[LU-15843] Crash when umount mdt targets lnet with llstat running.

Related — Fri, 13 May 2022 14:55:36 +0000

This issue was discovered while analyzing the root cause of LU-15759.

Reproducer

# llstat -i 5 mds.MDS.mdt.stats > /dev/null &
# umount -at lustre
(Crash ....)

Crash:

[151292.415792] Lustre: server umount lustrefs-MDT0001 complete
[151292.491875] LustreError: 11-0: lustrefs-MDT0001-osp-MDT0000: operation mds_disconnect to node 0@lo failed: rc = -107
[151295.103507] general protection fault: 0000 [#1] SMP 
[151295.104283] Modules linked in: osp(OE) mdd(OE) lod(OE) mdt(OE) lfsck(OE) mgs(OE) mgc(OE) osd_ldiskfs(OE) ldiskfs(OE) lquota(OE) lustre(OE) lmv(OE) mdc(OE) lov(OE) osc(OE) fid(OE) fld(OE) ksocklnd(OE) ptlrpc(OE) obdclass(OE) lnet(OE) libcfs(OE) dm_flakey mbcache jbd2 rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace fscache sunrpc dm_snapshot dm_bufio iosf_mbi ppdev crc32_pclmul snd_intel8x0 snd_ac97_codec ac97_bus snd_seq ghash_clmulni_intel snd_seq_device snd_pcm aesni_intel lrw gf128mul glue_helper ablk_helper cryptd snd_timer sg snd pcspkr i2c_piix4 soundcore parport_pc parport video ip_tables xfs libcrc32c sr_mod cdrom sd_mod crc_t10dif crct10dif_generic ata_generic pata_acpi vmwgfx drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm crct10dif_pclmul crct10dif_common
[151295.110133]  crc32c_intel ahci serio_raw libahci ata_piix drm e1000 libata drm_panel_orientation_quirks dm_mirror dm_region_hash dm_log dm_mod [last unloaded: libcfs]
[151295.111650] CPU: 1 PID: 6578 Comm: llstat Kdump: loaded Tainted: G           OE  ------------   3.10.0-1160.59.1.el7.centos.plus.x86_64 #1
[151295.113155] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[151295.113907] task: ffff931a5a4c9080 ti: ffff9319bc80c000 task.ti: ffff9319bc80c000
[151295.114663] RIP: 0010:[<ffffffffc0a129a1>]  [<ffffffffc0a129a1>] lprocfs_stats_collect+0xc1/0x140 [obdclass]
[151295.116235] RSP: 0018:ffff9319bc80fdd8  EFLAGS: 00010202
[151295.117017] RAX: 0000000000004669 RBX: ffff9319bc80fe10 RCX: 0000000000000006
[151295.117778] RDX: dead000000000100 RSI: dead000000000100 RDI: 0000000000000006
[151295.118573] RBP: ffff9319bc80fe00 R08: 0000000000000000 R09: 0000000000000000
[151295.119742] R10: 0000000000000000 R11: ffff9319bc80fc56 R12: ffff931a46694800
[151295.120652] R13: 0000000000000000 R14: ffff931a57bc7000 R15: ffff93199d70d0c0
[151295.121387] FS:  00007fe897599740(0000) GS:ffff931a9fc80000(0000) knlGS:0000000000000000
[151295.122150] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[151295.122888] CR2: 0000000001a11b24 CR3: 0000000069326000 CR4: 00000000000606e0
[151295.123637] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[151295.124336] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[151295.125016] Call Trace:
[151295.125709]  [<ffffffffc0a13308>] lprocfs_stats_seq_show+0x48/0x140 [obdclass]
[151295.126379]  [<ffffffff82476d10>] seq_read+0x130/0x450
[151295.127033]  [<ffffffff8244e3ff>] vfs_read+0x9f/0x170
[151295.127647]  [<ffffffff8244f27f>] SyS_read+0x7f/0xf0
[151295.128248]  [<ffffffff829aaed5>] ? system_call_after_swapgs+0xa2/0x13a
[151295.128839]  [<ffffffff829aaf92>] system_call_fastpath+0x25/0x2a
[151295.129439]  [<ffffffff829aaed5>] ? system_call_after_swapgs+0xa2/0x13a
[151295.130033] Code: c1 e0 03 0f 1f 80 00 00 00 00 48 63 d1 49 83 7c d4 20 00 74 48 4c 89 c2 49 03 54 fc 20 41 f6 44 24 04 02 4a 8d 34 0a 48 0f 45 d6 <48> 8b 32 48 01 33 48 8b 72 20 48 01 73 20 48 8b 72 08 48 3b 73 
[151295.131885] RIP  [<ffffffffc0a129a1>] lprocfs_stats_collect+0xc1/0x140 [obdclass]
[151295.132514]  RSP <ffff9319bc80fdd8>

Whamcloud Community JIRA

[LU-15843] Crash when umount mdt targets lnet with llstat running.