Bug report

Expected behavior and actual behavior

No warnings should appear in the log file. It seems that the pipelines are unable to detect failing jobs, instead waiting for the timeout before retrying the task, which significantly increases the runtime. This behaviour has been observed in nf-core/cutandrun and nf-core/scflow.

Steps to reproduce the problem

I believe all pipelines/runs with pbspro as their executors should be affected.

Program output

WARN: [PBSPRO] queue status cannot be fetched
- cmd executed: bash -c set -o pipefail; qstat -f $( qstat -B | grep -E -v '(^Server|^---)' | awk -v ORS=' ' '{print "@"$1}' ) | { grep -E '(Job Id:|job_state =)' || true; }
- exit status : 143
- output      :

Environment

• Nextflow version: 24.10.1
• Java version: 23.0.1
• Operating system: Red Hat Enterprise Linux (version 8.5)
• Bash version: GNU bash, version 4.4.20(1)-release (x86_64-redhat-linux-gnu)

Additional context

The problematic code snippet appears to be qstat -B | grep -E -v '(^Server|^---)' | awk -v ORS=' ' '{print "@"$1}', which produces two outputs in my case- @pbs and @.

While qstat -f @pbs executes as expected, qstat -f @ results in an error, which is the root cause of the issue.