DrDaveD

> > Another way to approach this would be to provide an allowlist to be used in conjunction with --cleanenv. > > Sorry I thought the method I am suggesting...

I didn't remember there was an existing issue because it wasn't marked with the 1.3.0 milestone. That's fixed now.

That commit was inherited from the https://github.com/apptainer/singularity repository. If you look at the [same commit](https://github.com/apptainer/singularity/commit/36f05c8c1b736628f17dac276d888dbc2cc8aa8a) there it leads to apptainer/singularity#2202 which has a little more context but not much. It...

Please make a PR to see if it works.

I doubt that the INFO message is related to the fatal error. This looks more like a race condition to me. Does the same script with the same $SLURM_JOB_ID run...

It wouldn't make sense to turn the fatal error about an already existing instance into an INFO message, because it can't go on and start another instance with the same...

But there would be nothing for it to go on to do at that point, it would need to exit. What you probably would like is some additional option that...

Yes, under one flock command.

If `/data` and/or `/data2` are on the host you may need `-B /data` and/or `-B /data2` options to bind them into the container.

Since none of the usual contributors to apptainer use nix, we would need you or another nix user to submit a PR that works on nix without breaking all the...