[LU-14579] GPF in lod_sub_declare_destroy

• Status: Resolved
• Project: Lustre
• Fix Version/s: Lustre 2.15.0
• Type: Bug
• Priority: Minor
• Reporter: Zhenyu Xu
• Assignee: Zhenyu Xu
• Resolution: Fixed
• Severity: 3

Description

Reproducable GPF in lod_sub_declare_destroy()

while true; do
  rm -f /mnt/lustre/f0
  lfs mirror create -N -E 1M -c -1 -E eof -N -p fast /mnt/lustre/f0
  echo xxx > f0
done &

while true; do lfs mirror split -d --mirror-id=1 /mnt/lustre/f0; done &

[ 2019.163545] general protection fault: 0000 [#1] SMP
...
[ 2019.197345] CPU: 3 PID: 21982 Comm: mdt01_003 Kdump: loaded Tainted: G        W  OE  ------------   3.10.0-1062.9.1.el7.x86_64.debug #1
[ 2019.199739] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-1ubuntu1.1 04/01/2014
[ 2019.201974] task: ffff90af5be84000 ti: ffff90af5aedc000 task.ti: ffff90af5aedc000
[ 2019.203824] RIP: 0010:[<ffffffffc1686ff6>]  [<ffffffffc1686ff6>] lod_sub_get_thandle+0xc6/0x400 [lod]
[ 2019.206282] RSP: 0018:ffff90af5aedf898  EFLAGS: 00010206
[ 2019.207642] RAX: 5a5a5a5a5a5a5a5a RBX: ffff90af7a2f8600 RCX: ffff90af5aedf8ef
[ 2019.209428] RDX: ffff90af5e295f00 RSI: ffff90af7a2f8600 RDI: ffff90af088b1640
[ 2019.211160] RBP: ffff90af5aedf8d8 R08: 0000000000000010 R09: 0000000000000000
[ 2019.212692] R10: 00000000000001e0 R11: 0000000000000000 R12: ffff90af5aedf8ef
[ 2019.214704] R13: ffff90af1b2d8000 R14: ffff90af6170e510 R15: ffff90af088b1640
[ 2019.216758] FS:  0000000000000000(0000) GS:ffff90af7d400000(0000) knlGS:0000000000000000
[ 2019.218930] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 2019.220566] CR2: 00007f8052161120 CR3: 000000001385a000 CR4: 0000000000160ee0
[ 2019.222484] Call Trace:
[ 2019.223106]  [<ffffffffc1688866>] lod_sub_declare_destroy+0x56/0x2d0 [lod]
[ 2019.224926]  [<ffffffffc1664dbd>] lod_obj_stripe_destroy_cb+0x8d/0xa0 [lod]
[ 2019.226579]  [<ffffffffc1671a71>] lod_obj_for_each_stripe+0x121/0x2e0 [lod]
[ 2019.228267]  [<ffffffffc1672898>] lod_declare_destroy+0x488/0x620 [lod]
[ 2019.229954]  [<ffffffffc167388a>] ? lod_declare_attr_set+0xba/0x6c0 [lod]
[ 2019.231650]  [<ffffffffc140c50b>] ? osd_declare_ref_del+0x3b/0x130 [osd_ldiskfs]
[ 2019.233270]  [<ffffffffc1664d30>] ? lod_xattr_list+0x150/0x150 [lod]
[ 2019.234816]  [<ffffffffc1510fe1>] mdd_declare_finish_unlink+0x91/0x210 [mdd]
[ 2019.236636]  [<ffffffffc1520e0f>] mdd_unlink+0x48f/0xaa0 [mdd]
[ 2019.238222]  [<ffffffffc159ccc5>] mdt_reint_unlink+0xc25/0x1540 [mdt]
[ 2019.240197]  [<ffffffffc0ea3f07>] ? lustre_msg_buf+0x17/0x60 [ptlrpc]
[ 2019.242185]  [<ffffffffc15a16b3>] mdt_reint_rec+0x83/0x220 [mdt]
[ 2019.243764]  [<ffffffffc157d4e1>] mdt_reint_internal+0x6e1/0xb00 [mdt]
[ 2019.245690]  [<ffffffffc1588cb7>] mdt_reint+0x67/0x140 [mdt]
...

Issue Links

• is related to: LU-14648 Oops: 0000 [#1] SMP DEBUG_PAGEALLOC i...

Comments

[Gerrit Updater]

Bobi Jam ([e-mail]) uploaded a new patch: https://review.whamcloud.com/43199
Subject: LU-14579 flr: GPF in lod_sub_declare_destroy
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 295704ce4248713d9513e3ea2953079edd3f5af1

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/43199/
Subject: LU-14579 flr: GPF in lod_sub_declare_destroy
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 55d5235354d49aee0a330ad64beef4ed9004a27f

[Peter Jones]

Landed for 2.15

[Gerrit Updater]

Bobi Jam ([e-mail]) uploaded a new patch: https://review.whamcloud.com/43369
Subject: LU-14579 flr: mirror unlink and split race
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 479612952b3b3dc4561d3799523635201d415543

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/43369/
Subject: LU-14579 flr: mirror unlink and split race
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: bd7a2f9938a7edf09afd133601ca4181e109a7d0

[Peter Jones]

Second patch landed for 2.15