Id1e67daa7b021fdfee49be4eb0beb2b86ca62c39
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50406
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Thomas Bertschinger [Fri, 17 Mar 2023 20:24:13 +0000 (16:24 -0400)]
LU-16503 utils: add --hex-idx option for lfs getstripe
The --hex-idx option for lfs getstripe and getdirstripe makes
them output OST and MDT indexes in hex.
In addition, this commit modifies the YAML parser to treat input
like "0x[0-9a-f]+" as a number so that lfs setstripe --yaml can
accept output produced by lfs getstripe --yaml --hex-idx.
Signed-off-by: Thomas Bertschinger <bertschinger@lanl.gov>
Change-Id: I98eaa9741db787bcf4516648e58e92c7520d8640
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50340
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Tue, 28 Feb 2023 18:48:04 +0000 (18:48 +0000)]
LU-16518 lnet: fix clang build errors
LNET_PID_ANY and LNET_NID_ANY were defined outside
the range of an u64. They were moved from -1 to the
maximum value for u64.
LNET_NID_ANY was used in an instance where LNET_NET_ANY
should have been used.
Initialize a variable which could potentially be left
uninitialized.
Fixed a minor style issue.
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I61c8b33d98fdaa4b7a4846d0061a483664e4f652
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50318
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Timothy Day [Tue, 14 Mar 2023 21:11:09 +0000 (21:11 +0000)]
LU-9619 misc: add Test-Parameters to git-hook
Add commented out example Test-Parameters lines
to git hook. This makes it easier to add them
in when appropriate.
Add short explaination for test parameters and
for Developer Certificate of Origin.
Update the file header to have the SPDX
license and use the standard format.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I6af3075339f15a3c4fc4134111f47a2899005c56
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50298
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Charlie Olmstead <charlie@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Serguei Smirnov [Thu, 9 Mar 2023 23:00:46 +0000 (15:00 -0800)]
LU-14668 tests: verify state of peer added with '--lock_prim'
Add peer state verification to sanity-lnet test_26:
check that peer state has corresponding bit set for a peer
created with '--lock_prim' option.
Test-Parameters: trivial testlist=sanity-lnet
Fixes:
05f7f6a0b ("LU-14668 lnet: add 'force' option to lnetctl peer del")
Signed-off-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Change-Id: Id5fde036907f9dd19a21e8e6611a070321310f0e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50249
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Timothy Day [Thu, 9 Mar 2023 04:55:26 +0000 (04:55 +0000)]
LU-16627 tests: add auster long options
Add long options to auster.
Update the file header to have the
SPDX license and use the standard
format.
Address a couple of small shellcheck
warnings.
Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I3f1089ce6e054cd40f0afb229f9a0f0935e96d2e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50237
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Charlie Olmstead <charlie@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Thu, 2 Mar 2023 17:44:44 +0000 (11:44 -0600)]
LU-16563 tests: Check peer NI health after link down
Validate whether peer NI health is decremented when a link is downed.
Test-Parameters: trivial testlist=sanity-lnet env=ONLY=303
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I5aa4a2f6efa2c0be2d7eb166b21ffab268e39941
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50188
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Patrick Farrell [Wed, 1 Mar 2023 21:05:02 +0000 (16:05 -0500)]
LU-16608 tests: Reduce lock count in 255c
Similar to LU-10443/LU-10136, test 255c is failing due to
lock reclamation during the test, this time on GK testing.
Reducing the lock count in the test from 500 to 100 should
dramatically reduce the frequency of failures, and is
still a good test.
Test-parameters: trivial testlist=sanity env=ONLY=255c,ONLY_REPEAT=100
Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: I47c9a60c8425b73c53093bae34995c08c8278f2b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50174
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Jian Yu [Fri, 31 Mar 2023 17:43:14 +0000 (10:43 -0700)]
LU-16221 kernel: RHEL 9.1 server support
This patch makes changes to support RHEL 9.1 release
with kernel 5.14.0-162.18.1.el9_1 for Lustre server.
Test-Parameters: trivial fstype=ldiskfs \
clientdistro=el9.1 serverdistro=el9.1 testlist=sanity
Change-Id: I4ed81f53aa6955eef06aff30a10a81f0e4124f79
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49921
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Etienne AUJAMES [Wed, 18 Jan 2023 09:42:54 +0000 (10:42 +0100)]
LU-16465 llite: fix LSOM blocks for ftruncate and close
LSOM is updated on close and setattr request.
For the setattr, clients do not know the numbers blocks yet (OSTs
setattr requests have to finish). So the blocks number is set to 1 by
the server.
The close request send after a ftruncate() will wrongly update LSOM
back to its old blocks number. This is because clients do not update
the inode.i_blocks after an OST setattr.
Then the MDS will denied a client close request to update LSOM to its
correct blocks number. Only truncates are allowed to decrease the
blocks number (server side).
This patch force the client inode update at the end of an OST setattr.
And it tries (if no contention on the inode_size) to update the inode
at the end of an OST fsync or a sync IO.
Update sanity test 806/807 for this use case.
