ISDN dialer watch using the dialer map statement fails to shut down when using host networks as the watched networks on my 4000's and 4500's. When the ISDN connection dials on the first line failure the watched address becomes a connected network (C) in the routing table (to BRI0) with an A.D. = 0. As a result, when the regular connection is reasserted, it will not pick up the watched route because its AD will be greater than 0, and the dialer watch will continue to determine that the primary route is down. As a result the ISDN line does not drop out. This is a problem when monitoring a loopback interface that is defined as a host network (/32). It is circumvented by making the loopback address space a four-address space (/30). The mapped address in the dialer map statement is still identified as a host route connected to BRI0 (/32), but when the primary line comes back up the dialer watch will identify the remote route (at /30) as correctly assigned to the primary line and the ISDN call will be dropped.
I have not seen this mentioned in any of the references, and I customarily use host network loopback addresses. Therefore, when I try to use the best practice of watching loopbacks, the line doesn't behave. Replaced IOS's, reset a lot of lines before it occurred to me. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62722&t=62722 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
