FreeNAS dead after starting (auto-scheduled) scrub

[asari]

Hi,

My FreeNAS server's uptime became 30 days, and automatic zpool scrub started on it (daily email sent to me), then somehow stopped in a few minutes (no ping/ssh/cifs response from it; my Nagios emailed me and complained that some mountpoints were invalid).

I woke up and found it stopped, so I pressed the power button twice and it restarted, restarted zpool scrub too, it stopped again in a few minutes.
I pressed the power button again and it restarts, then I logged into it via SSH, did `zpool scrub -s tank0`, and now it seems to work healthy.

Perhaps it will stop again tomorrow at 3:00. How can I get workaround on it (like temporarily disabling automated scrub)? Or, how can I troubleshoot (like turning on logging any messages)?

Sorry for my poor English, thanks in advance.

FreeNAS Build: FreeNAS-8.0.2-RELEASE-amd64 (8288)
Platform: Intel(R) Core(TM) i3 CPU 530 @ 2.93GHz
Memory: 6057MB
OS Version: FreeBSD 8.2-RELEASE-p3

Hardware: HP ML 110 G6 with 6 HDDs

 

[William Grzybowski]

As a workaround for now you can comment the line "daily_scrub_zfs_enable="YES"" in /conf/base/etc/periodic.conf

mount -uw / - to mount root as read-write

 

[asari]

As a workaround for now you can comment the line "daily_scrub_zfs_enable="YES"" in /conf/base/etc/periodic.conf

mount -uw / - to mount root as read-write

Thanks! I executed following commands; let me see what the daily mail will say at next 3:00 a.m.

# mount -uw /
# nano -w /conf/base/etc/periodic.conf # turn daily_scrub_zfs_enable to "NO"
# mount -ur / # this took a while

( just overwrote "YES" with "NO", because I didn't know how to comment it out...
which style of comment is available? "#", ";" or "//" ....? )

 

[asari]

I received a mail from my FreeNAS server today:

Scrubbing of zfs pools:
skipping scrubbing of pool 'tank0':
last scrubbing is 3 days ago, threshold is set to 30 days
skipping scrubbing of pool 'tank1':
last scrubbing is 29 days ago, threshold is set to 30 days​

It seems that the auto-scrub might be still scheduled. Let's wait for tomorrow and see if auto-scrub is really disabled.

 

[asari]

Although daily_scrub_zfs_enable is set to "NO", auto-scrub was not canceled and my FreeNAS died again today. Sad, I did `zpool scrub -s` manually.

Perhaps it dies because of hardware problem (like short of electricity), but how can I diagnose it? I can't see any log file in /var/log after restart...