Google Compute: task scsi_eh_0:103 blocked for more than 120 seconds

Question

Has anyone seen these errors before? Running a g1-small Debian instance.

The server has been online for about a week and is now putting these errors to the console.

Apr 14 19:32:49 box-1 kernel: [1226907.872051] sd 0:0:1:0: [sda] abort
[1227120.480055] INFO: task scsi_eh_0:103 blocked for more than 120 seconds.
[1227120.481458] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[1227120.482970] INFO: task jbd2/sda1-8:123 blocked for more than 120 seconds.
[1227120.484136] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[1227120.485619] INFO: task box-1d:10353 blocked for more than 120 seconds.
[1227120.486721] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 14 19:36:22 box-1 kernel: [1227120.480055] INFO: task scsi_eh_0:103 blocked for more than 120 seconds.
[1227240.488055] INFO: task scsi_eh_0:103 blocked for more than 120 seconds.
[1227240.489434] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[1227240.490740] INFO: task jbd2/sda1-8:123 blocked for more than 120 seconds.
[1227240.491806] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[1227240.493144] INFO: task box-1d:10353 blocked for more than 120 seconds.
[1227240.494159] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 14 19:36:22 box-1 kernel: [1227120.481458] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 14 19:36:22 box-1 kernel: [1227120.482842] scsi_eh_0       D ffff88006c813780     0   103      2 0x00000000
Apr 14 19:36:22 box-1 kernel: [1227120.482846]  ffff8800677f9160 0000000000000046 0000000000000000 ffffffff8160d020
Apr 14 19:36:22 box-1 kernel: [1227120.482849]  0000000000013780 ffff8800376abfd8 ffff8800376abfd8 ffff8800677f9160
Apr 14 19:36:22 box-1 kernel: [1227120.482851]  00205d3135303237 0000000000000000 0000000000000000 7fffffffffffffff
Apr 14 19:36:22 box-1 kernel: [1227120.482853] Call Trace:
Apr 14 19:36:22 box-1 kernel: [1227120.482890]  [<ffffffff8134eedb>] ? schedule_timeout+0x2c/0xdb
Apr 14 19:36:22 box-1 kernel: [1227120.482905]  [<ffffffffa0073070>] ? virtscsi_map_sgl+0x3f/0x5d [virtio_scsi]
Apr 14 19:36:22 box-1 kernel: [1227120.482908]  [<ffffffff8134fdc7>] ? _raw_spin_unlock_irqrestore+0xe/0xf
Apr 14 19:36:22 box-1 kernel: [1227120.482911]  [<ffffffffa0073239>] ? virtscsi_kick_cmd+0x1ab/0x1bd [virtio_scsi]
Apr 14 19:36:22 box-1 kernel: [1227120.482912]  [<ffffffff8134eb21>] ? wait_for_common+0xa0/0x119
Apr 14 19:36:22 box-1 kernel: [1227120.482928]  [<ffffffff8103f6ba>] ? try_to_wake_up+0x197/0x197
Apr 14 19:36:22 box-1 kernel: [1227120.482930]  [<ffffffffa00732bf>] ? virtscsi_tmf+0x74/0x9c [virtio_scsi]
Apr 14 19:36:22 box-1 kernel: [1227120.482945]  [<ffffffffa003234e>] ? scsi_error_handler+0x2eb/0x5f5 [scsi_mod]
Apr 14 19:36:22 box-1 kernel: [1227120.482949]  [<ffffffffa0032063>] ? scsi_eh_get_sense+0x176/0x176 [scsi_mod]
Apr 14 19:36:22 box-1 kernel: [1227120.482957]  [<ffffffff8105f681>] ? kthread+0x76/0x7e
Apr 14 19:36:22 box-1 kernel: [1227120.482964]  [<ffffffff81356ef4>] ? kernel_thread_helper+0x4/0x10
Apr 14 19:36:22 box-1 kernel: [1227120.482966]  [<ffffffff8105f60b>] ? kthread_worker_fn+0x139/0x139
Apr 14 19:36:22 box-1 kernel: [1227120.482968]  [<ffffffff81356ef0>] ? gs_change+0x13/0x13

Answer 1

I think it's the same problem as this: Hung Tasks in EXT4 on VM

I filed a bug report here: https://code.google.com/p/google-compute-engine/issues/detail?id=73