An expired NSX Manager certificate is replaced through the API: one certificate per node, the cluster VIP last, plan by default. The data plane is not touched.
Global Manager and Local Manager both show Aria Automation the same segment. Network profiles fill with duplicates, and IPAM does not always agree with either of them. Analyze, then plan, then execute.
VCF host commissioning rejects a host that is already in a cluster, already in a vCenter, or still in maintenance mode. The prep script plans by default and leaves those hosts alone.
A blue-to-green Avi move converges every virtual service onto the new site’s own cloud and SE group, then waits. DNS names flip one at a time, and a flipped application is frozen so the next run cannot drag it back.
A Global Manager to Global Manager firewall move should stage every policy disabled. API acceptance is not proof the rule is realized, and the tool refuses to recreate network objects it does not own.
Two VCF sites, two Avi controllers, one GSLB service. The build is a sequence of gates. The production names do not move until failover and failback have already been proved against the DNS virtual services.
In VCF 9.1, regional Harbor is the registry Supervisor services and VKS add-ons pull from. A region teardown can leave it mid-removal. Retrying the uninstall does not bring the cluster back.
A VCF on VxRail estate that has gone quiet does not need a credential rotation as the first move. It needs a fixed order of health checks, and a depot check that now applies even when the cluster is online.
After the management domain reached VCF 9.1.1, the workload domains showed every component on target and greyed out the plan buttons. The domain was not upgraded. A customized BOM had pinned it.
Three ways a kit still arrives at a disconnected site, and the VCF 9.1.1 depot path those older inventories should not be merged with.
