Hello I've been facing this issue since my last unraid update (==> 6.12.4), every 6 hours or so, my unraid server will become unreachable with no other option that hard reboot to get it back online.
I've looked into the syslog, I see errors related to the network bounding, but my networking understanding can't help me there.
Does anyone have faced the same issues and can someone have a look at the attached diag just in case I'm missing something ?

(My interface set up is : 2 mobo NIC, an Asus 10 Gbs NIC, all bounded with the 10gbs as main interface and one of the mobos as active backup).

Thanks a lot in advance for your help !
 

millesime-diagnostics-20231112-1422.zip

Enable the syslog server and post that after a crash, the network config issue is probably unrelated, but should still be corrected.

 

It happened again today, I had a look at the logs, but still can't tell what's causing this. The faulty sequence seems to have happened around NOV 14 07:26:45 in the logs.
Would appreciate any feedback.
For now my temporary fix is a user script that properly reboot the server every 4 hours (As I'm away from home quite often and can't afford server being down until I can manually reboot).

millesime-syslog-20231114-1528.zip

That's not the persistent syslog, but there's a clue anyway:

 

Nov 14 07:27:38 millesime kernel: macvlan_broadcast+0x10a/0x150 [macvlan]
Nov 14 07:27:38 millesime kernel: ? _raw_spin_unlock+0x14/0x29
Nov 14 07:27:38 millesime kernel: macvlan_process_broadcast+0xbc/0x12f [macvlan]

 

Macvlan call traces will usually end up crashing the server, switching to ipvlan should fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)), then reboot.
 

On 11/14/2023 at 5:27 PM, JorgeB said:

That's not the persistent syslog, but there's a clue anyway:

 

Nov 14 07:27:38 millesime kernel: macvlan_broadcast+0x10a/0x150 [macvlan]
Nov 14 07:27:38 millesime kernel: ? _raw_spin_unlock+0x14/0x29
Nov 14 07:27:38 millesime kernel: macvlan_process_broadcast+0xbc/0x12f [macvlan]

 

Macvlan call traces will usually end up crashing the server, switching to ipvlan should fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)), then reboot.
 

Thanks a lot, I've tried this switching to ipvlan and it seems to be working consistently now.

As for the persistant syslog, I was finally able to get it to work (forgot to input the server address as remote server).