FAQ and common troubleshooting

My pipeline yields error after a re-run at the kmindex step, what do I do ?

kmindex refuses to erase previous work. It produces three folders at different steps: G, index_workdir, output_querey_unitigs. Erase those to re-run. The default behaviour of the pipeline is to not erase the folders by default if they are not empty, hence this annoying issue.

The download of MicrobeAnnotator is intrerupted by errors, how do I solve them ?

What I found works best is to start the download with aspera connect as recommended. An error showed up around step 11-12 that was solved by using option --no-aspera to complete the database download and build.

The pipeline stops at the first step kmdiff

A known bug causes kmdiff to duplicate the partitions in memory, so very large datasets may be difficult to work with, you will need to play around with the parameter --nb-partitions. Write an issue for kmdiff with the log and correct information.