[LU-9097] sanity test_253 test_255: ZFS list corruption

Duplicate — Mon, 17 Apr 2017 19:09:40 +0000

This issue was created by maloo for Andreas Dilger <andreas.dilger@intel.com>

This issue relates to the following test suite run: https://testing.hpdd.intel.com/test_sets/937abbdc-e5e0-11e6-978f-5254006e85c2.

The sub-test test_253 timed out. Looking at the console log on the server, it appears to be ZFS list corruption:

00:50:58:[ 4422.775971] WARNING: at lib/list_debug.c:59 __list_del_entry+0xa1/0xd0()
00:50:58:[ 4422.775971] list_del corruption. prev->next should be ffffc900031a3010, but was           (null)
00:50:58:[ 4422.795082] CPU: 0 PID: 32 Comm: kswapd0 Tainted: P           OE  ------------   3.10.0-514.2.2.el7_lustre.x86_64 #1
00:50:58:[ 4422.815142] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2007
00:50:58:[ 4422.815142] Call Trace:
00:50:58:[ 4422.815142]  [<ffffffff81686318>] dump_stack+0x19/0x1b
00:50:58:[ 4422.815142]  [<ffffffff81085940>] warn_slowpath_common+0x70/0xb0
00:50:58:[ 4422.815142]  [<ffffffff810859dc>] warn_slowpath_fmt+0x5c/0x80
00:50:58:[ 4422.815142]  [<ffffffff81333301>] __list_del_entry+0xa1/0xd0
00:50:58:[ 4422.815142]  [<ffffffff8133333d>] list_del+0xd/0x30
00:50:58:[ 4422.815142]  [<ffffffffa065cf1d>] __spl_cache_flush+0xed/0x150 [spl]
00:50:58:[ 4422.815142]  [<ffffffffa065d046>] spl_cache_flush+0x36/0x50 [spl]
00:50:58:[ 4422.815142]  [<ffffffffa065d71f>] spl_kmem_cache_reap_now+0x10f/0x120 [spl]
00:50:58:[ 4422.815142]  [<ffffffffa070b3c9>] arc_kmem_reap_now+0x79/0xe0 [zfs]
00:50:58:[ 4422.815142]  [<ffffffffa0710bb7>] arc_shrinker_func+0x97/0x130 [zfs]
00:50:58:[ 4422.815142]  [<ffffffff81194213>] shrink_slab+0x163/0x330
00:50:58:[ 4422.815142]  [<ffffffff811f5361>] ? vmpressure+0x21/0x90
00:50:58:[ 4422.815142]  [<ffffffff81198001>] balance_pgdat+0x4b1/0x5e0
00:50:58:[ 4422.815142]  [<ffffffff811982a3>] kswapd+0x173/0x450

Please provide additional information about the failure here.

Info required for matching: sanity 253
Info required for matching: sanity 255

Whamcloud Community JIRA

[LU-9097] sanity test_253 test_255: ZFS list corruption