There is likely a problem with the input format, possibly introduced when moving the data between your local and Galaxy Main (truncated upload).
Or, the job is really too large to execute at Galaxy Main.
I would suggest running
Trimmomatic at Galaxy Main as well to eliminate the formatting issues being the problem. The output will be four datasets: two representing the F and R paired reads (the inputs to
Trimmomatic) and two representing the F and R unpaired reads.
Upload by FTP when moving the original fastq data, or paste in the public data URLs (if that is where you sourced it). I am assuming that your local is not publically web hosted, but if it is, there are more options I can share: Loading Data
Trimmomatic job fails again after that, then the job is too large. Options include:
- Subsample the reads. See the
- Upgrade your local Galaxy, if that is where
FastQC was run (??) to have sufficient resources to run the job
- Set up cloud Galaxy server with sufficient resources
More help for options 2 & 3:
FAQ about what to do to fix the different error cases: https://galaxyproject.org/support/