Known issues

Last updated: 2023-06-12 11:29

---

Resolved Issues

Slurm v23.02.0 bug with --ntasks-per-node

Resolved date: 2023-06-09

Relevant clusters/services: athena, boole, boyle, crusher, dalton, kelvin, neuro, parsons, pople, seagull, tinney

Description: Incorrect values for SLURM_TASKS_PER_NODE when the slurm job is submitted with flag --ntasks-per-node, due to bug in slurm v23.02.0. Full bug details:

-- Fix regression in 23.02.0rc1 which caused incorrect values for SLURM_TASKS_PER_NODE when the job requests --ntasks-per-node and --exclusive or --ntasks-per-core=1 (or CR_ONE_TASK_PER_CORE) and without requesting --ntasks. SLURM_TASKS_PER_NODE is used by mpirun, so this regression caused mpirun to launch the wrong number of tasks and to sometimes fail to launch tasks.