Looks like no one’s replied in a while. To start the conversation again, simply ask a new question.

Strange behavior with Quick Cluster and one shared machine

Hi, I've been struggling for a while with my compressor 4.0.7 distributed processing setup... Here is my configuration:


1 Mac Mini running as a Quick Cluster with services (3 instances of compressor shared)

1 Macbook Pro running Services Only (7 instances of compressor shared)

1 Macbook Pro (much older) running Services Only (2 instances of compressor shared)


All 3 systems are running Mavericks, but I had this same problem when all 3 were Mountain Lion.. Had to trash the app and all of the prefs in Library\Application Support\Apple Qmaster and Compressor and reinstall to get the app running on all of the upgraded machines, so a fresh start didn't fix this issue.


The Mac Mini and the older Macbook Pro can see the cluster, submit jobs and are both utilized for rendering when jobs are submitted from the Mini or the older Macbook Pro


The newer MBP can see the cluster in Share Monitor and in qadministrator, but cannot see any of the jobs in the history (the other Macbook Pro can see all details identically to the Mac Mini). When a job gets submitted from this system it appears as "Not Available" on the Mac Mini share monitor and it only utilizes 1 local process to do the rendering, which stalls out after about 1 minute. Activity Monitor shows all 7 instances of compressord are running and not frozen but have no activity.


Jobs submitted from the Mini and older MBP attempt to use the newer MBP for distributed rendering but stall out after about 30 seconds with a host error. The shared volume never appears on the newer MBP. Qadministrator on the Mini can see the newer MBP and all of the listed services as available.


Now here is the part that really blows my mind:


After submitting a job to the cluster from the newer MBP, which will stall out and need to be cancelled as mentioned above; submitting a job from the mini will actually successfully use the services on the newer MBP. Share monitor on the newer MBP still does not display any jobs on the server cluster. Rebooting the newer MBP puts me right back in the "I won't play with those other macs" tantrum.


Anyone else see this issue and have a fix for it? Workarounds are nice but this is very, very annoying when I get into crunch time.

Posted on Nov 7, 2013 7:41 AM

Reply

There are no replies.

Strange behavior with Quick Cluster and one shared machine

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple ID.