6.9.2 - Kernel Panic

[xRadeon]

Hello All,

 

I just had my first Unraid Kernel panic. I'm not too familiar with how to decode the panic output, so I'm unsure what caused the panic. Below is a screen shot of as much of the panic info as I can.

 

It crashed around ~0120 for me (I can tell from my Observium data stopping around that time). Not sure if any tasks ran at that time, however, I was finishing up a full BTRFS balance operation (after adding a new disk) to my 'hdd' BTRFS pool. There's a possibility it completed right around the time where the system crashed, so it's potentially a BTRFS issue.

 

I've ran a full memtest and it completed with no errors, I only did one full test though. I've also attached the diagnostics zip if needed.

 

Thanks for anyone who takes the time to look at this!

 

 

island-diagnostics-20210425-1003.zip

[JorgeB]

Can't see the complete call trace, but looks more like it's related to this, see if it applies to you:

 

https://forums.unraid.net/topic/70529-650-call-traces-when-assigning-ip-address-to-docker-containers/

 

See also here:

https://forums.unraid.net/bug-reports/stable-releases/690691-kernel-panic-due-to-netfilter-nf_nat_setup_info-docker-static-ip-macvlan-r1356/

 

If it keeps crashing you can enable the syslog server to see if it catches the traces.

[xRadeon]

Hmm. I'm not sure that it's the br0/nat issue. Most my dockers are bridge or host, none are br0 w/static.

 

It's only panic'd one time so perhaps it's not that big of a deal. If it happens again I'd start to be concerned.

 

Thank you for your reply!

[MariuszM]

It's happening for me too 🤔 on version 6.9.2 (before this one, everything was working smoothly). So after 4th Kernel Panic I sign up here. From the stack trace I can tell that something with network. @JorgeB thanks for some links I will check this.

 

[xRadeon]

I actually had a second panic the other day. I went through the link above and it seems like there isn't a solid conclusion as to what the problem is. It for sure is a networking issue related to docker, but it doesn't seem like it's 100% when you have dockers using the custom br0 net. Someone did post on that thread talking about the "Host access to custom networks" option causing the problem. I've disabled that and I'll see if I get another panic (I've been up for 2 1/2 days now).

 

[MariuszM]

I decided to turn off "Unifi-controller" docker app, because this was my new configuration. I think this was it, almost one week without kernel panic.

[Lilarcor]

On 5/3/2021 at 8:21 AM, xRadeon said:

I actually had a second panic the other day. I went through the link above and it seems like there isn't a solid conclusion as to what the problem is. It for sure is a networking issue related to docker, but it doesn't seem like it's 100% when you have dockers using the custom br0 net. Someone did post on that thread talking about the "Host access to custom networks" option causing the problem. I've disabled that and I'll see if I get another panic (I've been up for 2 1/2 days now).

 

Does your server you still have this issue

[xRadeon]

It was up for about 8 days then I had some bad luck over the weekend as the power went out for longer than my UPS lasted for. Hasn't crashed since the power going out, so that's good. I'm thinking if it can go about 14 days with out crashing, that's probably a good sign that it's fixed, so I'll update in about a week and half.

[xRadeon]

I guess I should update this.

 

I've been working really solid for weeks now. I did one test where I enabled "Access to Custom Networks" and within hours it had locked up with the same output as above. So for me, the solution was to disable access to custom networks in the Docker settings.