System crashes when removing large files on 3PAR disk that is mounted with discard option.

Solution Unverified - Updated -

Issue

  • When hung_task_panic is enabled, system crashes while removing large files on 3PAR disk that is mounted with discard option.
INFO: task rm:24315 blocked for more than 300 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
rm      D 0000000000000001     0 24315  13602 0x00000004
 ffff880818d11d08 0000000000000082 0000000000000000 ffff880813ce4dc8
 ffff880817ec4ae0 ffff88102593dc00 0000000000001000 ffff880822baaac0
 ffff88082a817ab8 ffff880818d11fd8 000000000000fb88 ffff88082a817ab8
Call Trace:
 [<ffffffffa013604a>] start_this_handle+0x25a/0x480 [jbd2]
 [<ffffffff81090857>] ? bit_waitqueue+0x17/0xd0
 [<ffffffff81090990>] ? autoremove_wake_function+0x0/0x40
 [<ffffffffa0136455>] jbd2_journal_start+0xb5/0x100 [jbd2]
 [<ffffffffa017dbd8>] ext4_journal_start_sb+0x58/0x90 [ext4]
 [<ffffffffa016cb7d>] ext4_unlink+0x9d/0x2b0 [ext4]
 [<ffffffff81183d5f>] vfs_unlink+0x9f/0xe0
 [<ffffffff81182aaa>] ? lookup_hash+0x3a/0x50
 [<ffffffff81186293>] do_unlinkat+0x183/0x1c0
 [<ffffffff811b5d0e>] ? dnotify_flush+0x7e/0x140
 [<ffffffff8117389d>] ? filp_close+0x5d/0x90
 [<ffffffff81186432>] sys_unlinkat+0x22/0x40
 [<ffffffff8100b072>] system_call_fastpath+0x16/0x1b
Kernel panic - not syncing: hung_task: blocked tasks
  • DISCARDs cause a continuous stream of errors if filesystem is mounted with "-o discard" and also cause mkfs to fail compeletely with the "-E discard" option.

Environment

  • Red Hat Enterprise Linux 6
  • HP Proliant BL460 Gen8
  • kernel 2.6.32-279.19.1.el6

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content