Host stopped

Hi,
The host with Percona-Server-5.5.16 stopped with kernel panic.

Pid: 18409, comm: mysqld Tainted: G M D ---------------- 2.6.32-71.29.1.el6.x86_64 #1
Call Trace:
[] panic+0x78/0x137
[] ? print_oops_end_marker+0x23/0x30
[] die_nmi+0xfc/0x100
[] nmi_watchdog_tick+0x1aa/0x200
[] do_nmi+0x1a3/0x2d0
[] nmi+0x20/0x30
[] ? _spin_lock+0x1e/0x30
<> [] task_rq_lock+0x5d/0xa0
[] try_to_wake_up+0x51/0x380
[] wake_up_state+0x10/0x20
[] wake_futex+0x40/0x60
[] futex_requeue+0x4b8/0x890
[] do_futex+0xa6/0xb00
[] ? xfs_log_reserve+0xf0/0x150 [xfs]
[] ? xfs_trans_reserve+0xa0/0x210 [xfs]
[] ? xfs_trans_unlocked_item+0x39/0x60 [xfs]
[] ? xfs_iunlock+0x86/0x100 [xfs]
[] ? xfs_fsync+0x168/0x1c0 [xfs]
[] sys_futex+0x7b/0x170
[] ? audit_syscall_entry+0x272/0x2a0
[] system_call_fastpath+0x16/0x1b

There’s no reason to ask my question here but ‘mysqld’ at the top.
I’ve got extra worry 'cause the same panic occured 2 weeks ago on another host with the same Percona server.
Please, what can cause the failures?

CentOS 6.0
DB files lie on XFS, as you can suspect.

Hi,

Can you provide the full mysql error log? Initially it doesn’t look like MySQL issue. Check below links, many people faced this kind of issue and it resolved by replacing RAM.
http://bugs.mysql.com/bug.php?id=47733
http://www.howtoforge.com/forums/showthread.php?t=48298
https://www.centos.org/forums/viewtopic.php?t=29306