We have been experiencing some issues with writable volumes on AV 2.14.2 using instant clones on vSphere 6.5 up2 and Horizon 7.5.1. We have slowly rolled out users in this environment over the past month, and we are constantly having to delete WV's because the VM cannot access it, it get's locked, or it cannot find it. The issue appears to be related to the datastore's that are assigned in the storage groups for WV's. We have an all flash DellEMC unity 350f connected via 16gb FC with 2 2TB luns dedicated for WV's. Once all users are connected, I have noticed that I cannot browse the datastore (HTML or Flash client, works via SSH), and vRealize is reporting over 200+ms on Max VM Disk Latency for the WV datastores. When I check the SAN, everything appears fine with no bottlenecks or system utilization spikes. All the other datastores (instant clones, replica, app stacks) all report little to no latency and are performing as expected. There are ~130 desktops which each datastore holding about 65 WV's a piece. The storage group is setup as round-robin/direct. The OS is Windows 10 x64 1803 with the latest VMware tools (10.3.2) and AppVolumes (2.14.2). We have 5 ESXi hosts (Dell R740), and 1 App Volumes manager. If anyone has any experience with this issue, or has any suggestions I am all ears. I've had a ticket in with VMware for over 2 weeks and have gotten no where, so I am desperate.
Thanks in advance!
