[help] How to get target name in the slurm log name

[eheinzen]

Help

• I understand and agree to https://books.ropensci.org/targets/help.html.

Description

I'm just starting with crew, and I can't seem to figure out: is there a way to get the name of the target getting submitted in my slurm job? For example, I've got

crew.cluster::crew_controller_slurm(
    name = "bigger_slurm",
    workers = 6,
    seconds_idle = 2,
    options_cluster = crew.cluster::crew_options_slurm(
      script_lines = "module load R/4/4.2.2",
      script_directory = here::here("logs/submission_scripts"),
      memory_gigabytes_required = 100,
      cpus_per_task = cpus,
      partition = "himem",
      log_output = here::here("logs/parallel_%j.log"),  ## this is where I want the target name to get inserted
      log_error = NULL,
      time_minutes = NULL
    )
  )

but there are so many parallel jobs that I can't find the individual logs of each target where things are getting printed out (e.g., in this case, details on the convergence of a gradient descent). Any ideas on how to get the target name in the log's file name?

[eheinzen]

Ah hang on, I misunderstood that crew uses persistent workers, so multiple targets get sent to each slurm job. I had in my head more of the future.batchtools approach where they are transient and hence one-job-per-target. Thanks for the great work and amazing packages!