On Mon, Feb 6, 2012 at 23:22, Patrick Lists <centos-list at puzzled.xs4all.nl> wrote: > On 07-02-12 04:28, Bob Hoffman wrote: >> I put this page together just so I won't spam the board anymore begging >> for help..lol >> http://bobhoffman.com/vmissue.html > > According to http://wiki.centos.org/TipsAndTricks/BondingInterfaces > there should not be a HWADDR=<mac_address> in ifcfg-eth0. > > Regards, > Patrick I second that. What may be happening is that the VM host, which is explicitly set to use br0 as it's main host interface, works fine when bond0 communicates using the eth0 interface, and maintains the connection while eth0's MAC is active for br0 ip address, but at some point outbound traffic from the VM host ceases and the MAC address for eth0 times out of br0. At that point inbound traffic can go to whichever interface answers first in bond0, and if it is eth1, traffic will time out since the main host is probably only using eth0 and not the br0 as its own interface. I would be curious to know what the main hosts routing table has in it, if it is using eth0 or br0 as it's communication interface. try these: # route -n # ip route show table all