careful-dress-39510
04/20/2023, 10:52 PMvictorious-lawyer-58417
04/20/2023, 10:55 PMrun --with batch:queue=...victorious-lawyer-58417
04/20/2023, 10:55 PMvictorious-lawyer-58417
04/20/2023, 10:57 PMMETAFLOW_BATCH_JOB_QUEUE if you want it to have an effect only on steps that are already decorated with @batch.
run --with batch applies to all stepscareful-dress-39510
04/20/2023, 10:59 PMcareful-dress-39510
04/21/2023, 7:51 AMvictorious-lawyer-58417
04/21/2023, 7:54 AMcareful-dress-39510
04/21/2023, 7:54 AMcareful-dress-39510
04/21/2023, 10:58 PMpython src/my_flow.py --environment=conda resume --with batch:queue=new-queue --origin-run-id sfn-c12757fd-d8f9-4bb7-b9ad-01bf9a39ae93 --tag prod startvictorious-lawyer-58417
04/21/2023, 11:00 PMcareful-dress-39510
04/21/2023, 11:00 PMcareful-dress-39510
04/21/2023, 11:01 PMvictorious-lawyer-58417
04/21/2023, 11:04 PM@batch in the code?careful-dress-39510
04/21/2023, 11:05 PMcareful-dress-39510
04/21/2023, 11:05 PMvictorious-lawyer-58417
04/21/2023, 11:06 PM@resources instead? Or you can try the environment variable instead
unfortunately the command line --with batch won't override the @batch in the codecareful-dress-39510
04/21/2023, 11:07 PMvictorious-lawyer-58417
04/21/2023, 11:08 PM@resources it'll run on batch when you do run --with batch. It should work exactly as before, unless you need to hardcore some batch-specific attributes in the code (besides the usual cpu/memory)careful-dress-39510
04/21/2023, 11:10 PMstep functions create so I guess we just need to ensure that has the --with batch yeah?victorious-lawyer-58417
04/21/2023, 11:14 PMstep functions will automatically execute everything on batch, so --with batch is not needed in that case
Just when you do run manually