Unicycler - Remote job server indicated a problem running or monitoring this job.


I used Unicycler on Galaxy in the past but with sequences from Illumina only.
This time I am using it with long reads obtained from a MinION.
I tried to run it twice but the two times I got the same error message:
Remote job server indicated a problem running or monitoring this job.
Is it a problem with the server or is it a problem with my data?

Thanks for help,

1 Like

Hi @vanina.guernier

It is difficult to know from the information given. I am going to assume that you are actually working at Galaxy Main https://usegalaxy.org since that is how your post is labeled (and the error message is one that we report for this tool). Check the URL for where you are working and write back if it is different (and clarify).

I’m running quick tests to see if Unicycler is working at all (there could be a server-side issue). Both with and without long reads, just to make sure incorporating long reads at all isn’t a distinct problem for some reason (shouldn’t be, and would be a new bug).

Please also run two tests:

  1. Run just the short reads, without the long reads.
  2. Run the original job again that includes both.

If both fail, the data could be too large to run OR there could be a server issue. I’ll be able to let you know if a server issue is a factor once my test completes.

If the first succeeds, and the second fails, there could be an input problem with the long reads OR including them now creates a job that is too large to run OR there is a tool bug involving long reads (my tests will catch that, too).

If both now succeed, this was a transient server issue that resolved itself.

More feedback after my tests complete. Please also post back what happens with your test runs. We can follow up from there.

Thanks for reporting the problem!


Unicycler is executing fine under all conditions with smaller test cases (including and not including long reads).

If your job failed again, start by double-checking the format and content of your inputs. Be sure to not skip read QA steps. Once done and if and the job is still too large to execute, then your choices are to 1) test running with different parameters 2) consider downsampling the inputs and/or 3) moving to your own Galaxy server where more resources can be allocated.

I added some tags to your post that will help you to find prior Q&A that covers these choices with more details.