Issue with 4-quiver/merge-reads step

[BenjaminSchwessinger]

Hi there, I am running falcon-2018.07.25-07.01-py2.7-ucs${UCS_VER}-beta.tar.gz installed with the wrapper found here https://pb-falcon.readthedocs.io/en/latest/quick_start.html .

Falcon and unzip all work fine till the merge-reads step in quiver.

When the pipeline tries to execute the following script it fails.

#!/bin/bash
printenv
echo
set -x
python2.7 -m pwatcher.mains.fs_heartbeat --directory=/home/smrtanalysis/benjaminS_analysis/Pst_104E_v2/FALCON_180725/run_1/4-quiver/merge-reads --heartbeat-file=/home/smrtanalysis/benjaminS_analysis/Pst_104E_v2/FALCON_180725/run_1/mypwatcher/heartbeats/heartbeat-Pfa7ea4563f981b --exit-file=/home/smrtanalysis/benjaminS_analysis/Pst_104E_v2/FALCON_180725/run_1/mypwatcher/exits/exit-Pfa7ea4563f981b --rate=10.0 /bin/bash run.sh || echo 99 >| /home/smrtanalysis/benjaminS_analysis/Pst_104E_v2/FALCON_180725/run_1/mypwatcher/exits/exit-Pfa7ea4563f981b

with exit code 256.

If I submit the same job manual with the full qsub command. The exit code is 0.

I am struggling to trouble shoot this any further at the moment as I cannot find the full error message anywhere. The bam files with the reads were generated with bax2bam. The input_bam.fofn also contained the srcapps.bam. Might this have been the issue.

I am re-running the quiver step with the latest version of falcon falcon-2018.08.08-21.41-py2.7-ucs4-beta.tar.gz and without the scrapps.bams in the input_bam.fofn and got the same issue.

Any pointers what the issue could be would be greatly appreciated.

[BenjaminSchwessinger]

A bit off sleep fixed my issue. My input_bam.fofn also had the bam.bai files in them which generated an issue with this version of falcon unzip while previous version just handled this fine. So found the issue. Running now.

[pb-cdunn]

When there is a bad file in input_bam.fofn, should we raise an Exception or print a Warning?

[BenjaminSchwessinger]

I think this would be useful. Old versions of Falcon didn't care about this, hence I ran into this issue.

[pb-cdunn]

But which? An exception, or a warning? I'm not sure.

[BenjaminSchwessinger]

I would use an exception.