"Fatal trap 12" following power cut

[brunnenguk]

I've run in to the following issue with an HP Microserver that I've had running as a FreeNAS box:

Following a power cut earlier on in the day, the box wouldn't come back up again. The box in question normally runs headless (so far FreeNAS has been very stable so I've rarely needed to connect a monitor).

After connecting the monitor I saw that the following error was showing on screen: "Fatal trap 12: page fault while in kernel mode"

I've done some searching for this and most sources suggest that this could be a memory issue. I'm a little uncertain as there are a series of "SyntaxError: Invalid syntax" errors that were also presented.

For example:
File: "/usr/local/lib/python2.7/os.py", line 398, in <module> import UserDict

The first point where this occurs is just after:
Mounting local file systems

The next line I saw is attached as an image.

Can anyone suggest any possible causes, resolutions and whether my data is probably lost? (with 2.5Tb worth of storage, I haven't found any suitable method of backing up so I've been relying on ZFS mirrors, not a good idea in hindsight)

 

[danb35]

Unlikely your data is lost. Try a clean install of FreeNAS onto a fresh USB stick and see if it will boot OK. If so, see if it will auto-import your pool. If so, upload your saved config file and the system should be back as it was. If not, some more detail (version of FreeNAS, details of your hardware and pool configuration, etc.) would likely be helpful.

 

[brunnenguk]

Thanks for the input danb35

I believe my version is 9.2.1.8, as FreeNAS-9.2.1.8-RELEASE-x64.GUI_Upgrade.txz is the most recent update file in my downloads folder. I believe I took a configuration backup prior to upgrading, but I'm not sure whether I have a post upgrade configuration file available (I've been quite pedantic about keeping up to date config filebackups in the past, stupidly I've become sadly lax lately)

If I lose my configuration, will that be limited to users, network shares, plugins etc. If so, I think I could live with it as recovering the data is my priority, I can potentially set up things like that later.

As I have a spare box, would it be worth cloning the disks attempting a zpool import in the other box? (a little bit like a lifeboat type scenario)

I'll grab together as much information about hardware/software configurations as I can (although from memory, my zpool consists of 4 disks, 2x3Tb and 2x2Tb. each pair is mirrored and the two pairs are treated as separate volumes. beyond that, I'll try to get as much data as I can)

 

[danb35]

You should be able to do a clean install of 9.2.1.8 and upload your pre-9.2.1.8 config file. Most likely (and least destructive) scenario is that your USB stick went flaky since your last boot. Clean install onto a fresh USB stick will mostly likely solve it.

 

[brunnenguk]

Thanks for the help.

I didn't have a config file to hand but I was able to import my zpool. Due to the lack of configuration file (entirely my fault), I had to set up things like CIFS shares again but that was only a minor inconvenience in the grand scheme of things.