Migration to new nodes fails: "VLAN aware" bug?

NdK73

Renowned Member
Jul 19, 2012
97
6
73
Bologna, Italy
www.csshl.net
Hello all.
At the end of 2021 I upgraded all the six existing nodes to 7.1 (migrating VMs back and forth w/ no issues, except higher load).
While at it, I also installed 3 new nodes and added 'em to the cluster.
The cluster now sees the 9 nodes online and OK.
But when I try to live migrate a VM from one of the older nodes to a new one I get:

Code:
2022-01-10 09:29:54 starting migration of VM 142 to node 'virt7' (192.168.1.37)
2022-01-10 09:29:54 starting VM 142 on remote node 'virt7'
2022-01-10 09:29:58 [virt7] ovs-vsctl: no bridge named vmbr1
2022-01-10 09:29:58 [virt7] can't add ovs port 'fwln142o0' - command '/usr/bin/ovs-vsctl -- add-port vmbr1 fwln142o0 -- set Interface fwln142o0 'type=internal'' failed: exit code 1
2022-01-10 09:29:58 [virt7]
2022-01-10 09:29:58 [virt7] kvm: -netdev type=tap,id=net0,ifname=tap142i0,script=/var/lib/qemu-server/pve-bridge,downscript=/var/lib/qemu-server/pve-bridgedown,vhost=on: network script /var/lib/qemu-server/pve-bridge failed with status 256
2022-01-10 09:29:58 [virt7] start failed: QEMU exited with code 1
2022-01-10 09:29:58 ERROR: online migrate failure - remote command failed with exit code 255
2022-01-10 09:29:58 aborting phase 2 - cleanup resources
2022-01-10 09:29:58 migrate_cancel
2022-01-10 09:30:00 ERROR: migration finished with problems (duration 00:00:07)
TASK ERROR: migration problems

Obviously I checked the network config to be sure vmbr1 exists also on the target node and maps to the same VLAN. The strangest thing is that it's a Linux bridge and we never used OVS for anything.

BUT if I remove these two spurious lines:
Code:
root@virt7:/etc/network# diff -u interfaces{bad,}
--- interfacesbad       2022-01-10 11:10:45.975357535 +0100
+++ interfaces  2022-01-10 11:11:13.987952707 +0100
@@ -30,8 +30,6 @@
        bridge-ports enp3s0
        bridge-stp off
        bridge-fd 0
-       vlan-id 1
-       vlan-raw-device enp3s0
 #Public net
 
 auto vmbr2
the error goes away.
Maybe it's a bug when removing the "VLAN aware" flag? We experimented a bit with it. Removing those lines from the GUI implies removing an recreating the affected interfaces from scratch, removing "vlan aware" flag is not enough.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!