Dear HPC experts, Ever since the baobab maintenance finished last week, my interactive jobs started to get killed before they are meant to expire. For example, today I submitted jobs with 8 hours runtime: salloc -n1 -c16 --partition=public-cpu,private-dpnc-cpu,shared-cpu --time=08:00:00 --mem=20G …

Dear @Vilius.Cepaitis , Thank you very much for the details provided. :folded_hands: The issue you are experiencing appears to be similar to the one discussed here: Job timeout despite not hitting the timelimit - #8 by Michael.Sonner . Based on the logs: [2025-12-08T14:56:12.460] sched: _slurm_rpc…

Hi @Adrien.Albert , Thank you very much for your response and for looking into this. I was running a vscode server and snakemake on the node in question, but perhaps the scheduler thought the job was idle. Also, I ran salloc within tmux when it got killed. Best regards, Vilius

Hello, I’m still experiencing this issue. I need my jobs to run non-interrupted as I have snakemake running on the allocated job managing my workflow, which can take up to a day to finish. Do you have any suggestions what to do? Best regards, Vilius salloc: Job allocation 6547165 has been revo…

Hello; Could you please give me your salloc/srun cmd ?

Hi @Adrien.Albert , It’s typically something like this salloc -n1 -c16 --partition=public-cpu,private-dpnc-cpu,shared-cpu --time=24:00:00 --mem=30G Cheers, Vilius

srun -p shared-gpu,private-kalousis-gpu,private-cui-gpu --gres=gpu:4,VramPerGpu:48G --time=7-00:00:00 --mem=128000 --cpus-per-task=32 [...]

Hello Vilius; [2026-01-14T17:30:45.011] job_time_limit: inactivity time limit reached for JobId=6547165 [2026-01-14T17:38:35.992] cleanup_completing: JobId=6547165 completion process took 470 seconds Your job has been killed for ‘inactivity’. The SLURM FAQ: Why is my job killed prematurely? S…

we also test with salloc during one hour doing nothing and the job continue without being killed for inactivity.

@maciej.falkiewicz are you using Tmux or another tool instead shell ? It could be a network issue beetween your machine and the cluster impacting the ssh process. yes @Adrien.Albert , I do use tmux. There can be a network issue between login node and compute node, but the termination wouldn’t ha…

Interactive jobs killed prematurely

HPC Support HPC issues

Yann.Sagon January 19, 2026, 1:29pm 20