Update: July 26, 2019
This section of the forum is now closed; we are working on a new support model for WDL that we will share here shortly. For Cromwell-specific issues, see the Cromwell docs and post questions on Github.

Adding GCloud labels to WDL / Cromwell tasks.

I am running my WDL on Google cloud via the google alpha genomics pipelines command:

gcloud alpha genomics pipelines run \
  --pipeline-file haplotypecaller.yaml \
  --logging gs://my_bucket/logging \
  --inputs-from-file WDL=haplotypecaller.wdl \
  --inputs-from-file WORKFLOW_INPUTS=inputs.json \
  --inputs-from-file WORKFLOW_OPTIONS=options.json \
  --inputs WORKSPACE=gs://my_bucket/workspace \
  --inputs OUTPUTS=gs://my_bucket/GATK_HaplotypeCaller/output
  --label project=sample

The YAML looks like this:

name: WDL Runner
description: Run a workflow defined by a WDL file

- name: WDL
  description: Workflow definition
  description: Workflow inputs
  description: Workflow options

  description: Cloud Storage path for intermediate files
- name: OUTPUTS
  description: Cloud Storage path for output files

  imageName: gcr.io/broad-dsde-outreach/wdl_runner

  cmd: >

  minimumRamGb: 1

Everything runs just fine, and I get my expected output. However, I would like to track costs of each run. However, using the labels of the genomics pipeline only tracks the VM usage of the head node of cromwell, but not the underlying VMs. Is there a means to tracking the VMs spawned by wdl_runner with google cloud labels?

Best Answer


Sign In or Register to comment.