Jobs that donât specify request_memory and exceed 128 MB are getting killed. We would like to increase and have tried setting
JOB_DEFAULT_REQUESTMEMORY = 2048
in configurations files on the noded submitted from, our access point, our central manager and our execute nodes. It does not seem to be having any effect.
condor_config_val JOB_DEFAULT_REQUESTMEMORY does output 2048
Hi Gerard:
JOB_DEFAULT_REQUESTMEMORY only needs to be set on the access
point, but takes effect only for jobs that are submitted after the
parameter is set.
-greg