|
|
Workarounds for PBS Job Failures Due to MPT Timeouts31 Jan, 2019
Recently, there has been an increase in PBS jobs failures due to MPT timeout problems at startup. If this happens to your PBS job, you will typically see several error messages similar to the following: MPT: xmpi_net_accept_timeo/accept() timeout System administrators are working on the problem, and more information will be posted when the next step in the resolution plan is scheduled. In the meantime, please see the following Knowledge Base article to learn how to augment your PBS job scripts to help avoid this problem: MPT Startup Failures: Workarounds We apologize for the inconvenience. If you have any questions, please contact the Control Room at (800) 331-8737, (650) 604-4444, or support@nas.nasa.gov. |
