>>But we should only accept packages which originates from VMs? 

I see 3 cases:

ethX->tap-in : 
--------------
incoming ethX is not firewall
tap-in do the ACCEPT

tap out->tap in :
----------------
tap-out do the RETURN
tap-in do the ACCEPT

tap out->ethX :
---------------
tap-out do the RETURN,
so we need an accept for ethX



I have look at cloudstack, they are doing

-A vmbr0 -m physdev --physdev-is-bridged --physdev-out $physdev -j ACCEPT

where $physdev is ethX,bondX plugged in the bridge


but maybe

-A vmbr0 -m physdev --physdev-is-out --physdev-is-bridged -j ACCEPT

is enough ?



----- Mail original ----- 

De: "Dietmar Maurer" <[email protected]> 
À: "Alexandre Derumier" <[email protected]>, [email protected] 
Envoyé: Mardi 25 Février 2014 16:25:00 
Objet: RE: [pve-devel] [PATCH 2/2] bridge rules : -j ACCEPT for physical 
interfaces 

> We need to accept traffic at the end of bridge rules for outgoing packets 
> from tap->ethX, as we don't do ACCEPT in tap-out rules. 

But we should only accept packages which originates from VMs? 
_______________________________________________
pve-devel mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Reply via email to