bridge problems between br100 and eth0, stops working
If I reboot my compute node, the bridged network works for my guests but after 4-5 minutes it stops working, if I run tcpdump on eth0 (public) i can se icmp requests coming in but no replies, if I run tcpdump on br100 I can see icmp requests coming in and replies being sent but the traffic does not seem to makes it to eth0, any ideas?
If i do the ping from the same local subnet it works all the time, but it's traffic outside the subnet that stoppes working.