Ok that sounds better. So would it be best to use the same eth0 (or whatever the physical interface name is) as the untagged network management device, or should I somehow create a virtual interface via OVS?
Also, if I: 1. Start the database server with an empty database 2. Start ovs-vswitchd 3. Configure OVS to use eth0 4. Start using eth0 normally 5. Stop the database server 6. Swap out the database to the persistent one 7. Start the database server What happens after #7? Is there something I have to do to get the previous network state restored, or does ovs-vswitchd just magically notice that the actual running state vs. saved db state is different, and correct the actual running state to match? Also, when the database server is stopped, will the network on eth0 be disrupted at all? On Tue, May 14, 2013 at 8:17 AM, Jesse Gross <[email protected]> wrote: > On Mon, May 13, 2013 at 9:24 PM, Nick <[email protected]> wrote: > > 5. Start ovsdb-server on one node (on a floating IP which would be an > alias > > on eth0.mgmt), with a persistent storage data directory. If this server > > dies it will be automatically master elected (using our internal > automation > > framework) on another node and the floating IP moved to the new node. > > 6. Start ovs-vswitchd on all nodes, pointing them to the single > > ovsdb-server IP. > > This part isn't right: you can't have a single database server for the > entire network, you need one per host. > > By default, when OVS comes up it will allow untagged traffic to pass. > This should allow you to bootstrap and then you can have it connect > outwards to retrieve its configuration. >
_______________________________________________ discuss mailing list [email protected] http://openvswitch.org/mailman/listinfo/discuss
