[LU-8048] LBUG during cfs_hash_destroy()

• Type: Bug
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.9.0
• Severity: 3
• Assignee: nasf (Inactive)
• Reporter: nasf (Inactive)

Description

We hit the following LBUG during sanity test:

LustreError: 10045:0:(osd_handler.c:1044:osd_device_free()) header@ffff88007c
356980[0x0, 1, [0x200000003:0x0:0x0] hash exist]

{ LustreError: 10045:0:(osd_handler.c:1044:osd_device_free()) ....local_storage @ffff88007c3569d0 LustreError: 10045:0:(osd_handler.c:1044:osd_device_free()) ....osdzfs@ffff8 800287a1cd0osd-zfs-object@ffff8800287a1cd0 LustreError: 10045:0:(osd_handler.c:1044:osd_device_free()) }

header@ffff8800
7c356980

LustreError: 10045:0:(hash.c:1138:cfs_hash_destroy()) ASSERTION( !cfs_hash_with_assert_empty(hs) ) failed: hash lu_site_osd-zfs bucket 3(1) is not empty: 1 items left
LustreError: 10045:0:(hash.c:1138:cfs_hash_destroy()) LBUG
Pid: 10045, comm: obd_zombid

Call Trace:
[<ffffffffa04a9895>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
[<ffffffffa04a9e97>] lbug_with_loc+0x47/0xb0 [libcfs]
[<ffffffffa04c0a25>] cfs_hash_putref+0x3a5/0x480 [libcfs]
[<ffffffff8106c8ca>] ? __cond_resched+0x2a/0x40
[<ffffffffa05f02c1>] lu_site_fini+0x51/0xa0 [obdclass]
[<ffffffffa05f0370>] ? lu_cdebug_printer+0x0/0x140 [obdclass]
[<ffffffffa1388a0b>] osd_device_free+0x15b/0x3b0 [osd_zfs]
[<ffffffffa05d66bd>] class_decref+0x3ed/0x4d0 [obdclass]
[<ffffffffa05bdc6c>] obd_zombie_impexp_cull+0x61c/0xac0 [obdclass]
[<ffffffffa05be175>] obd_zombie_impexp_thread+0x65/0x190 [obdclass]
[<ffffffff81064c00>] ? default_wake_function+0x0/0x20
[<ffffffffa05be110>] ? obd_zombie_impexp_thread+0x0/0x190 [obdclass]
[<ffffffff8109e78e>] kthread+0x9e/0xc0
[<ffffffff8100c28a>] child_rip+0xa/0x20
[<ffffffff8109e6f0>] ? kthread+0x0/0xc0
[<ffffffff8100c280>] ? child_rip+0x0/0x20

Kernel panic - not syncing: LBUG
Pid: 10045, comm: obd_zombid Tainted: P --------------- 2.6.32-5
04.30.3.el6_lustre.gd2dd53d.x86_64 #1
Call Trace:
[<ffffffff8152a81c>] ? panic+0xa7/0x16f
[<ffffffffa04a9eeb>] ? lbug_with_loc+0x9b/0xb0 [libcfs]
[<ffffffffa04c0a25>] ? cfs_hash_putref+0x3a5/0x480 [libcfs]
[<ffffffff8106c8ca>] ? __cond_resched+0x2a/0x40
[<ffffffffa05f02c1>] ? lu_site_fini+0x51/0xa0 [obdclass]
[<ffffffffa05f0370>] ? lu_cdebug_printer+0x0/0x140 [obdclass]
[<ffffffffa1388a0b>] ? osd_device_free+0x15b/0x3b0 [osd_zfs]
[<ffffffffa05d66bd>] ? class_decref+0x3ed/0x4d0 [obdclass]
[<ffffffffa05bdc6c>] ? obd_zombie_impexp_cull+0x61c/0xac0 [obdclass]
[<ffffffffa05be175>] ? obd_zombie_impexp_thread+0x65/0x190 [obdclass]
[<ffffffff81064c00>] ? default_wake_function+0x0/0x20
[<ffffffffa05be110>] ? obd_zombie_impexp_thread+0x0/0x190 [obdclass]
[<ffffffff8109e78e>] ? kthread+0x9e/0xc0
[<ffffffff8100c28a>] ? child_rip+0xa/0x20
[<ffffffff8109e6f0>] ? kthread+0x0/0xc0
[<ffffffff8100c280>] ? child_rip+0x0/0x20