Skype Room Systems v2: known perf issue for Surface Pro (2017)-based devices

So according to the engineer handling our support ticket, the deployment of 3.1.112.0 is at 100% now. We've seen some units update however the overall impression so far is pretty average. It doesn't appear to have made the units more unstable, but it hasn't changed the situation much when it comes to HDMI ingest. Our experience is essentially the same and we're still seeing frequent lockups too (usually when a meeting isn't even running), sometimes the app is able to recover itself after 15 minutes or so of being locked up but sometimes it'll remain locked up until someone manually intervenes by rebooting the unit.

The behavior of the application when it is locked up is really strange too. Yesterday I logged into a frozen unit via RDP as 'admin' and used procexp to look at the frozen SRS process. It looks like right around when it crashed the application suddenly started spawning a new thread approx. every second until it reached over 500 threads and suddenly stopped. The process didn't show as suspended and all the threads appeared to be idle, there were no open network connections and the DesktopAPIService log indicated that there was no activity from the app to the desktop bridge.

I'm going to monitor the situation when it comes to thread count and see if there's a correlation between a high thread count and crashes.

David GroomIf I recall correctly, the freezing issue was acknowledged and attributed to a previous update which a subsequent update was supposed to have resolved. This freezing issue is far less frequent than it was back then but we're still seeing at least 4 units each day lock up. Has your team considered this issue resolved or are you aware of an ongoing issue still?