In some cases, restarting Volicon Encoder Services can cause a spike in host Memory usage which will not allow the Volicon services to start correctly. The solution is to modify an Encoder Configuration file so that the Observer Encoder service can be restarted with the individual Encoders in a down state. Once the Probe or Encoder Service is restarted, then the Encoders can be restarted manually one at a time to avoid the spike in Memory usage.
To do this:
1. Make sure Volicon Probe Manager is not open. On the Volicon Probe, as an Administrator user, open in a text editor the following file. It is recommended to use Notepad++ which is installed on Volicon Servers by default:
c:\Program Files\Volicon\Encoder Service\Encoder Config\EncoderAgent.xml
2. In this file there is a section labeled <encoders>. Underneath it will be an entry for each encoder on the Probe similar to the following:
<encoder name="encoder" run="true" enabled="true" dummy="false" creationxml="C:\Program Files\Volicon\Encoder Service\Encoder config\encoder.xml"/>
3. On each <encoder name=...> entry edit the "run" and "enabled" tags from "true" to "false". Save these changes. It might be necessary to elevate to a Windows user with Administrative privileges to do this.
4. In Windows Services (services.msc), stop the Observer Encoder service. It may be necessary to stop the Observer Encoder process in Windows Task Manager if it does not respond to being stopped in Services. Do not use the service restart option. Make sure to fully stop the Observer Encoder Service.
5. Re-start the Observer Encoder Service.
6. Open the Volicon Probe Manager. Locate the Observer Encoder item. Click on the + sign to expand the available Encoders.
7. As you highlight each Encoder, on the right hand side of Probe Manager there is an option to "Start Encoder". Manually start each encoder one-by-one. Make sure that an encoder fully starts before trying to start the next encoder.
As each Volicon Encoder starts, it consumes a large amount of system memory. Once it has started, this memory usage will decrease an amount. If the encoders are restarted manually one-by-one as instructed above, you should be able to avoid this spike in memory usage and resume normal operation of the Probe.