Jump to content
Welcome to our new Citrix community!
  • 0

VSAN Datastore Capacity Issues Following XenDesktop VDA 7.15 to 1912 Update


Jason Nash1709151307

Question

Last Wednesday evening we upgraded the VDA version on our master image from 7.15 to 1912. Since this point when machines boot we see a huge spike in disk utilisation when VMs are running. We  also upgraded our Storefront and FAS instances but as of yet we have not upgraded our controllers to 1912 they are still on 7.15.

 

We took a VSAN host into maintenance due to a fan failure today and the drop in disk space caused over 150 desktops to fall over and go unregistered due to the datastore running out of space. Usually we can take a host out of the cluster for maintenance with no issues but we are now using over 50% more space during the working day.

 

The attached graph shows disk space utilisation on our VSAN datastore over a 1 month period. Ignoring the top line the first two peaks are Thursday and Friday last week following the VDA update then this flattens as we are closed for the weekend. You can then can see yesterday and today to the far right. Prior to the 7.15 to 1912 VDA update there was no peak at all during the day and the line was fairly flat all be it a gradual increase.

 

I have reached out to support but has anyone seen this before or have any ideas?

Capture.PNG

 

**Edit** 

 

 - Just to add the hypervisor is VMware (VMware ESXi, 6.5.0, 15256549)

 

 - Catalog is MCS Random allocation with temporary cache to memory (384MB)/overflow to disk (15GB). It appears the temporary cache disk is the issue here, these are now significantly larger on disk that they were before. For  example I have a machine that is currently using 89MB on the disk cache drive but within VMware the VMDK is over 10GB. We can also see that the cache disk is also now visible within Windows Explorer (MCSWCDisk)

Link to comment

1 answer to this question

Recommended Posts

  • 0

We resolved this one in house in the end by uninstalling the old VDA and then reinstalling, on reinstall ensuring the MCSIO filter driver optional component was selected. We also upgraded the catalogs and delivery groups and deleted and reprovisioned the VMs. There were two issues the main one was that the MCSIO driver was not present and the second one the MCSIO drive was not correctly formatted, this may has been as we have upgraded and downgraded the VDA a few times in our test catalog.

 

I can now see the MCSIO drive which contains the vhdx, we no longer see snapshot data against these VMs.

  • Like 1
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...