Failed Primary Route Periodically Reappears in the
Routing Table
A primary connection fails; network monitoring detects the failure and
removes the primary route. A backup connection becomes active, and traffic
once again reaches its destination. A minute or two later, the router begins
dropping traffic again. Then after several minutes, traffic can again reach its
destination. Users complain that their connections keep going up and down.
The problem is that PBR has not been configured for probe traffic, so the probe
begins testing the backup route rather than the primary route. The track, on
the other hand, is still monitoring the primary route. Because the backup route
is good, the track believes incorrectly that the primary route is valid and
should be added to the routing table.
To fix this problem, make sure that you have correctly configured a PBR route
map for the probe. See "Implementing PBR to Route Probe Traffic" on page
9-34. Check that:
The route map has been applied to router traffic as the local policy.
A route map entry matches the correct ACL configured for the probe.
The destination address in the ACL matches the probe's destination. For
TCP and HTTP probes, the destination port must also match.
Network Monitoring
Troubleshooting Network Monitoring
9-61