I recently tested the parallelism of the code in the cluster. At first, I used 224 cores on the four nodes node02, node03, node04, and node05， the test was no problem. Later, after adding node06, the following problems occurred. I tried a variety of methods It didn’t solve the problem, do you know what could be the cause of this?
Since I just registered and can’t insert too many links, the detailed errors can be seen on my github account.
Here are details about error
Please try setting the limits on the pbs_mom (compute node)
/opt/pbs/lib/init.d/limits.pbs_mom # $PBS_EXEC/lib/init.d/limits.pbs_mom
#restart of pbs services on the compute node is required after updating this file with limits.
or you can set the stack size in the /etc/security/limits.conf at the system level and check the /proc/pid_of_the_process/limits whether this has been applied.