I'm seeing a hang in ZFS-related processes, occasionally when various nighttime jobs are running. This is on 9.2-RELEASE/amd64. I have two ZFS pools. The maxwell pool is exporting one NFS filesystem; the paydirt filesystem doesn't export anything.
Here's what ps shows me for the three hung processes:
Code:
pool: maxwell
state: ONLINE
scan: scrub repaired 0 in 0h49m with 0 errors on Mon Dec 2 04:01:04 2013
config:
NAME STATE READ WRITE CKSUM
maxwell ONLINE 0 0 0
mirror-0 ONLINE 0 0 0
gpt/max1 ONLINE 0 0 0
gpt/max0 ONLINE 0 0 0
errors: No known data errors
pool: paydirt
state: ONLINE
status: One or more devices are faulted in response to IO failures.
action: Make sure the affected devices are connected, then run 'zpool clear'.
see: http://illumos.org/msg/ZFS-8000-HC
scan: scrub repaired 0 in 2h8m with 0 errors on Tue Dec 24 05:21:06 2013
config:
NAME STATE READ WRITE CKSUM
paydirt ONLINE 2 0 0
mirror-0 ONLINE 10 0 0
gpt/paydirt2 ONLINE 4 0 0
gpt/paydirt1 ONLINE 2 0 0
errors: 2 data errors, use '-v' for a list
Here's what ps shows me for the three hung processes:
Code:
UID PID PPID CPU PRI NI VSZ RSS MWCHAN STAT TT TIME COMMAND
0 5626 5624 0 52 0 37624 2508 db->db_c D ?? 0:00.01 zfs list -H
0 816 1 0 20 0 12080 1276 zio->io_ Ds ?? 0:00.48 /usr/sbin/mountd -r /etc/exports /etc/zfs/exports
65534 93232 93224 0 52 5 9944 2860 db->db_c DN ?? 0:00.28 find -s / ! ( -fstype ufs -or -fstype zfs ) -prune -or -path /tmp -prune -or -path /usr/tmp -prune -or -path /var/tmp -prune -or -path /var/db/portsnap -prune -or -name .zfs -type d -prune -or -print