Merge branch 'master' into squeue-msg

ci/pbs.sh

@@ -9,6 +9,7 @@ cd -
 docker exec pbs_master /bin/bash -c "chmod -R 777 /shared_space"
 docker exec pbs_master /bin/bash -c "chown -R pbsuser:pbsuser /home/pbsuser"
 
+docker exec pbs_master /bin/bash -c "yum install -y procps"
 docker exec pbs_master /bin/bash -c "cd /dpdispatcher && pip install uv && uv pip install --system .[test] coverage && chown -R pbsuser ."
 docker exec -u pbsuser pbs_master /bin/bash -c "cd /dpdispatcher && coverage run --source=./dpdispatcher -m unittest -v && coverage report"
 docker exec -u pbsuser --env-file <(env | grep GITHUB) pbs_master /bin/bash -c "cd /dpdispatcher && curl -Os https://uploader.codecov.io/latest/linux/codecov && chmod +x codecov && ./codecov"

ci/slurm.sh

@@ -6,5 +6,6 @@ docker compose pull
 ./start-slurm.sh
 cd -
 
+docker exec slurmctld /bin/bash -c "yum install -y procps"
 docker exec slurmctld /bin/bash -c "cd dpdispatcher && pip install uv && uv pip install --system .[test] coverage && coverage run --source=./dpdispatcher -m unittest -v && coverage report"
 docker exec --env-file <(env | grep -e GITHUB -e CODECOV) slurmctld /bin/bash -c "cd dpdispatcher && curl -Os https://uploader.codecov.io/latest/linux/codecov && chmod +x codecov && ./codecov"

dpdispatcher/machines/shell.py

@@ -74,7 +74,10 @@ def check_status(self, job):
             return JobStatus.unsubmitted
 
         # mark defunct process as terminated
-        cmd = f"if ps -p {job_id} > /dev/null && ! (ps -o command -p {job_id} | grep defunct >/dev/null) ; then echo 1; fi"
+        cmd = (
+            r"""command -v ps >/dev/null 2>&1 || { echo >&2 "I require ps but it's not installed. Aborting."; exit 1; };"""
+            f"if ps -p {job_id} > /dev/null && ! (ps -o command -p {job_id} | grep defunct >/dev/null) ; then echo 1; fi"
+        )
         ret, stdin, stdout, stderr = self.context.block_call(
             cmd
         )