Hi all,
I am currently trying to run Salmonquant for paired data set collection from each of my RNA-seq sample. I have 40 samples (20 R1/ R2 each) and I set them up as paired collection and run each Salmonquant job separately.
However, I notice that most jobs stuck in queue and for job that run, it is rather random and managed to start processing within 15 minutes.
Are there any known issues with job queue for salmonquant tools?
Welcome @Ldaminh1912
It sounds like your jobs queued and are now being sent out to cluster nodes to process. All of those sub-jobs are independent. This means that the first sample in your batched collection listing is not necessarily the first job that will process and complete. So, all sounds normal to me.
Later on when running a workflow, you might even have some samples complete through several early first few tools while other samples are still pending in the first tool. Once a tool in the workflow is dependent on all of the upstream samples together, then it will queue until all of upstream jobs are completed, and finally that tool will run. You can watch this occur in your workflow invocation view.
The job queue works in batches. Some of your jobs run, some of other people’s jobs run, more of yours, repeat. Getting your jobs into the queue is the important part. Using collections and workflows are the fastest way to do that.
More details are in this guide. →
Most jobs will complete within 24 hours. You can always ask if you notice longer queue timing and we’ll provide some feedback. A shared link to the history would be appreciated since the server, tools, and data involved usually matter.
We hope this helps! 