Stuck routes that require clear

Dieg0M

Hey guys,
I have encountered in several occasions on production networks that a specific route after a failover or failback would not route the proper way or would not be redistributed properly. I would need to do a clear ip route x.x.x.x or a hard reset of the neighbor adjacency to fix the issue. No configuration changes need to be done to fix the problem. Most of these problems involve routers and switches that do redistribution between BGP and an IGP (most often with EIGRP). Also, I have noticed that most of these cases involve devices that have an uptime of several years. Have any of you encountered similar problems? What do you think this might be related to?

My colleagues and I have different theories on this and I would like to hear all your advice on this.

Thank you

networker050184

Probably something in the code and not something you could configure around or a standard way of the protocols reacting. Either that or the CPU get's so taxed it just misses cycles.

keenon

BGP config
IOS version
Hardware resources

--Not a fan of eigrp

Dieg0M

Different BGP configs and different IOS's. Some are 6500, some 4500, some 3750's. Issue seems random and not recurrent.