[PVE-User] Strange problem on bridge after upgrade to proxmox 7

[Simone Piccardi]

On 19/05/22 20:58, Richard Lucassen wrote:
> I have no idea if this can have something to do with it, but not a very
> long time ago I had two Dell R210 servers connected through a simple
> failover bond0. The issue I found was that somehow these bond0 devices
> on two *different* servers got the *same* fixed MAC address. After some
> searching I stumbled upon this:
> 
> https://blog.sigterm.se/posts/a-bonding-exercise/
> 
That's a very interesting reading, thank's for the link.

Anyway I checked all bridge mac address on all server, and they are all 
different (they were anyway installed indipendently and have different 
machine-id).


> I had some discussion afterward with Patrik and I ended up in adding a
> fixed MAC address in the /etc/network/interfaces stanza, e.g.:
> 
> hwaddress ether 4a:89:66:60:e4:97
> 

I think this can solve MAC conflicts (I'll try anyway) but I'm not 
seeing any duplicate MAC, and I got problem only between the VM inside a 
bridge and this specific host. The host is perfectly reachable from 
everywere.

What I got is the same problem described here (I found this just this 
morning):

https://forum.proxmox.com/threads/ve-7-1-10-slow-to-forward-arp-replies-over-bridge.106429/

a VM has a port inside the bridge that does not match the one used by 
its tap interface,

I'll investigate the issue explained in this article referenced there:

https://bugs.launchpad.net/neutron/+bug/1738659


Simone
-- 
Simone Piccardi                                 Truelite Srl
piccardi at truelite.it (email/jabber)             Via Monferrato, 6
Tel. +39-347-1032433                            50142 Firenze
http://www.truelite.it                          Tel. +39-055-7879597