Kernel panic when importing mirrored pool

[Yunior]

Hi all,

Two days ago I found out my freenas had crashed. I was forced to reinstall the OS and since them I've been unable to import my mirrored zpool (Raid-1). I keeps giving me this message that it was previously accessed by another system with a different hostid that i can try using -f to import the pool. But every time I try to (zpool import -f mypool) freenas crashes seconds later due to a kernel panic. The drives are healthy and freenas is letting import other pools I have without issues. Please help I'm running out of options I have been looking for a solution for the past 2 days, this pool holds all my critical data including years of pictures of my kids.

 

[nojohnny101]

Please post your hardware details so help can be provided.

And raid 1 doesn't mean anything here. Did you mean raidz1?

 

[Yunior]

pool: NAS
id: 5340985150341367641
state: ONLINE
status: The pool was last accessed by another system.
action: The pool can be imported using its name or numeric identifier and
the '-f' flag.
see: http://illumos.org/msg/ZFS-8000-EY
config:

NAS ONLINE
mirror-0 ONLINE
gptid/f9a483a8-8a8f-11e2-b974-0013d45bb373 ONLINE
gptid/fa0b5e5f-8a8f-11e2-b974-0013d45bb373 ONLINE

cannot import 'NAS': pool may be in use from other system, it was last accessed by freenas.home.local (hostid: 0xe8915981) on Fri Ju
n 10 08:50:42 2016
use '-f' to import anyway

Note: zpool import -f mypoolname, will not work system crashes.

 

[nojohnny101]

what is your hardware?

 

[Yunior]

FreeBSD 10.3-STABLE amd64

CPU: Intel(R) Pentium(R) 4 CPU 3.00GHz (3000.97-MHz K8-class CPU)
FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
cpu0: <ACPI CPU> on acpi0
cpu1: <ACPI CPU> on acpi0
SMP: AP CPU #1 Launched!
real memory = 3346792448 (3191 MB)
avail memory = 3206336512 (3057 MB)
agp0: aperture size is 256M, detected 7932k stolen memory
ioapic0 <Version 2.0> irqs 0-23 on motherboard
cryptosoft0: <software crypto> on motherboard
acpi0: <A M I OEMRSDT> on motherboard

 

[nojohnny101]

well let's tackle one thing at a time. Why are you running 10.x? This is in alpha currently which screams "don't touch me with a 10 foot pole". There is no stable train for that so I'm not sure why you wrote that unless that was a typo.

Was your system running 10.x before it crashed (I doubt it).

 

[danb35]

Less than half the minimum RAM isn't going to help.

 

[Yunior]

Typo :) my bad

Build FreeNAS-9.10-STABLE-201606072003 (696eba7)

Hostname freenas.local
Build FreeNAS-9.10-STABLE-201606072003 (696eba7)
Platform Intel(R) Pentium(R) 4 CPU 3.00GHz
Memory 3162MB
System Time Tue Jun 14 17:34:05 PDT 2016
Uptime 5:34PM up 4:22, 0 users
Load Average 0.29, 0.24, 0.19

 

[Ericloewe]

Memory 3162MB

That is why your system is panicking.

 

[Yunior]

But I had it working fine for months with same amount of RAM? so question upgrading the ram is a must for me to recover that 2 TB pool?

 

[danb35]

Yes, it's pretty common for a substandard system to work fine, until it doesn't. The minimums are what they are for a reason. Your CPU is also far from adequate, but isn't likely the cause of your immediate problem.

 

[Yunior]

Thanks for the fast reply, really appreciated. Will upgrade to another board with 8 GB RAM and a AMD dual core @ 4.0 Ghz see if that works better.

 

[danb35]

That would be an improvement. More of an improvement would be an Intel board with a suitable CPU and ECC RAM.

 

[Yunior]

Definitively will look into that once my budget allows it :):D

 

[joeschmuck]

More of an improvement would be an Intel board with a suitable CPU and ECC RAM.

Hey, I heard that!

I'm the AMD Defender and you can use a proper AMD solution (my old server motherboard/CPU/RAM is just sitting on my desk in front of me waiting for a new purpose, maybe as another FreeNAS server for one of my kids), I did for years and others followed my lead. But @danb35 is correct, most of us run Intel CPU systems and I did jump ship but it had nothing to do with liking Intel more, it had to do with the feature set of the motherboard.

And yea, you need more RAM as that is what's causing the panics. If you did not update the pool when you updated to 9.10, you could reinstall 9.3 and "try" to access your data. If you have any jails, plugins, etc... turn them off. But in reality 9.2 needed more RAM and 9.3 it was very solid on 8GB. 7GB will actually work too.

 

[Yunior]

Unfortunately I tried 9.3.1 already and that didn't stop the kernel to scream like a girl.

 

[Robert Trevellyan]

If all else fails, try importing the pool read-only, copying your data elsewhere, then starting over with more suitable hardware.

 

[Robert Trevellyan]

I had it working fine for months with same amount of RAM? so question upgrading the ram is a must for me to recover that 2 TB pool?

An unclean shutdown can leave a pool in a state where more RAM is required to import it and clean up the mess ... or something like that ... plus it's quite likely that lack of RAM cause the initial crash, leading to a corrupted pool.

 

[Yunior]

Nope not working either, my freenas just screamed "aahh!!! a mouse" and it crashed after that

 

[Robert Trevellyan]

What is the output of zpool import -fFn? Please post it between CODE tags.