Job submission fails via Xenon CLI on Slurm v19

[lsantuari]

I am getting the following error:

slurm adaptor: Got invalid key/value pair in output: Cgroup Support Configuration:

when running sv-callers with slurm on the HPC, as reported here.

[arnikz]

Thanks @lsantuari for reporting. I could confirm this issue on two HPC clusters (with Slurm 19.05.0 and 19.05.5). In fact, Xenon is not tested with the latest Slurm (only up to v17).

$ xenon --version
Xenon CLI v3.0.4, Xenon library v3.0.4, Xenon cloud library v3.0.2

$ snakemake -C echo_run=0 mode=p enable_callers="['manta','delly','lumpy','gridss']" --use-conda --latency-wait 30 --jobs 14 \
--cluster 'xenon -vvv scheduler slurm --location local:// submit --name smk.{rule} --inherit-env --cores-per-task {threads} --max-run-time 5 --max-memory {resources.mem_mb} --working-directory . --stderr stderr-%j.log --stdout stdout-%j.log' &>smk.log&

$ cat smk.log
...
slurm adaptor: Got invalid key/value pair in output: Cgroup Support Configuration:
Error submitting jobscript (exit code 1):

[arnikz]

For time being, you could use the sbatch instead of xenon command for snakemake --cluster arg:

sbatch -J smk.{rule} -n {threads} --mem={resources.mem_mb} -t 5 -D . --error=stderr-%j.log --output=stdout-%j.log

[arnikz]

Xenon CLI v3.0.5 fixes this issue. Include this fix in the next workflow release.