Tardi Gradus

@izahn Yes, I saw #82, but that refers to R 3.2.0 and I don't know to what degree this is still an issue with subsequent versions of R core.

@ltierney Here's the stacktrace: ``` > traceback() 12: .Call(C_R_igraph_add_edges, graph, as.igraph.vs(graph, edges) - 1) 11: add_edges(res, edges = t(as.matrix(el[, 1:2])), attr = weight) 10: graph.adjacency.sparse(adjmatrix, mode = mode, weighted =...

I didn't persue this any further, as in my case it was possible to partition the data such that the error didn't occur for the subsets.

The problems seem to be to do with the splitting off of ```keyrings.alt``` from ```keyrings```. If I don't clean up the build directory and then set ```PYTHONPATH``` to include the...

> I don't think that benchmark is accurate and looking at the implementation, I'm quite convinced that the problem is caused by wrong use of the `maxForks` directive > >...

Each job array has a single working directory, but for each element of a Slurm job array, the environment variable ```SLURM_ARRAY_TASK_ID``` contains the value of the index. This can then...

Looking at this again, I wonder whether it is not already possible to use Slurm job arrays. If I write ```clusterOptions = '--array=0-999'```, can't I make use of ```SLURM_ARRAY_TASK_ID``` in...

> This could work provided, but I guess the aim of this issue is to provide more native integration with jobs array. > > Somehow related to this there's an...

I hadn't been aware of the potential problem with backfilling. That is also something we would take very seriously as Nextflow allows relatively inexperienced users, who might overestimate their job...

Would it be possible to have a release with this fix?