Grid computing - spread workload

RE: Grid computing - spread workload

leo stalin — Sat, 03 Feb 2018 05:39:37 GMT

I followed the same instruction to run grid computing in interconnected system in a same LAN. When i submit job in master, node engine showing ignoring job ; project not found continuously. How to solve this issue in node system.

RE: Grid computing - spread workload

Casey Bentley — Wed, 23 Mar 2016 15:37:32 GMT

I would check the engine on the computer running the job that failed and see if it is telling you why it failed. I found when setting up a cluster, the most common problem was making sure that all the files (photos and project file) were shared so that they could be accessed by all the computers in the cluster.

RE: Grid computing - spread workload

Richard Bath — Tue, 22 Mar 2016 10:04:03 GMT

Thanks for the response Pascal,

I have followed what you've said and the secondary engine shows up in the job que on the main CC master but when the secondary engine tries to run each pending job it fails after 1 second.

Hopefully someone has come across this problem before and can offer some advice.

many thanks

Richard

RE: Grid computing - spread workload

Pascal Cloutier — Tue, 15 Mar 2016 18:32:54 GMT

Hi Richard,

First, you'll need to place your project and images in a shared folder that can be accessed by all computers. You will also need to place the job queue in a shared folder than can be accessed by all machines.

Then, on each computer that is running an engine, open "ContextCapture Settings" and select the "Configuration" tab. From there, select the job queue shared folder. Exit the dialog and start ContextCapture Engine.

Finally, on the computer where the project is located, open ContextCapture Master. Select the top node in the project tree and select the "Options" tab. Toggle ON the "Use proxy UNC path..." option. And, update the job queue shared folder.

HTH,

Pascal