The workflow hanging after the GTDBtk step

[sgalkina]

Dear authors,

Thanks for the great tool. I'm currently running the following command on Computerome:

maginator -v <...>
-r <...>
-c <...>
-o <...>
-g "/home/databases/gtdbtk/release207_v2"
--cluster qsub --cluster_info "-l nodes=1:ppn={cores},mem={mem_gb}gb,walltime={runtime}"
--max_cores=10 --max_mem=150

The command itself was also run through qsub system.
The GTDBtk step seem to be completed succesfully, but then nothing more hapens. The main job is still running, and all the gtdbtk spawned job are finished. The output folder now looks like that:

total 98
drwxr-xr-x    5 svekut ku_00200  4096 Aug 19 15:54 .
drwxrwxr-x   43 svekut ku_00200 16384 Aug 19 16:05 ..
drwxrwxr-x 1328 svekut svekut   65536 Aug 19 15:51 clusters
drwxrwxr-x    2 svekut svekut   16384 Aug 19 15:58 gtdbtk
drwxr-xr-x    4 svekut ku_00200  4096 Aug 19 14:08 logs
-rw-r--r--    1 svekut ku_00200   979 Aug 19 15:38 parameters.tab

The latest update in the folder was yesterday, and now the main job is running for more than 24 hours. It looks like it is hanging, but I see no errors.

Is it something you could help with?

[trinezac]

It is difficult to know what went wrong. Can you try to run the GTDB-snake "manually" to see if any og the clusters seems to not have completed successfully?

snakemake --use-conda --latency-wait 150 -s MAGinator/maginator/workflow/gtdbtk.Snakefile --rerun-triggers mtime --config wd= reads= contigs= vamb= params=