"Failed to run medaka consensus" about medaka

[dominik-handler]

Hi,

I have the same problem but interestingly only on one of our cluster environments. I use the very same singularity container on both systems so medaka should be experiencing the very same environment on both systems. The biggest difference might be the CPU version, do you think this can be the cause of this?

Working:
NAME="CentOS Linux"
VERSION="7 (Core)"
CPU=Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz

Not working:
NAME="CentOS Linux"
VERSION="7 (Core)"
CPU=Intel(R) Xeon(R) Gold 6138 CPU @ 2.00GH

[cjw85]

Hi @sxh1136 and @dominik-handler

Apologies for the lack of reply until now.

We have not been able to reproduce this error; I have a suspicion it may be how the conda package is built. Admittedly we do not use the conda builds too often, though I know other users (@wdecoster?) do.

You may wish to try building from source (check the build-wheels.sh script in the repository), or to use the pypi package instead. The pypi package does not require compilation but unlike the conda package will require you to install samtools, minimap2, tabix, and bgzip yourself

It may be possible to install the pypi package into a conda environment, though I'm not sure if this will resolve your issue.

[sxh1136]

Hi @cjw85

I was able to solve my issue by installing medaka in a conda environment with a virtualenv.

[wdecoster]

In my case, I also use a conda environment, on a CentOS7.