Bug#532386: Same crash but on SAN-storage
Hi all,
we had the same problem already twice in two month. Although there are some
differences. We use the etch-backport kernel:
linux-image-2.6.26-bpo.2-amd64 2.6.26-17~bpo40+1 Linux 2.6.26 image on AMD64
We write to a SAN device which is attached via FC. Our server didn't crash but
all application writing do the affected device blocked. In the below log it was
Apache. The only way to get rid of this problem was to reboot the server. We
couldn't do a umount because the process writing to the device couldn't be
killed.
Below the the kernel log. The message only repeated so I just put the first
lines here. If you need any further information please ask.
[2964960.883183] INFO: task httpd:3888 blocked for more than 120 seconds.
[2964960.883211] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
this message.
[2964960.883246] httpd D 0000000000000000 0 3888 18809
[2964960.883274] ffff81012a16b8e8 0000000000000086 0000000000000000
0000000000000000
[2964960.883312] ffff8102189e29d0 ffffffff804fa480 ffff8102189e2c58 0000000000000000
[2964960.883366] 0000000000000000 0000000000000000 00000000ffffffff
0000000000000000
[2964960.883395] Call Trace:
[2964960.883447] [<ffffffff8042a116>] schedule_timeout+0x1e/0xad
[2964960.883488] [<ffffffff8042a87c>] __down+0x64/0x91
[2964960.883511] [<ffffffff80249a7e>] down+0x27/0x36
[2964960.883565] [<ffffffffa02a4204>] :xfs:xfs_buf_iowait+0x3c/0x45
[2964960.883611] [<ffffffffa02a5a73>] :xfs:xfs_buf_read_flags+0x4e/0x7f
[2964960.887103] [<ffffffffa029adbf>] :xfs:xfs_trans_read_buf+0x47/0x2af
[2964960.887103] [<ffffffffa0288afe>] :xfs:xfs_imap_to_bp+0x3f/0xfa
[2964960.887103] [<ffffffffa028836c>] :xfs:xfs_imap+0x6a/0x134
[2964960.891260] [<ffffffffa0288c59>] :xfs:xfs_itobp+0xa0/0xe7
[2964960.891308] [<ffffffffa028ae61>] :xfs:xfs_iread+0x79/0x1f0
[2964960.891356] [<ffffffffa0286541>] :xfs:xfs_iget_core+0x2de/0x51d
[2964960.891407] [<ffffffffa0286859>] :xfs:xfs_iget+0xd9/0x17c
[2964960.891454] [<ffffffffa029df22>] :xfs:xfs_lookup+0x73/0x91
[2964960.891501] [<ffffffffa02a8e99>] :xfs:xfs_vn_lookup+0x3b/0x7a
[2964960.891528] [<ffffffff802a2435>] do_lookup+0xd7/0x1c1
[2964960.891554] [<ffffffff802a45d1>] __link_path_walk+0x88e/0xd1e
[2964960.891580] [<ffffffff803b0d09>] sock_aio_read+0x103/0x11b
[2964960.891607] [<ffffffff802a4aa7>] path_walk+0x46/0x8b
[2964960.891630] [<ffffffff802a4dd3>] do_path_lookup+0x158/0x1cf
[2964960.891656] [<ffffffff802a3bbd>] getname+0x140/0x1a7
[2964960.891680] [<ffffffff802a5742>] __user_walk_fd+0x37/0x4c
[2964960.891704] [<ffffffff8029e81b>] vfs_stat_fd+0x1b/0x4a
[2964960.891727] [<ffffffff802464c1>] autoremove_wake_function+0x0/0x2e
[2964960.891755] [<ffffffff803b1fdf>] move_addr_to_user+0x5d/0x78
[2964960.891780] [<ffffffff8029e8a6>] sys_newstat+0x19/0x31
[2964960.891806] [<ffffffff8029bd20>] vfs_read+0x11e/0x152
[2964960.891830] [<ffffffff8029c0a8>] sys_read+0x60/0x6e
[2964960.891853] [<ffffffff8020beca>] system_call_after_swapgs+0x8a/0x8f
Regards and cheers, Markus
Reply to: