These are chat archives for nextflow-io/nextflow

25th
Apr 2019
Paolo Di Tommaso
@pditommaso
Apr 25 05:30
This a know problem see here
that check needs to be improved but real issue is that the job fails during the submission
you are using slurm as well, right?
Steve Marshall
@stevemmarshall
Apr 25 05:46
I'm trying to launch some jobs via AWS Batch which works fine however I need to ensure there is sufficient space inside the container for downloading large files. I have an AMI that has a mounted drive 250GB under /scratch I've configured the job definition to map that host volume to a container volume... but it doesn't seem like docker is using that volume and my nextflow task is running under /tmp/...
Paolo Di Tommaso
@pditommaso
Apr 25 05:54
so far, it's only possibile using your own JobDefinition
Pierre Lindenbaum
@lindenb
Apr 25 06:20
@pditommaso yes it's a wrapper over slurm.
@pditommaso do you think there is anything I could quickly implement in my custom executor ? what would be the best place in the code to implement this ? Thanks.
Paolo Di Tommaso
@pditommaso
Apr 25 10:43
it should be possible improving the GridTaskHandler#checkIfRunning logic, so that it also check the job status in the cluster somehow similar to how is done here
Adrian Viehweger
@phiweger
Apr 25 14:21
say I create an assembly from some Illumina paired reads, and want to map some Nanopore data to this assembly. I want to do this for several samples. how can I ensure nextflow coordinates the right assembly with the corresponding Nanopore reads?
Sinisa Ivkovic
@sivkovic
Apr 25 16:51
Hi @pditommaso , I have a quick question concerning resume function https://www.nextflow.io/docs/latest/getstarted.html?highlight=resume#modify-and-resume. Would -resume work if you copy work directory to another location? I tried that and I get WARN: It appears you have never run this project before -- Option -resume is ignored, so I just wanted to check is this expected behavior?