Dear all, On the small heterogenous clusters of several customers, i have often the following need. I want to know on each node the job that will terminate the latest and get the planed date of completion ( it corresponds to the date when we could pass from draining status to drained status if no new jobs and no premature end) -eventually filtered by partition, node state, on specific list of nodes-
Unless I am mistaken, there is no direct way to obtain it because I is a mix of squeue (for job data) and sinfo (for data nodes) So I have performed it in a bash script : https://github.com/bellenco/HPC-Utils/blob/main/slurm_sinfo_w_end_last_exec_by_nodes.sh Any idea to do it in an easier way ? Could it be an addition feature of a future extended sinfo ? Thanks a lot ! F.
