- This topic has 5 replies, 2 voices, and was last updated 3 years, 3 months ago by Anonymous.
BloodMemberAug 22, 2017 at 7:56 am #167161
We use a Windows 2012 Standard member server in our domain to control remote access. This has been fine for some time but a few months ago staff reported not being able to establish a VPN connection. The error was the typical ‘connection may have been interrupted’ or similar. I tracked it down to the Routing and Remote Access service not starting after Windows Updates have been installed after the server is restarted. The services console showed the service as Stopped. Starting it allowed VPN connections to be made.
This server takes a while to restart. It runs WSUS and hosts the Sophos Enterprise Console and a stand-alone installation of Sharepoint Search 2010 (free version). It has 8GB RAM and an i3 CPU 2 core/4 logical processors @ 3.10GHz. When running, memory utilisation is about 6GB and CPU is negligible. Opening and closing e.g. Sophos Enterprise Console takes about 20secs so it is not the fastest of servers…
After identifying the failed service issue I used the services console to configure the Routing and Remote Access service to restart after 1st, 2nd and 3rd failures. This did not work. Next, I created a scheduled task that runs 15mins after startup and which runs a batch file with a command to start the Routing and Remote Access service. This worked fine on the next two reboots, but after last week’s round of updates, the service failed to start.
Trawling through the system log I can see the following related to remote access etc. (some services are reported as stopped without a start event log entry preceeding it):
Shutdown initiated at 19:22
Service Control Manager logs services stopping for the next 2mins.
System restarts at 19:26
Network Connectivity Assistant enters stopped state at 19:32
Remote Access Management service enters running state at 19:35
Remote Access Connection Manager service hung on starting at 19:36Remote Access Connection Manager dependencies:Secure Socket Tunneling Protocol service: entered stopped state at 19:24Secure Socket Tunneling Protocol service: entered running state at 19:35Telephony entered running state at 19:35RPC entered running state at 19:26
The Routing and Remote Access service depends on the Remote Access Connection Manager service which failed to start because of the following error: service hung in a start-pending state at 19:36
Remote Access Connection Manager service hung on starting at 19:39
The Routing and Remote Access service depends on the Remote Access Connection Manager service… at 19:39
Remote Access Connection Manager service entered the running state at 19:40
The scheduled task ran at 19:37 and failed with error code 0x8007042C (The dependency service or group failed to start). The remainder of the entries in the system log comprise ‘… service starting’ and ‘Windows installer’ notifications.
I can see that the Remote Access Connection Manager starts at about the same time as the Secure Socket Tunneling Protocol service and the Telephony service.
I have now extended the trigger time for the scheduled task to run 20mins after the system starts. I’ve not had a lot of luck finding information about what might be causing the RACM to hang. Once the RACM service gets going the Routing and Remote Access service starts without any problems.
Anyone had anything similar and know what might be causing this? Is it simply that the server is over-burdened installing Windows Updates and its i3 processor is struggling to keep up?
All thoughts welcome
You must be logged in to reply to this topic.