ZFS raidz1 redundancy
Boris Kochergin
spawk at acm.poly.edu
Thu Oct 11 19:38:18 PDT 2007
Hi. I'm running running an i386 -CURRENT built on October 2nd. I have a
raidz1 pool consisting of seven 400-GiB PATA disks. ad4 and ad5 are part
of the pool. This afternoon, the following happened:
Oct 11 19:05:27 exodus kernel: ad4: timeout waiting to issue command
Oct 11 19:05:27 exodus kernel: ad4: error issuing READ_DMA command
Oct 11 19:05:27 exodus root: ZFS: vdev I/O failure, zpool=home
path=/dev/ad4 offset=70362711040 size=21504 error=5
The machine proceeded to panic after that, and when it rebooted, the
following happened after a while:
Oct 11 19:11:40 exodus kernel: ad5: detached
Oct 11 19:11:40 exodus kernel: ad4: TIMEOUT - READ_DMA retrying (1 retry
left) LBA=32
It crashed again half an hour after that, and when it came back up, ad4
was no longer detected by the ATA controller. The output of "zpool
status" is as follows:
pool: home
state: FAULTED
status: One or more devices could not be opened. There are insufficient
replicas for the pool to continue functioning.
action: Attach the missing device and online it using 'zpool online'.
see: http://www.sun.com/msg/ZFS-8000-D3
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
home FAULTED 6 0 0 corrupted data
raidz1 DEGRADED 6 0 0
ad4 UNAVAIL 0 0 0 cannot open
ad5 ONLINE 0 0 0
ad10 ONLINE 0 0 0
ad11 ONLINE 0 0 0
ad8 ONLINE 0 0 0
ad9 ONLINE 0 0 0
ad6 ONLINE 0 0 0
Is it possible that the data on ad5, in the midst of the failured of
ad4, has become inconsistent with the other members of the pool, and
that I need to bring ad4 online (I'm fairly sure that it's a
motherboard- or power-related issue and that the drive is OK) to be able
to access the data on the pool?
-Boris
More information about the freebsd-current
mailing list