Create a Post
cancel
Showing results for 
Search instead for 
Did you mean: 
Dmitry_Barantse
Participant

ClusterXL OSPF

Hi Team.

I have a question about OSPF work in HA pair (R80.10).

Do I need configure graceful restart or other special settings for seamless failover?

I tested failover scenario and after I disable current active member with clusterxl admin down new active member stopped forwarding traffic because of lack of routing information

3 Replies
PhoneBoy
Admin
Admin

As far as I know, you do not need Graceful Restart unless you are using VRRP.

With regular ClusterXL, routing information should sync between cluster members.

The router-id should be configured as the Cluster IP, though.

See also: OSPF on Gaia 

(1)
vinodsh
Explorer

If Router-id is not configured as Cluster IP then LSA on peer device are not updated. I could see router LSA with unreachable metric as soon as I changed the router-id to Cluster IP, it is working fine. 

0 Kudos
_Val_
Admin
Admin

It looks like OSPF synchronization is not working for you. 

It is done on TCP port 2010. You can start troubleshooting with this SK: How to troubleshoot failovers in ClusterXL - Advanced Guide 

Some additional recommendations concerning FW rules you may need to put in place are here: ClusterXL member is "Down" due to Critical device "FIB" Mind this SK is not directly applicable to your case, but the FW rules mentioned there are.

0 Kudos