I am trying to find out how to remove two datastores marked “inaccessible” from the VCenter appliance version 6.0update2
We have two datacenters, each with it’s own VCenter appliance. We had a host in one VCenter that had access to two datastores attached over fiber channel. That host was imported into the second VCenter using the "add host", so it wasn’t properly removed from the first VCenter.
The result is that there are two datastores in the first vcenter that have no hosts associated with them. They’re showing as “inaccessible” and are grayed out. Most the functions such as “delete” “unmount”, etc. are also grayed out.
Is there any easy, safe way to remove them? I did some searching and all I seemed to find were references to NFS datastores, and/or modifying the database on a Windows based VCenter. I found articles about removing a datastore from a host, but in this case there are no hosts attached to this datastore.
I'm looking for something like the "remove from inventory" option that exists for hosts
Hi, could you try to:
1. Connect directly to ESXi using vSphere Client and try to remove the datastore from ESXi
2. Connect directly to ESXi using vSphere Client, try to rename the datastore then remove the datastore from ESXi
There are still running virtual machines on the old vCenter with that may have reference (like mounted ISO, even on a old snapshot) to the datastore that was moved to the new vCenter?
The problem is that there is no ESXI host on that VCenter that is connected to that datastore. The ESXI host was moved to a different VCenter. VCenter shows the datastore, in it's list of datastores, but there are no VM's, or hosts associated with it.
I figured it out. I was double checking on VM's on those datastores. Under "related objects", there were no VM's or hosts, but there were two old templates that were still referenced by the original VCenter. When I right clicked on the template and selected "remove from inventory", the data stores disappeared.
Also take a look at VMs that were installed from an ISO on one host, then cloned to another host. If the ISO "CD" is not unmounted (eg: CD/DVD drive isn't say set to "Client Device"), then it will hold onto the storage. This has been my problem more than once.
You may have to walk all of your VMs across your hosts to find the offending VM that still has the ISO remote mounted.
Guys, but I'll tell you what I did to solve it:
1st - Place the host in maintenance so that all the Vm's are removed from the Cluster;
2nd - Remove the host from the Cluster:
Click on connection then on disconnect;
3rd click on remove from inventory;
4th Access the isolated esxi host and try to remove the datastore with problem. Often you don't even need to remove it. If the datastore is inaccessible it is automatically removed.
After that, just insert the host in the vcenter again and assign it to the Cluster.
This worked for me I had a similar problem.
I had to deal with this today. Old datastore from ESXi host that had been ungracefully removed was sticking around after the ESXi host was removed from inventory.
Ended up being two running VM's had snapshots that were from that old ESXi host. Also, one of the VM's had a CD path to the old datastore.
Step1: Validate if any ISO / Template resides in that Datastore
Step2: Validate if the datastore is used by any host as Swap Datastore
Step3: Validate if the datastore is used as HA Heartbeat Datastore
Step4: Login to all the hosts via IP or FQDN and see if any host has assess to that datastore. If yes, remove them and it should refect in vCenter
Step5: Restart the vpxd service on vcenter "service-control --stop vpxd && service-control --start vpxd" this will refresh the inventory and do a complete sync