| I am not sure if this would help but it did for me:
The art script does output the failure reason, but it's gobbled up and not stored anywhere. You can manually run the script in Terminal and find out the reason it's returning 0. The best way to do this is to use a PCP submission that's already prepped but not processed. Since you're getting an art score of 0, I'm guessing you have some to choose from. ;)
1) Log into one of the failing grid agents as the user running the grid jobs (eg: ssh email@hidden).
2) Pick one of the waiting recordings. You'll find them on the shared file system in the Recordings folder (eg: /Media/Recordings).
3) In the recording's folder, open the properties property-list file (properties.plist), and find the Server UUID.
3.9) You'll need the following information for step 4:
- The <path to the shared file system> (eg: /Media)
- The <Server UUID> (from step 3)
- The <workflow bundle name> for a workflow you're trying to run (eg: Blog.pwf)
4) Enter the following command, replacing the angled bracket arguments with the values from step 3.9.
ruby <path to the shared file system>/Caches/Workflows/<workflow bundle name>/Contents/Resources/Tools/art.rb <path to the shared file system> <Server UUID>
Eg: ruby /Media/Caches/Workflows/Blog.pwf/Contents/Resources/Tools/art.rb /Media 15A5F943-EDDD-4DE2-AAED-0F23A5056E43
5) You should get a message indicating why the art score is 0.
King Chung Huang
Information Technologies
University of Calgary
|
