ClusterXL probing wrong IP?

[Ryan_Ryan]

We have two virtual gateways in cluster running on an on premise vmware. Everything as working perfectly until one of the hosts was rebooted.

The standby member keeps flapping between status of standby and down. The reason - because it says the outbound direction of one of its interfaces is down every 30 seconds or so then comes back up again (inbound stays up). It always happens on the same interface and not any of the others.

although I suspect the issue is not the firewall itself I have noticed one thing, the ccp packets on that interface are being sent (unicast) to the gateway address of the upstream router. When I look at all the other interfaces, the ccp packets are directed at its own physical interface and its partner physical interface. 

I have done a get topology again, confirmed the vip and mask etc is ok, pushed policy again, (presumably if any of that was wrong it would not work at all rather than intermittent). also this only happens on one of the firewalls not the primary

any ideas?!

[Chris_Atkinson · Accepted solution]

Hi Ryan, for context can you share the Gateway/Cluster version please?

Requirements for clustering on VMware have changed for the better in recent versions (R80.40+) compared with past releases that required ESXi side tweaks (sk101214)

[Ryan_Ryan]

Hello,

thanks for the document because we resolved it, somehow mac address learning was disabled on one side of the port groups. (not sure how an incident caused that but happy its fixed - thanks)

[Chris_Atkinson]

Happy it's solved, please also consider your upgrade plans prior to September as R80.20 will be End of Support then.