I have a site with a RB4011 connected via an old M5 Powerbridge, and an
EPMP PTP425. The 4011 is running ROSv7, everything else is on ROSv6.
Connecting to a CCR1036 and CCR1009 on the other ends of the radios.
Everything was working perfectly for several months until a storm came
through, and now the F425 OSPF Session will not stay stable. The OSPF
Weights are set to prefer the F425 link. The OSPF neighbor never goes
down, Adjacency/state never change, but the RB4011 will update it's
entire routing table about every 5 seconds and flap back and forth
between the Powerbridge and the F425. The remote routers never update
the route to the 4011, it is stable over the F425. If I disable the
F425 neighbor, everything is stable over the powerbridge. If I disable
the Powerbridge neighbor, the router will lose connectivity for a second
every 10 or 15 seconds, like it loses it's routing table.
There is no packet loss across the F425 link, and it will Tik to Tik bw
test at 100MB sync no problem and no packet loss. I've even tried
changing ports on the 4011, plugging the F425 into a completely separate
OSPF router on the remote end. I've also put the traffic through a VLAN
on the F425 to see if the F425 was doing something weird with the
packets. The F425 radios on both ends have been power cycled. I've
also updated ROS to the latest 7beta, and no change.
I'd think this is an issue with ROSv7/6 since the routing table keeps
updating, but the powerbridge link is always fine, so it feels like
something with the F425. Especially since it happened after a storm.
But at the same time, how could a failure of the F425 be affecting
OSPF? When I do OSPF Logging on the 4011, nothing is looking out of the
ordinary. I guess I could try physically changing the 4011, but I've
never had a failure like that before.
Mainly I just want to vent that I'm probably missing something stupid, I
just haven't been able to figure out what it is yet.
--
AF mailing list
AF@af.afmug.com
http://af.afmug.com/mailman/listinfo/af_af.afmug.com