VMware Cloud Community
MiKLVMware
Contributor
Contributor

vCenter - Incorrect Datastore capacity after increase

Hi,

I have for some time a problem with vCenter and about datastores resizing.

Symptoms

A have a datastore with the orignal size : 1To.

I increase VV on the SAN Storage : +1To

On vCenter, I check the Device size on each ESXi : 2To

On one ESXi, I increase the datastore. OK.

On vCenter, I rescan and check the Datastore size on each ESXi : 2To

After a moment (minutes, hours), the Datastore size on each ESXi is incorrect ; it's the old value : 1To. I refresh and the new size is displayed...but... after a moment, the Datastore size on each ESXi is incorrect. The Device size is always OK.

When I log in directly on the Hosts, the size is always OK.

Environment

vCenter 6.0.0 (u2a) build 4541947 - 2016-11-22

DataCenter A - Cluster 1 :

All ESXi 6.0.0 build 4600944 - 2016-11-22 => Checked. I read this topic vmware esxi - VSphere web-client shows incorrect datastore capacity - Server Fault

DataCenter B - Cluster 1 :

ESXi in 5.1 version, but they are no connection to the Datacenter 1 SAN Storage.

Thanks for your help.Smiley Happy

Reply
0 Kudos
5 Replies
Finikiez
Champion
Champion

I would recommend to perform a rescan on _each_ host connected to your datastore and restart management agents on __each__ host  just in case.

cholmqvist
Contributor
Contributor

hi

Did you solve your problem? I got the same problem, increased our datastor +1 TB but vsphere show wrong status 5 minutes after refreshing.

Reply
0 Kudos
Finikiez
Champion
Champion

As I wrote - do rescan on each ESXi host where the datastore is connected. This will help.

Reply
0 Kudos
ameyd
Contributor
Contributor

Discrepancy in the vCenter server DB and Host is causing this problem, Showing an incorrect value.

Solution 1 - Perform a cluster level re-scan

Solution 2 - Perform a re-scan individually on each host.

Reply
0 Kudos
ThompsG
Virtuoso
Virtuoso

Hi there,

Sorry to interject but it’s not always cosmetic. The vCenter view now picks an object to view the space from rather the actual object. This can mean that one of the hosts actually hasn’t seen the additional space due to the rescan not happening on a host. Part of our standard OP now is to rescan on each host rather than at the cluster and this does mean at the console sometimes.

This particular feature has burnt us more than once. There is meant to be a fix getting released (was scheduled for end of last year) to resolve. We haven’t patched yet so not sure this resolves the problem. 

Kind regards.

Reply
0 Kudos