Hi! We've encounter this issue quite a few time recently with

HaplotypeCaller

and

mutect2

when we run sarek 3.8.1 on our HPC w/o internet access.

OMP: Error #15: Initializing libomp.so, but found unknown library already initialized.
  OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see <http://openmp.llvm.org/>

If we resume the runs it seems like we can get past it and for most runs it doesn't happen at all. So we can still get the work done but it's a bit frustrating. Any idea what could be causing this? We've been running this sarek version for quite some time w/o problems so I'm a bit confused why this is suddenly appearing..

I would say there's 2 OMP runtimes at the same time

But is that something we can control, or do we just run it and hope for the best?

yeah, but that's still the same pipeline you're running, so nothing has changed there

Is this w docker or conda or …?

singularity

No experience with singularity, but I think there’s a possible clash between system libs and the ones in the image

Did something change on the cluster Jonas?

No, I can't think of anything that could have changed. We have only seen this during the last week and the cluster have been under heavy use so I don't know if that can be connected somehow. Anyway, since it's seems kind of random and we still can get the analyses done I don't think we'll do much about it atm

your guess is as good as mine, but that looks probable, it could even be one mutect2 and 1 haplotypecaller