Safest way to shut down with Fatal Trap 12 while in Kernel Mode / next steps to troubleshoot

Status
Not open for further replies.

NASbox

Guru
Joined
May 8, 2012
Messages
650
Any insight or suggestions as to next steps would be much appreciated as to the best way to proceed.

I was powering off my FreeNAS box and I got the error
'Fatal Trap 12 while in Kernel Mode'
Fault code 'supervisor read data' page not present'
and the box is hung with a prompt db>

For now I am leaving the box powered on in this condition until I get some input from someone who understands the finer details of FreeBSD.

I got 'GEOM_ELI: Detached xxxxx on last close.' for all drives in the ZFS pools, so I would assume that these means the ZFS pools are closed successfully. Correct assumption?

What is the safest way to shut down or is there no choice but to just kill the power?


I am running FreeNAS 8.3.1.

Any suggestions as to cause? Is this likely transient, or should I be looking for a hardware problem or some sort?

Important Background:
We had a power failure yesterday (all systems on UPS). The first box I shut down was the FreeNAS, and then the power came back on. There was a windows XP box connected to the FreeNAS, no I/O happening, just CIFS shares pointing to a couple of directories on the ZFS pool.

The reason I shut the system down was that I got an error "device not configured" when accessing (attempted ls on a couple of directories) a small USB drive formatted in UFS that is used for scripts (doesn't see much write action, mainly read).

The last thing I did prior to shutting down was a zpool status, and everything was OK on the pools.

My plan is to reboot everything and see if the problem clears.

Thanks in advance for any help or advice.
 
Last edited by a moderator:

rs225

Guru
Joined
Jun 28, 2014
Messages
878
That is a debug prompt, which can get some information about the crash. Probably not useful, since it isn't a current OS version.

Unless you think you want to get some of that debug info, you just reboot. If the problem continues, it means your pool is probably corrupt. Importing in readonly mode often works so you can copy data out.

Other things to do is run a memory test, run a scrub, and consider upgrading the OS.

Nothing you described sounds like a direct cause of this problem, but do keep looking for clues.
 

NASbox

Guru
Joined
May 8, 2012
Messages
650
Thanks rs225

Is there any debug info you think would be helpful? If so how do I get it?

Thanks
 

NASbox

Guru
Joined
May 8, 2012
Messages
650
Thanks rs225

Machine froze (I think it was because I have a USB KVM switch on the box and I left it unswitched from the kb for too long... and I've had programs like MemTest86+ and other programs that don't have a heavy duty OS either crash or fail to respond to the keyboard after it has been switched away for too long. I use only one keyboard and monitor due to space limitations, and the box is normally run headless) so the only thing I could do was hit the reset button.

Scrub was good before I shut down and this problem occurred.

Finished 3 passes though MemTest86+ without any errors.

I suspect it might be an issue with the UFS formatted USB that I use for my custom scripts.
What is the best way to check it?

I rebooted the system and everything seems fine now. If it doesn't happen again, then no problem. If the problem reoccurs, then I need to start looking.

Thanks for your help.
 
Last edited:
Status
Not open for further replies.
Top