We are setting up our windows 10 pool linked clone pool - using a master image Win10 1809 - I have optimized the image with all of the standard tools- I have managed to Nuke to ... My login times and my times to "unlock" sessions are horrible compare to Win 7 32 VM's
Averaging about 40-55 seconds when using either imprivata or HTML.
The issues seems to be with windows session persistence - I can watch (via) console as the Vm "Spins up" everytime i lock and unlock a session.
I do have the pool settings
"Refresh OS disk on Logoff" set to "Never"
"Automatic logoff after disconnect" "after 240 min"
using PCOIP
Etc...
all the standard stuff checked..
Does anyone have any advice on how to Improve the long on and lock unlock time for windows 10
what we got
Win 10 1809 4 Gig 4 CPU 2 Sockets per core 4 Gig RAM 40 Gig HD- View Agent 7.9 Golden Image
Vmware Horizon 7.9 Build 13956742
ESXI: VMware ESXi, 6.5.0, 8294253
VmTools 10.2.1
- If you remove the Imprivata agent from the image and deploy to a test pool, do you still see the long reconnect times?
- Where does it seem to sit the longest on the reconnect process (eg, black screen, blue Windows Welcome screen, Horizon client connecting)?
Is this connecting from a ProveID Embedded endpoint as well? Type 1 Imprivata agent on image?
When you say “unlock” - are you referring to when a user reconnects to the desktop?
I have Imprivata 6.2 on the Golden Image, which is being used to supply the 4 subnet pool of VM.s - We use Wyse (Thin) clients as our "dumb" terminals to pull up the desktop, when the user swipes their badge.
So the user will start the day and login - this always takes a bit of time, they will launch the apps that they need (Outlook, Etc..)
they will finish up with that patient and the swipe their badge (Which Locks the work station) and move on to the next room/patient, where they swipe their badge and the terminal unlocks the session. My current production windows 7 vm’s will unlock in about 9 seconds, compared to Windows 10 40+ seconds to unlock.
Using Wyse endpoints (thin client)
I have tested and have the same times using HTML(connection server URL) verses the Thin client.
- If you remove the Imprivata agent from the image and deploy to a test pool, do you still see the long reconnect times?
- Where does it seem to sit the longest on the reconnect process (eg, black screen, blue Windows Welcome screen, Horizon client connecting)?
Removed Imprivata Agent and started to get improved times, called them and found out that the Appliance I was on needed a build update and the Agent that I was using on the client need to be downgraded to one that they don't "officially support" but works.
So I edit the Golden image and now there is an update on the Golden image that keeps getting rolled back and not installed, and it's screwing up the recompose process so now I am getting Errors displayed in the pool
Error expanded
but removing Imprivata did solve that issue.
had to bump the appliance to "6.3.102.23" that resolved the issue.