iPHoP job running for 6+ days on Galaxy Europe (880 genomes)

Hello,

I submitted an iPHoP “Host prediction” job on Galaxy Europe about 6 days ago using a FASTA file with ~880 viral genomes. The job is still shown as “running”, but when I check the job information there is no standard output, no error, and no exit code.

  • Galaxy server: usegalaxy.eu

  • Tool: iPHoP predict (v1.3.3+galaxy0)

  • Job state: “running” for >6 days

  • Job API ID: 11ac94870d0bb33a65cc958d2710008e
    Since the dataset is quite large, I expected it to take some time, but it seems unusual that nothing has appeared in the logs after so long. Could you please check whether this job is actually still running on the backend, or if it is stuck in the queue?

Thank you very much for your help!

Welcome @afiaanjum_7

It sounds like your job is in the execution phase (yellow dataset) which is great! That means it has passed through queueing (gray dataset) and has been dispatched to the clusters.

While a job is running, there won’t be any output to report back to the Galaxy application except for some timestamps on the job details view. This is a bit different from running a tool on your own computer where there might be standard error logs reporting progression through the algorithm. Once your job completes, you’ll get back those logs, and the standard outputs, plus the output datasets.

Then for this part, you are correct!

The job runtime of 6 days for ~880 viruses with this tool doesn’t seem unusual when working at the public clusters. The EU team didn’t reply yet, probably because this doesn’t seem odd so far. There isn’t an easy way to check the scientific progress of a running job and if there was a technical problem, it would fail and report that back. So, the best advice is to let this run to completion! Even if it eventually fails, that is how to get the logs.

And, the server itself seems to be processing lots of work! → How to see the UseGalaxy.eu job queue statistics

I see a job running with your target tool_id. This is probably yours! :slight_smile:

I know this isn’t super satisfying but it is best I can provide unless the EU team jumps in with more advice. The EU server is a bit special in that jobs can run for an indefinite amount of time. I would let this keep running since stopping and restarting will lose the progress and the logs.

Hope this helps! And we can follow up more if this runs for too much longer, ok? The first few days were likely queuing time, so this might have only been “executing” for a day or two so far. :slight_smile:

Thank you very much for the detailed response. I just got the results this morning.

Great! Glad that worked out and thanks for letting us know :rocket: