January 28, 2000
SC2200 version 4.2(x)
The information in this field notice is for sites implementing Redundant Signalling Link functionality on an SC2200 system.
In order to implement fully redundant signalling paths between the SC2200 and the network access servers (NASs) it controls, it is necessary to be able to route packets to the NASs via a primary and a backup Ethernet interface; the SC2200 requires routing to two different IP addresses corresponding to each NAS. Due to limitations of the Solaris operating system used by the SC2200 and certain routing protocols, it is not possible to automatically distribute routes to both of the NAS addresses if they are part of the same classful (class A, B, or C) network.
The SC2200 and NAS use a protocol known as Redundant Link Manager (RLM) to provide a robust communication path between the SC2200 and associated NASs. RLM uses separate interfaces and IP addresses on each end of the link to provide a rapid failover to a second signalling path in the event of a failure of the primary path.
In many cases, network design considerations require the primary and secondary IP addresses of the NASs to be on different subnets of the same classful network. The Solaris 2.5 operating system used in SC2200 version 4.2.x has a very limited ability to perform subnet routing of outgoing packets. Furthermore, the RIP version 1 protocol supported by the standard "routed" routing daemon does not support propagation of other than classful network routes.
Following failure of the primary link in a redundant link SC2200-to-NAS connection, the secondary signalling link does not provide signalling continuity as expected. This normally results in dropping all calls on the affected NASs.
If the primary and secondary NAS IP addresses are in the same classful network, it is normally necessary to configure a host route to either the primary or secondary NAS IP address. You can usually provide a default route or other network route to send packets to the primary NAS address(es), but an individual route to each secondary NAS address is required. This is configured by the super-user or in one of the Solaris initialization files with a number of commands such as:
route add host /NAS1secondary/ /next-hop-router/ 1 route add host /NAS2secondary/ /next-hop-router/ 1 route add host /NAS3secondary/ /next-hop-router/ 1
In the syntax above "/NAS?secondary/" represents the IP address of the secondary interface IP address of a given NAS. The "/next-hop-router/" syntax represents the address of the router connected to the secondary Ethernet interface of the SC2200 which will route packets to the NAS.
In order to verify the solution, it may be useful to verify connectivity and routing to each of the interfaces via the Solaris traceroute command. The routes to the primary and secondary IP addresses of each NAS should be compared to ensure that they have no IP addresses in common. If common addresses are found for a given NAS, this indicates that the routing paths are not fully independent and that a source for a single point failure may exist.
In cases where it is possible to use a dynamic routing protocol, it is also useful to verify the paths to the primary and secondary NAS IP addresses to ensure that the routing protocol has not introduced unexpected route dependencies.
For more information, refer to the Solaris operating system documentation for the route and traceroute commands.
For More Information
If you require further assistance, or if you have any further questions regarding this field notice, please contact the Cisco Systems Technical Assistance Center (TAC) by one of the following methods:
Receive Email Notification For New Field Notices
Product Alert Tool - Set up a profile to receive email updates about reliability, safety, network security, and end-of-sale issues for the Cisco products you specify.