I’ve been randomly having Docker, I think, crash the last few weeks requiring unclean reboots to recover from. I have started logging, as of earlier today, syslog to USB drive. Server has not went down since I began logging to USB. 
 

I have been watching server and came across this though— see attached pic and error below: 

 

Feb 27 19:44:35 Tower kernel: traps: php-fpm7[1107] general protection fault ip:1519630a2c88 sp:7ffe596f23a0 error:0 in ld-musl-x86_64.so.1[151963094000+48000]

 

Could this have anything to do with my crashes lately? I was able to find one thread in forum with similar error but seems was radarr specific. 
 

Attached is syslog as well as diagnostics. Any help would be greatly appreciated. Hopefully all required info is provided. 

 

tower-syslog-20220228-0300.zip tower-diagnostics-20220227-2102.zip

Macvlan call traces are usually the result of having dockers with a custom IP address, upgrading to v6.10 and switching to ipvlan might fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enable, top right)), or see below for more info.

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/

1 hour ago, JorgeB said:

Macvlan call traces are usually the result of having dockers with a custom IP address, upgrading to v6.10 and switching to ipvlan might fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enable, top right)), or see below for more info.

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/

Is 6.10 likely to bring other issues to the table though? 🤔🤔

Try it and see

10 minutes ago, plantsandbinary said:

Try it and see

Depending on how this week looks I may see if I can find the time. Do I just need to backup USB drive ahead of time in case things go awry? Change branch to "Next"?

32 minutes ago, blaine07 said:

Depending on how this week looks I may see if I can find the time. Do I just need to backup USB drive ahead of time in case things go awry? Change branch to "Next"?

Yes.

I had the same issue with at least weekly crashes and unclean shutdowns.  The docker network fix suggested by JorgeB was what was recommended to me and after doing it, I've not had the same issue.

 

I'm not happy about having to run a RC OS on my main data storage, but it is what it is.

 

(have had no other issues, AFAIK) - been running probably 1-2 months that way now.

Well, as advised, here I am on 6.10 RC2. Server went down again this evening...I even was writing logs to USB and nothing useful there. Like nothing. So crazy. Anyways, I got jammed up on all the MyServers stuff...Everything back up though and changed to IPVLAN

 

Cross fingers for me that this is the stability I need, please? LOL 🙂

Long shot: Is anyone aware if changing to IPVLAN would mess with containers being assigned BR0? I have issues with Remote Access with Unifi Docker; changing from BR0 to “Bridge” and issue goes away. It’s possible container is mucked up but is it change to ipvlan could’ve caused an issue unbeknownst to me?

Is there further setup required when switching to IPVLAN from MACVLAN? This Unifi Controller… seems to not allow “remote access” to work since I’ve turned IPVLAN on. I change container to anything but BR0 and remote access immediately works. Not sure if container is broke or change from Macvlan to ipvlan and not sure where to start [emoji30]

On 3/8/2022 at 6:50 AM, blaine07 said:

Is there further setup required when switching to IPVLAN from MACVLAN? This Unifi Controller… seems to not allow “remote access” to work since I’ve turned IPVLAN on. I change container to anything but BR0 and remote access immediately works. Not sure if container is broke or change from Macvlan to ipvlan and not sure where to start

FWIW: I had had static IPs set in pfSense from when I was using MACVLAN in Unraid; once I deleted the MAC address associated with previous static IP assignment(s) this issue went away. I was able to keep static ip, just couldn't have previous MAC Address from MACVLAN associated it with it. No NO Mac Address is associated with static IP assignment in pfSense.

 

tldr; figured it out; my own idiocracy.

I am experiencing similar kind of issues after switching from macvlan to ipvlan, which i have mentioned here: 

 

@blaine07 May I ask you how you have resolved those issues?

 

3 hours ago, greenflash24 said:

I am experiencing similar kind of issues after switching from macvlan to ipvlan, which i have mentioned here: 

 

@blaine07 May I ask you how you have resolved those issues?

 

Sure.

 

I used to have everything with STATIC IP in pfSense. When I converted to IPVLAN stuff was all borked. In DHCP Server>One of the listings for something on IPVLAN had to UNCHECK "Create an ARP Table Static Entry for this MAC & IP Address pair" AND remove the MAC ADDRESS for the items setup with IPVLAN. You can keep static entry; just have to uncheck THAT specific box.

 

Here is a example for something I have a static IP on:

 

Hopefully this makes some sense. 🙂