FreeNAS Mini on 11.2 freezes / locks up

[hoba]

I have upgraded to 11.2U3. There is one line that catches my attention in the changelog: https://www.ixsystems.com/blog/library/freenas-11-2-u3/

Bug Separate ARC reclamation threads to prevent blocking disk I/O under heavy use

I'll give it another try with U3 tonight and report back.

 

[hoba]

Actually the system survived a complete backuprun last night. Maybe U3 really did the trick. Will watch it closely the next few days and report back again.

 

[Meyers]

Awesome. Mine just rebooted itself again the other night. I've also updated to U3 so we'll see.

 

[hoba]

3 days in a row without locking up (and it did lockup every single day for me previously with U2). I really think U3 fixed it.

 

[hoba]

Still no more issues since the upgrade to U3. For me it's fixed.

 

[wdanen]

Yes, U3 seems to have fixed my lockups as well. On U2.1 I would lockup within 1-2 days run time, max. Been running U3 for over a week now with normal usage and no lockup. Still puzzled as to what might have been the cause, but in any case, seems like it's fixed.

 

[Meyers]

Mine just rebooted itself again. Looks like it may have frozen up first then eventually reset itself. No logs anywhere I can find to explain this.

 

[Meyers]

So they closed the bug report because they couldn't find anything. This is still happening. It's seemingly random. I can go weeks w/o a reboot then I'll have two in a week. Thankfully it's not freezing up now; it's just rebooting itself.

This is what I've done so far:

-Checked every log file I can find. Nothing reported at all anywhere.

-Checked and rechecked CPU and disk temps. All well within acceptable ranges.

-Removed ZIL. No change. Added it back.

-Upgraded to 11.2-U5.

-Increased the fan speeds too 100% full just in case it MIGHT be temps (even though there is zero evidence that temps are a problem).

-Ran stress for days see if CPU temps were an issue. Could NOT get the thing to reset itself even after 100% CPU for days.

-Checked the BMC event log. Nothing.

-BMC and BIOS are at latest versions.

-Probably tons of other things I can't recall.

This system ran fine for years so this is all very puzzling to me. I really only started having problems after upgrading to 11.2. What I don't get is why there isn't a log entry anywhere indicating why this thing is rebooting. I've never seen anything like this before.

I'd really like to get the bottom of this so if anyone has any input I'd really appreciate it!

 

[hoba]

My System has an uptime of 96 days and runs fine since U3. I'm still at that version though and I'm not using encryption.

 

[vindi-799]

I'm having a similar issue, running 2 mirrored 1tb samsung qvo ssds. I have an nfs share with some VMs on it and also an iSCSCI share out to Xenserver. If I copy a VM over 100gb to the pool it crashes every single time. I never had this problem in 11.1.

Any sustained writes will cause the pool to lock up.

Running 11.2 U5. Dell pe510 64gb RAM.

 

[Meyers]

I'm having a similar issue

You'll probably want to open a ticket up and upload a debug log.

 

[vindi-799]

You'll probably want to open a ticket up and upload a debug log.

I figured it out. I was running QLC drives without knowing the cache was only 42gb. Everything was fine until I hit that limit then things would crash.