Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. If you'd prefer to explore, try our test area to get started. And see here for current known issues.

New Member

HSRP / Cat 3524 / Stonebeat Full Cluster - Problem


i have th foloowing problem:

we have 2 72xx Router with are running hsrp and both are connected to a catalyst 3524 Switch. To this switch there is also a checkpoint firewall-1 connected: the fireall is a cluster solution running stonebeat full cluster. For this cluster there is a virtual ip and mac address. The mac address is out of the multicast range 01:00:5e:xx:xx:xx. There is the foloowing problem.:

If i try to telnet the switch i have to try several times until i get a connection to this. Both routers are also flapping. On the switch i have configured static mac addresses for the outgoing ports to the firewall. The switch is sending tcp rst packets. I know the multicast problems with the cluster software and catalyst switches. There are other switches wich are working fine with static mac entries and the firewall cluster. But with this i don´t have any idea. Does anybody have nearly the same problems or uses stonebeat cluster software too? thanks al lot....

  • Other Network Infrastructure Subjects
New Member

Re: HSRP / Cat 3524 / Stonebeat Full Cluster - Problem


I had the similar problems at a customers place with stone beat software and multicast address as stone beat proposes.

The 2948 work well and 6500 with Cat OS 5.x work also newer releases like 6.3.5 no more

The new 2950 and 3550 wont work.



New Member

Re: HSRP / Cat 3524 / Stonebeat Full Cluster - Problem

In my experience SBFC + multicast addresses does not work at all if you use HSRP. Here is the solution we use:

-HSRP on your routers, nothing special.

-Use unicast addresses on the SBFC side. The downside to this is that you will only be able to reach one firewall at a time (the active one) on this NIC. The way we overcame this is to use reachable addresses (a /30) on the hearbeat network. This network should be statically routed on your routers to the virtual IP of the SBFC cluster.

-Static mac flooding config on the switches. Something like:

mac-address-table dead.beef.face fast 0/0 fast 0/3 fast 0/4

mac-address-table dead.beef.face fast 0/1 fast 0/3 fast 0/4

This is assuming that dead.beef.face is the unicast mac of the SBFC, your 7200's are on fast 0/0-1 and SBFC is on fast 0/3-4. This will flood the traffic from your 7200s to both of the sbfc ports simultaneously. Also keep in mind your other VLANs with SBFC interfaces in them will need similar config, with all client ports listed as the input port.

In my opinion this is a really nasty config, and I cringe doing it but this is pretty much from Stonesoft themselves.