Test-Parameters: testlist=sanity env=ONLY=806,807,ONLY_REPEAT=20
Test-Parameters: fstype=zfs testlist=sanity-flr env=ONLY=70,ONLY_REPEAT=10
Signed-off-by: Etienne AUJAMES <eaujames@ddn.com>
Change-Id: Ib1afde97071ebae56f0b413ec444403c3cdebd02
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49675
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Mon, 12 Dec 2022 05:36:07 +0000 (16:36 +1100)]
LU-16382 spec: add dependency on kernel-%1
The SUSE OBS rpmlint tool wants all kmps to depend on the base kernel.
As RH uses didn't package naming to SUSE, these dependencies must be
conditional.
Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I4f8525dd2ba35b824bcfbd4097505bdfe7b11e0f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49371
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Mon, 12 Dec 2022 05:20:32 +0000 (16:20 +1100)]
LU-16382 spec: specify all directories.
All directories added by the package should be explicitly specified.
This allows rpm to know which package owns which directories, so it
knows when to safely remove them.
Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I0ed03ca09f57587d19cb350777f2ece1ed303c7e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49370
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Mon, 12 Dec 2022 04:51:59 +0000 (15:51 +1100)]
LU-16382 spec: add more dependencies for lustre-tests
Without these packages/tools, the lustre sanity.sh test suite can fail
unexpectedly.
If dbench or iozone are absent the relevant tests are skipped rather
than failing, so that can be Recommends.
Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib30731cdb11525e4f8875734314cc955f8fe1ae4
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49368
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Shaun Tancheff [Sun, 2 Apr 2023 09:09:29 +0000 (04:09 -0500)]
LU-16350 osd-ldiskfs: no_llseek removed, dquot_transfer
Linux commit v5.19-rc2-6-g868941b14441
fs: remove no_llseek
With the removal of no_llseek, leaving .llseek set to NULL
is functionally equivalent. Only provide no_llseek if it exists.
Linux commit v5.19-rc3-6-g71e7b535b890
quota: port quota helpers mount ids
dquot_transfer adds a user namespace argument. Provide an
osd_dquot_transfer() wrapper to discard the additional
argument for older kernels.
Test-Parameters: trivial
HPE-bug-id: LUS-11376
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: If3165aed0d7b827b90e26d9f0174137d087ce57a
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49266
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shaun Tancheff [Sun, 2 Apr 2023 11:47:51 +0000 (06:47 -0500)]
LU-16328 llite: migrate_folio, vfs_setxattr
Linux commit v5.19-rc3-392-g5490da4f06d1
fs: Add aops->migrate_folio
Linux commit v5.19-rc4-52-ge33c267ab70d
mm: shrinkers: provide shrinkers with names
From Linux commit v5.19-rc5-17-g0c5fd887d2bb
acl: move idmapped mount fixup into vfs_{g,s}etxattr()
Until Linux commit v6.0-rc3-6-g6344e66970c6
xattr: constify value argument in vfs_setxattr()
Cast away const when required.
Linux commit v5.19-10313-geba2d3d79829
get rid of non-advancing variants
iov_iter_get_pages_alloc2() replaces iov_iter_get_pages_alloc()
Test-Parameters: trivial
HPE-bug-id: LUS-11358
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Id1fa6db94172c0a61008ba4b066907950bdd6473
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49265
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Shaun Tancheff [Sun, 2 Apr 2023 11:45:49 +0000 (06:45 -0500)]
LU-16327 llite: read_folio, release_folio, filler_t
Linux commit v5.18-rc5-221-gb7446e7cf15f
fs: Remove aop flags parameter from grab_cache_page_write_begin()
flags have been dropped from write_begin() and
grab_cache_page_write_begin()
Linux commit v5.18-rc5-241-g5efe7448a142
fs: Introduce aops->read_folio
Provide a ll_read_folio handler around ll_readpage
Linux commit v5.18-rc5-280-ge9b5b23e957e
fs: Change the type of filler_t
Affects read_cache_page, provides a wrapper for read_cache_page
and wrappers for filler functions
Linux commit v5.18-rc5-282-gfa29000b6b26
fs: Add aops->release_folio
Provide an ll_release_folio function based on ll_releasepage
Test-Parameters: trivial
HPE-bug-id: LUS-11357
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Ibd4ec1133c80cd0eb8400c4cd07b50e421dd35c5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49199
Tested-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Chris Horn [Tue, 4 Oct 2022 10:53:48 +0000 (05:53 -0500)]
LU-16217 tests: Add routed lst test
Modify sanity-lnet test_220 to perform two quick LST tests.
Test-Parameters: trivial
Test-Parameters: testlist=sanity-lnet env=ONLY=220
HPE-bug-id: LUS-10279
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I6b9ce85fcd232eb20c5b3e1d25eeca73856f9a82
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48800
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Kevin Zhao [Fri, 28 Oct 2022 02:05:24 +0000 (10:05 +0800)]
LU-11785 tests: fix conf-sanity/98 mount check on 64K page
This patch fix the mount option length check expectation
fail on 64K page. Since the maxopt_len is the minmium
value of page_size or 64K page_size, but the test cases
only hard code the length of option to the 4K one. This
patch add the mount options according to the page size.
