3 weeks ago
we have a weird problem with a 2 node fabric attached stretch metrocluster and a DR site.
Our Ontap 9 metrocluster is peered with a disaster recovery (DR) cluster.
Checking metrocluster LIFs health in a 2-node stretch metrocluster returns a warning if data LIFs from 3rd peered cluster is unreachable. This could affect basic operation like switchover and switchback.
In order to separate environment we tried to:
use different vlan for the purpose of cluster peering (metrocluster and DR) in the same ipspace
use different vlan for the purpose of cluster peering (metrocluster and DR) in different ipspace at metrocluster side.
Since the DR site is not part of the metrocluster, I would expect the DR site data LIFs would not be checked when running metrocluster check run, but they show up instead. Furthermore, when communication with the DR site is interrupted, the metrocluster check command gives a warning about LIFs because DR cluster data LIFs are unreachable.
There is an internal bug opened for this problem
But I want to be sure we aren't doing any misconfiguration