Currently when jobs crash after cluster submission, nothing is printed to the log files, rather we need to fish for the execution command and submit it manually before the file can be read.
We should figure out why
First question should be whether the same jobs print logs on local submission
Currently when jobs crash after cluster submission, nothing is printed to the log files, rather we need to fish for the execution command and submit it manually before the file can be read. We should figure out why
First question should be whether the same jobs print logs on local submission