RHEL6.5: soft lockups with HP perfd / scopeux in on a machine with a lot of disks
Issue
- Seeing periodic soft lockup backtraces which eventually clear, all look similar to the below, with
disk_seqf_start -> class_dev_iter_next -> klist_next -> _spin_lock
Aug 20 20:19:22 localhost kernel: <EOI>
Aug 20 20:19:22 localhost kernel: [<ffffffff811ae790>] ? seq_read+0x0/0x400
Aug 20 20:19:22 localhost kernel: [<ffffffff8150dcc0>] ? klist_release+0x0/0xd0
Aug 20 20:19:22 localhost kernel: [<ffffffff8152bcc1>] ? _spin_lock+0x21/0x30
Aug 20 20:19:22 localhost kernel: [<ffffffff8150db0f>] ? klist_next+0x2f/0xf0
Aug 20 20:19:22 localhost kernel: [<ffffffff8136f610>] ? class_dev_iter_next+0x20/0x60
Aug 20 20:19:22 localhost kernel: [<ffffffff8127140c>] ? disk_seqf_start+0x8c/0xb0
Aug 20 20:19:22 localhost kernel: [<ffffffff811ae826>] ? seq_read+0x96/0x400
Aug 20 20:19:22 localhost kernel: [<ffffffff8105af90>] ? __dequeue_entity+0x30/0x50
Aug 20 20:19:22 localhost kernel: [<ffffffff81069290>] ? pick_next_task_fair+0xd0/0x130
Aug 20 20:19:22 localhost kernel: [<ffffffff811f3f8e>] ? proc_reg_read+0x7e/0xc0
Aug 20 20:19:22 localhost kernel: [<ffffffff81189715>] ? vfs_read+0xb5/0x1a0
Aug 20 20:19:22 localhost kernel: [<ffffffff81189851>] ? sys_read+0x51/0x90
Aug 20 20:19:22 localhost kernel: [<ffffffff8100b072>] ? system_call_fastpath+0x16/0x1b
Environment
- Red Hat Enterprise Linux 6.5
- seen on kernel 2.6.32-431.63.1.el6
- HP performance monitoring tools (perfd, scopeux, etc)
- thousands of disks
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.
Close
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.