My Oracle Support Banner

OL7: XFS deadlocks in UEK4 kernel + xfs_log_force + xfs_log_worker (Doc ID 2336764.1)

Last updated on AUGUST 04, 2018

Applies to:

Oracle VM - Version 3.2.1 and later
Linux OS - Version Oracle Linux 7.2 with Unbreakable Enterprise Kerne [4.1.12] to Oracle Linux 7.4 with Unbreakable Enterprise Kerne [4.1.12] [Release OL7U2 to OL7U4]
Linux x86-64

Symptoms

Oracle Linux (OL) 7 server may present below behavior:

After vmcore analysis are presented follows patterns in calltraces:

xfs_log_worker()
xfs_log_force()
_xfs_log_force()
xlog_cil_force()
xlog_cil_force_lsn()
xlog_cil_push_now()
flush_work()

Example

[9146280.671820] INFO: task kworker/0:0H:9294 blocked for more than 120 seconds.
[9146280.674773] Tainted: G O 4.1.12-61.1.25.el7uek.x86_64 #2
[9146280.677791] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[9146280.681412] kworker/0:0H D 0000000000000000 0 9294 2 0x00000080
[9146280.681522] Workqueue: xfs-log/dm-1 xfs_log_worker [xfs]
[9146280.681526] ffff880001d97cb8 0000000000000046 ffff880085c59c00 ffff8800bfe7f000
[9146280.681529] ffff880035e24128 ffff880001d98000 ffff880035e24128 0000000000000000
[9146280.681532] 0000000000000000 ffff8800bfe7f000 ffff880001d97cd8 ffffffff8171d597
[9146280.681534] Call Trace:
[9146280.681541] [<ffffffff8171d597>] schedule+0x37/0x90
[9146280.681584] [<ffffffffa02470df>] _xfs_log_force_lsn+0x29f/0x2f0 [xfs]
[9146280.681589] [<ffffffff810b49c0>] ? wake_up_state+0x20/0x20
[9146280.681622] [<ffffffffa0242903>] xfs_trans_commit+0x1f3/0x240 [xfs]
[9146280.681651] [<ffffffffa0219aec>] xfs_sync_sb+0x5c/0xa0 [xfs]
[9146280.681683] [<ffffffffa0246e3d>] xfs_log_worker+0x4d/0x50 [xfs]
[9146280.681690] [<ffffffff8109f145>] process_one_work+0x155/0x460
[9146280.681693] [<ffffffff8109f942>] worker_thread+0x112/0x520
[9146280.681696] [<ffffffff8109f830>] ? rescuer_thread+0x3e0/0x3e0
[9146280.681700] [<ffffffff810a5718>] kthread+0xd8/0xf0
[9146280.681703] [<ffffffff810a5640>] ? kthread_create_on_node+0x1b0/0x1b0
[9146280.681707] [<ffffffff81721b62>] ret_from_fork+0x42/0x70
[9146280.681710] [<ffffffff810a5640>] ? kthread_create_on_node+0x1b0/0x1b0

 

 

 

 

  

Changes

No changes.

Cause

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


In this Document
Symptoms
Changes
Cause
Solution
References


My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.