When configuring VDS LACP, the active uplink uplink1 is set to unavailable by mistake, resulting in

[Aizenwong]

Hello everyone
When setting the VDS LACP, I add the Lag1 to the active uplink when configuring the binding and failover of the distributed port group. At this time, I only configure the LACP of one esxi host, and none of the other hosts are configured. This operation makes all esxi unable to connect to vsan storage, including VCSA (VCSA virtual machine storage is vsan); You can't recover by using the esxcli command line. Please tell us what other methods you can use to recover.

[TheBobkin]

@Aizenwong "You can't recover by using the esxcli command line." Yes you can, Did you partition the cluster or just vCenter lost it's port? If the latter then follow https://kb.vmware.com/s/article/83906, if the former then take the uplinks off the vDS used for vSAN, make a new vSS, add the uplinks to it, attach the vsan-vmk to it (or make a new one and enable vsan-traffic on it). If you are changing IPs or anything like that then you will have to manually change the unicastagent entries on each node https://kb.vmware.com/s/article/2150303.

Stand your vCenter back up and then consider when making changes in future with vCenter stored on the vsanDatastore of the cluster it is managing to either be more careful or move the vCenter off vsanDatastore while making changes that can possibly partition your cluster.