I just applied the Patch Tuesday Windows Server 2008r2 patches to a remote server and at the reboot prompt said "yes." The RDP session closed as expected but the sites running on the server continue to function and I can no longer remote into the server.
I have the option to have the hosting company power cycle the server but my fear is that the server is in one of those "don't switch off, currently applying patches" states and a recycle may do serious damage and that's the last I'll see of the server. The server's been in this state for about 2 hours now.
Is there anyway to remotely probe the server to see what it's up to?
There are a number of tricks you can do in this situation to diagnose the problem and hasten the reboot, in the absence of a Remote Management card.
You can try remotely managing the server with your local management tools. For instance, you can run services.msc and connect to the remote computer to inspect the running services. You could run eventvwr and connect to the remote computer to look at the current event logs (perhaps a service is hung while trying to restart?)
You can try using the SysInternals tool "PSExec" to connect to the system, and kill processes manually, or abort the current reboot and re-issue the shutdown command.
I have experienced a similar issue with a Hyper-V R2 SP1 cluster where one of the Hyper-V hosts would regularly take about 2 hours to reboot. RDP sessions would drop, but the host would get stuck on "Shutting down cluster service". In my case, I was able to fix the problem by remotely killing the cluster service with the following command:
taskkill /S Hyper-V-Host /IM clussvce.exe
The server would reboot instantly.
Turns out that the server eventually restarted and I just had to wait for 2 hours.
I'm leaving this question and answer here in case it helps/saves someone else from the current patch Tuesday cycle which appears to have an extraordinary long apply cycle before restarting.