[LU-15671] conf-sanity test_30b: MDS assertion in osp_precreate_send req->rq_transno == 0

• Type: Bug
• Resolution: Fixed
• Priority: Critical
• Fix Version/s: Lustre 2.16.0
• Affects Version/s: Lustre 2.16.0, Lustre 2.15.3
• Labels: LTS15 failing_tests
• Severity: 3
• Assignee: Sergey Cheremencev
• Reporter: Maloo

Description

This issue was created by maloo for S Buisson <[e-mail]>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/d3ce7316-3e9e-4249-abb1-799a5eb3934c

test_30b failed with the following error:

trevis-67vm4, trevis-67vm5 crashed during conf-sanity test_30b

Both MDSes crashed with the following stack trace:

[ 5524.915839] LustreError: 167-0: lustre-OST0000-osc-MDT0002: This client was evicted by lustre-OST0000; in progress operations using this service will fail.
[ 5524.918788] LustreError: Skipped 1 previous similar message
[ 5524.920502] Lustre: lustre-OST0000-osc-MDT0002: Connection restored to 10.240.41.194@tcp (at 10.240.41.194@tcp)
[ 5524.922306] Lustre: Skipped 1 previous similar message
[ 5524.954841] LustreError: 17882:0:(osp_precreate.c:683:osp_precreate_send()) ASSERTION( req->rq_transno == 0 ) failed: 
[ 5524.956895] LustreError: 17882:0:(osp_precreate.c:683:osp_precreate_send()) LBUG
[ 5524.958304] Pid: 17882, comm: osp-pre-0-2 3.10.0-1160.59.1.el7_lustre.ddn16.x86_64 #1 SMP Wed Mar 9 19:03:32 UTC 2022
[ 5524.960268] Call Trace:
[ 5524.960880] [<0>] libcfs_call_trace+0x90/0xf0 [libcfs]
[ 5524.961867] [<0>] lbug_with_loc+0x4c/0xa0 [libcfs]
[ 5524.962863] [<0>] osp_precreate_send+0x1126/0x11d0 [osp]
[ 5524.963851] [<0>] osp_precreate_thread+0x6ba/0x13b0 [osp]
[ 5524.964945] [<0>] kthread+0xd1/0xe0
[ 5524.965682] [<0>] ret_from_fork_nospec_begin+0x21/0x21
[ 5524.966771] [<0>] 0xfffffffffffffffe
[ 5524.967479] Kernel panic - not syncing: LBUG
[ 5524.968288] CPU: 1 PID: 17882 Comm: osp-pre-0-2 Kdump: loaded Tainted: G           OE  ------------   3.10.0-1160.59.1.el7_lustre.ddn16.x86_64 #1
[ 5524.970699] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
[ 5524.971780] Call Trace:
[ 5524.972295]  [<ffffffff9d9865b9>] dump_stack+0x19/0x1b
[ 5524.973283]  [<ffffffff9d9802c1>] panic+0xe8/0x21f
[ 5524.974207]  [<ffffffffc0845a1b>] lbug_with_loc+0x9b/0xa0 [libcfs]
[ 5524.975364]  [<ffffffffc13c3326>] osp_precreate_send+0x1126/0x11d0 [osp]
[ 5524.976632]  [<ffffffffc13c47ca>] osp_precreate_thread+0x6ba/0x13b0 [osp]
[ 5524.977901]  [<ffffffff9d2c6f50>] ? wake_up_atomic_t+0x30/0x30
[ 5524.978998]  [<ffffffffc13c4110>] ? osp_init_pre_fid+0x630/0x630 [osp]
[ 5524.980214]  [<ffffffff9d2c5e61>] kthread+0xd1/0xe0
[ 5524.981143]  [<ffffffff9d2c5d90>] ? insert_kthread_work+0x40/0x40
[ 5524.982283]  [<ffffffff9d999df7>] ret_from_fork_nospec_begin+0x21/0x21
[ 5524.983492]  [<ffffffff9d2c5d90>] ? insert_kthread_work+0x40/0x40

VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
conf-sanity test_30b - trevis-67vm4, trevis-67vm5 crashed during conf-sanity test_30b

Issue Links

• is duplicated by: LU-16264 assertion in osp_precreate_send req->rq_transno == 0
• is related to: LU-10594 conf-sanity test_30b: FAIL: check lustre-OST0000.failover.node failed!