- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi All,
We have a strange issue that we have been seeing occurring over the past couple of months (possibly more).
We run Horizon 2203, with NVIDIA GRID M10s, Windows 10 Instant Clones, with AppVolumes and FSLogix, across 2 completely separate sites.
We have got complaints from users, that all of a sudden the response times on the VDI drops dramatically. They say it becomes very laggy. We have checked the network latency and their latency is like 10ms. there are no issue on their network nor can we see anything on our networks (We have done numerous captures). There is no resource constraints (I had this occur to my own VDI when there were only 10 VDIs across the whole farm). The hardware and software across both sites is exactly the same (Same Golden image), Yet users experience this strange thing. Sometimes a user will log on to a VDI, and it will be perfectly fine, but then maybe a day later, even a week later, things just grind. It is almost like rendering slows down, and you look like you are getting 1fps if that. If the user logs off and logs back on to a new VDI it may "fix" itself, but sometimes you get on to a couple of VDIs doing the same thing. Takes a few log off and on agains for it to come right.
We are using BLAST, only H264 Encoding (I thought it was adaptive encoding that may have been causing the issue, but it still happens after I removed that setting).
I am pretty stumped and wondering if anyone else has experienced this, or have any suggestions of what to try.
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I've seen issues with web sites, especially Google maps pasting the image while moving around. I tracked it down to the OSOT tool adjusting the Visual Effects settings on the parent. Try settings the Visual Effects to "adjust for best appearance" to see if it makes any difference.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Does this happen after the VMs run continuously for a time period? How often are the VMs refreshed/restarted?
Worth checking whether App Volumes is causing this issue by disabling it for a couple of users.
-- If you find this reply helpful, please consider accepting it as a solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@Jubish-Jose Well that is the weird thing, sometimes it happens after a few hours, sometimes a user can have a session for a few days. There is no way we can reproduce the issue manually. I had it happen to 3 consecutive logins I did, until I logged in on the 4th time and the VM was fine, then 3 days later I got the issue.
Most users are saying it seems to happen mainly around the O365 product suite. Other apps seem to be ok.
I was also thinking if this could be an app volume related issue, and maybe to exclude all the office apps from the snapvol.cfg file as they are all installed on the golden image. And maybe also to exclude all the NVIDIA related paths and processes. What is your opinion on that?
We have tested with PCoIP and the issue still happens, so it does not seem to be a BLAST specific issue.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Are you using App Volumes for Appstack or Writable? IF Appstacks, what is the configuration (number of apps, type of apps etc.)? Is App Volumes config same for both sites?
If I have to test without App Volumes, I would disable it completely rather than adding exclusions.
May be worth trying this fling: https://flings.vmware.com/horizon-helpdesk-utility. This will show a user experience matrix which gives useful information on what is causing a poor session experience.
Also, since this happens only for users in a particular site, I would also suspect something with networking specific to that site. I would contact VMware support if nothing else resolves the issue.
-- If you find this reply helpful, please consider accepting it as a solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This sounds like a case I had,
NVidia License Problem, either you have a too high LicenseIntervall set and stale computers are blocking new machines from getting new licenses or you forgot to add the license server while doing nv updates, verify your config please.
Nvidia blocks a machine if it doesnt recieve a correct license and throttles it down.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yeah ... Got it sorted out last week.
There was a licensing issue. the NVIDIA DLS Server stopped handing out licenses. Looked like we hit some bug / memory leak of sorts. I rebooted the servers and it all came back to life again.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Been there, good that you fixed it.
Regards