Test-Parameters: trivial testlist=conf-sanity env=ONLY=98
Test-Parameters: testlist=conf-sanity env=ONLY=98 clientarch=aarch64 clientdistro=el8.6
Signed-off-by: Kevin Zhao <kevin.zhao@linaro.org>
Change-Id: Icdeb8b73308056e216c3f4ce71907b0c928d2c30
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48177
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Lai Siyao [Fri, 23 Sep 2022 02:45:19 +0000 (22:45 -0400)]
LU-11047 mdt: standardize mdt object locking
* object LOOKUP lock should be taken from parent MDT if object and
its parent are located on different MDTs, but current mdt object
locking doesn't handle this explicitly, instead it implies its
parent is on local MDT.
* PDO locking should be put into a separate function.
* with the above changes, mdt_object_lock_internal() becomes a simple
wrapper: it calls mdt_remote_object_lock_try() if object is remote,
and mdt_fid_lock() otherwise.
* the MDT object locking functions are as below:
. mdt_object_lock(): lock object, this is used in most places, and
normally lock ibits doesn't contain LOOKUP, unless the caller
knows its parent is on the same MDT.
. mdt_object_check_lock(): lock object with LOOKUP and other
ibits, it needs to check whether parent is on different MDT,
if so, take LOOKUP lock on parent MDT, and then lock other
ibits on child MDT.
. mdt_parent_lock(): take parent UPDATE lock with specific mode,
if parent is local, take PDO lock, otherwise take regular
lock.
. mdt_object_stripes_lock(): lock object which should be local,
and if it's a striped directory, lock its stripes. This is
called in operations which modify dir object and its stripes.
. mdt_object_lock_try(): lock object with trybits, the trybits
contains optional inode lock bits that can be granted. This is
called by getattr/open to fetch more inode lock bits to
client, and is also called by dir migration to lock link
parent in non-block mode to avoid deadlock.
. rename/migrate source object are locked in specific functions,
because source object and its parent may be located on
different remote MDTs.
Test-Parameters: mdscount=2 mdtcount=4 testlist=racer,racer,racer
Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: I8225cbee4a1f5db8f77399866061f12e0a4cbb47
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40764
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Qian Yingjin [Fri, 20 Nov 2020 10:38:02 +0000 (18:38 +0800)]
LU-14139 statahead: batched statahead processing
Batched metadata processing can get a big performance boost.
In this patch, it implements a batched statahead mechanism which
can also increase the performance for a directory traverse or
listing such as the command 'ls'.
For the batched statahead, one batch getattr() RPC equals to
'N' normal lookup/getattr RPCs. It can pack a number of dentry
name getting from the readdir() call and prepared lock handles
one client side lock namespace into one large batched RPC
transfering via bulk I/O to obtain ibits DLM locks and
associated attributes for a lot of files in one blow.
When MDS receives a batched getattr() RPC, it executes the sub
requests in it one by one serially.
A tunable parameter named "statahead_batch_max" is defined, it
means the maximal items can be batched and processed within one
aggregate RPC. Once the number of sub requests exceeds this
predefined limit, it will pack and trigger the batched RPC.
The batched RPC will also be triggered explictly when the
readdir() call comes to the end position of the directory or
the statahead thread exits abnormally.
Batched metadata processing can get a big performance boost.
The mdtest performance results without/with this patch series are
as follow:
mdtest-easy-stat 720.562369 kIOPS : time 118.695 seconds
mdtest-easy-stat 1218.290192 kIOPS : time 70.656 seconds
In this patch, we set statahead_batch_max=0 and disabled batched
statahead by default. It will enable accordingly once some
subsequent fixes about batched RPC have been merged.
Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I5a80c2c377093dc8b8e21341f440e3038f017ca8
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40720
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>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Alex Zhuravlev [Thu, 20 Oct 2022 05:27:48 +0000 (08:27 +0300)]
LU-11404 llite: only first sync to MDS matter
fsync() is supposed to sync metadata and data, but given file's
layout the only first MDS_SYNC matters to ensure the file creation
has been committed, everything else go to OSTs (data and attributes)
also, uid/gid/mode, EAs and ACLs must be subject to sync.
Change-Id: Iaa9095a9ace319a6dfa8840313dc536e39786847
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/33175
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Mr NeilBrown [Thu, 26 May 2022 04:34:10 +0000 (14:34 +1000)]
LU-10391 ptlrpc: change cc_nid in nrs to be struct lnet_nid
switch to struct lnet_nid and adjust accordingly.
Note the return value of nrs_crrn_cmpfn() is changed from a bool to
"0 or negative error" to make it more clear which is success and which
is failure. The end result is the same.
Test-Parameters:trivial testlist=sanityn envdefinitions=ONLY=77b
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I3e5f2e9d88d7aeccb68ebfb4c814cc4e388cbc42
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50100
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: jsimmons <jsimmons@infradead.org>
Reviewed-by: Nikitas Angelinas <nikitas.angelinas@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
git://git.whamcloud.com / fs / lustre-release.git / log