Random server freezing/lagging issues

[Crimson Unraider]

Hey guys, I could use a little help.  I have been getting a lot of random issues in the last week, High CPU usage, all threads stuck 100%, Slow access to files, Plex locking up, other docker randomly freezing and I get Server Error (no code) when I try to stop or restart the docker.  When I reboot the server using the reboot button it takes more than 20 min and when it comes back up it starts a Parity check due to unclean shut down. I stopped docker and ran mover, but mover is really slow also.  The only warnings I get from fix common problems is my Network adapters are configured with a non-standard MTU of 9014. I have a 10Gbs nic installed and onboard nic disabled but it gives the warning for both eth0 and eth1.  Only eth0 shows on network settings.  I tried downgrading to 6.10.2 but it didn't help. VM seems to be fine. My diagnostics are attached but it is after restart due to frozen GUI.

crimson-diagnostics-20220917-0138.zip

[JorgeB]

We highly encourage to use MTU 1500 for the main NIC, eth0, you can then use 9000 for the 10GbE NICs, I for example always have a gigabit NIC with MTU 1500 as eth0 and then my 10GbE LAN with MTU 9000, if you want to keep using that make sure everything in the LAN supports that and is correctly configured.

 

You should upgrade back to v6.10.3, delete/rename /boot/network.cfg and reboot to see if both NICs get detected.

 

Also switch to ipvlan (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)), I see some call traces in the log related to that.

 

After that post new diags after you see the problem.

[Crimson Unraider]

Ok I'll try it and let you know. Thanks for the quick response.  It won't detect both lans because I disabled the 1G nic in the Bios.  Can I configure with the 1G and the 10G on the same network? My switch is an unmanaged switch with 2 10G and 8 1G ports.

Edited September 17, 2022 by Crimson Unraider

[Crimson Unraider]

Jorge,  I changed back to 1G nic for awile and I still had the same issue.  Then I swatched back to the 10Gbe nic and changed the Docker custom network type -> ipvlan and it has been running pretty good for the last couple days.  I will let it run like this for a few more days but I think that was the problem.  Also, I just upgraded to 6.11.0 so that may have also been the fix.  Thanks for the help, I'll let you know.