matlabpool success/failure depending on number of workers

4 views (last 30 days)
Linux cluster with two users - first user has created a matlabpool with 96 workers. second user should be able to create a matlabpool with 48 workers, but can't get more than 10. Any thoughts on possible causes and/or how to troubleshoot would be helpful.
Matlab version: R2013a
Configuration: 18 nodes, 144 workers (8 per), all tests passed in admincenter and validate cluster
User1: matlabpool(96) = success - currently actively running jobs
User2: matlabpool(10) = success
User2: matlabpool(20) = failure
Error using
parallel.internal.pool.InteractiveClient>iThrowIfBadParallelJobStatus
(line 780)
The interactive communicating job errored with the following
message: Cannot rerun task because there are no rerun attempts
left (task has no rerun attempts left).
Original cancel message:
The worker MATLAB exited or was stopped during task evaluation.
MATLAB ended with exit status 9.
  1 Comment
Edric Ellis
Edric Ellis on 3 Nov 2015
This is not expected - I suggest you contact support to help resolve this problem.

Sign in to comment.

Answers (0)

Categories

Find more on MATLAB Parallel Server in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!