Really Long time running stacks

[frsepulveda]

Hi!

I’m running denovomap from stack since Friday (13/04/2020) on my 87 samples and it took a really long time. How can I know if my analysis still running? or how can speed up?
I used stack because stacks2 has not -m option on the denovo analysis.

[jorge_arias]

hola francisco. Estoy con un problema similar al tuyo en 2020. Estoy trabajando con 48 individuos y pasados unos días Galaxy me dice “cluster could not end job”. Quería saber si en su momento a usted le resulto correr Stacks. Por ahora supongo que es una cuestión del tiempo de corrida y que por algun motivo esto falla. Saludos

[jennaj]

Hi @frsepulveda

Maybe this prior answer helps?

@jennaj

An orange/yellow dataset means that the job has been dispatched to a cluster and the final result is not back yet. Quitting out at that stage limits what kind of feedback you’ll get in the job logs. So, unless you already know the job will fail for some reason, it is best to let those run to completion.

Jobs will take as long to execute in Galaxy as they would anywhere else given the same input/tool/parameters. Why? Galaxy itself is doing exactly that – sending the job to a cluster.

A long running job could be for many reasons: large data, computationally expensive parameters or the tool itself is computationally expensive always.

The help section of tool forms generally have link outs that provide more author or publication details about expected runtimes given resources. Once your job finishes, you can go into the Job Details view to review what Galaxy was using, and estimate from there.

Several FAQs cover how jobs run at public servers: Galaxy Training! and I also added a tag to your topic.