Panic solaris assert: c < (1ull <<24) >> 9, rebooting freenas system

[tomcatcw]

Hello Community,

please be gracious, but as the title says I'm fully of panic myself.

I searched the forum for a similar failure as I have, but could not find appropriate threads. In case this thread is wrong please move it to the right place.

Here is my situation:
Quietly unexpected my freenas box started to reboot. This happened a lot of time, but always came up normal, so that I had access to the webui. I had a certain dataset in suspicion, because whenever I accessed a certain directory via ssh within this dataset, the server restarted. I then decided to start a scrub on that pool. This runs to approx. 93% then the server restarted again, but could import the pool normally. I decided to destroy this particular dataset in the hope of fixing the error. But I was wrong. From now on the servers restarts all the time an leads to the error message see screenshot. Even the db prompt is not seen. Always reboots as soon the error message comes up.

Currently I pressed ctrl+c so that this faulty pool ist not imported. The server starts normally to the cli. I imported pool then with read only. The server runs stable. So I am about to backup the data.

Can you give me advice what to do next. The data is saved so that the pool can be destroyed and started from scratch. But as soon as I import the pool without read only the server crashes and reboots.

What else can I do to destroy this pool and create a new one with die existing hdds. Or maybe the corrupted data can be repaired. I don’t think those disks are damaged. In my opinion this is only caused by corrupted data.

Some information about my system:

RAM: 8 GB non-ecc
HDD: 4 drives WD-Red 3TB, Raid-Z1
FreeNAS: 11.3-U3.1
motherboard an cpu: low budget itx-board (add later on, currently in office)

If you need more information, please tell me.

Thank you in advance.

Greeting
Christian

 

[Tony-1971]

Hi,

8GB on non ECC ram is not recommended: can you perform a memory test?
There is also a new version of FreeNAS-11.3-U3.2 , you can try to upgrade.

Best Regards,
Antonio

 

[tomcatcw]

Hi Antonio,

I upgraded to 11.3-U3.2 already which does not helped and leads to the same failure. I then changed the boot environment to U3.1 hoping that this helps. But it doesnt. How can I do a memory test? The FreeNAS Boot Menu does not contain such tests or havent I found that?

Thanx.

Christian

 

[Tony-1971]

Hi Christian,

There is no memory test in FreeNAS menu.
You can use something like https://www.memtest86.com/

Best Regards,
Antonio

 

[jlpellet]

If I understand, you have a data copy so can start fresh. From your description, I think the pool is corrupt. Have you tried importing read-only then export/destroy the pool? If that works, you can create a new pool & see if it recurs. You can alos pull the drives to wipe in a different system, if needed. Assuming you have a config copy, if this does not work I've start a fresh install. Good luck.
John

 

[tomcatcw]

Hi,

@Tony-1971 :
I will give it try to verify that the ram is ok.

@jlpellet :
Exactly: I have a data copy and could generally start from scratch. But there was some hope that the corrupted pool could be saved.
Is that possible to export a pool an then destroy it?