Hello, We are running subsecond OSPF in our enterprise LAN and just upgraded our core switches. When one core was being brought up, we lost 30 seconds of pings to our server farm (which is hanging of the core switches redundantly in L3). The setup is a bit more complicated, but analysing the boot, i noticed that all OSPF links came up rather "randomly" and i think the downtime came from blackholing traffic because neighborships were still missing. For example: the server farm OSPF neighborship gets up first. Since this is a stubby area, the only route added is a default route. The server farm starts to loadbalance to the two cores, but on one of the core switches the downstream links are not yet up, resulting in blackholing traffic. So my question is: can i put a WAIT timer on the routing process so that it waits until all linecards are booted and then initiates the routing process (preferably in a certain predetermined sequence) ? Or on the server farm core, can i put a HOLD timer on the ospf neighborship with the core ? ( a bit like you can configure a PRE-EMPT delay on HSRP neighbors ?)
regards, Geert _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
