Excessive logging from liquibase.changelog

[ivakru]

Apologies, but I can not currently access jira.

First time user question:

The problem

When running cromwell locally, I get an excessive number of messages (thousands) from liquibase. Here is an example:

Jan 31, 2022 5:36:07 PM liquibase.changelog
INFO: Custom SQL executed
Jan 31, 2022 5:36:07 PM liquibase.changelog
INFO: ChangeSet metadata_changesets/remove_non_summarizable_metadata_from_queue.xml::delete_non_summarizable_metadata_from_queue::mcovarr ran successfully in 1ms
Jan 31, 2022 5:36:07 PM liquibase.changelog
INFO: Index IX_WORKFLOW_METADATA_SUMMARY_ENTRY_MAS dropped from table WORKFLOW_METADATA_SUMMARY_ENTRY

Is it possible to control these from cromwell? Is this a liquibase issue? The standard -DLOG_LEVEL=WARN does not seem to effect log messages.

• Version: cromwell-74.jar
• Backend: local
• Java version: 17.0.2+8 (azul)

Workflow:

version 1.0

task say_hello {
  input {
    String name
  }

  command {
    set -euxo pipefail
    echo "Hello ~{name}"
    echo "Hello ~{name}" > greeting.txt
  }

  output {
    File greeting = "greeting.txt"
  }

  runtime {
    docker: "debian:bullseye-slim"
  }
}

workflow hello {
  input {
    String name
  }

  call say_hello {
    input: 
      name = name
  }

  output {
    File greeting = say_hello.greeting
  }
}

[natechols]

Just noticed the same thing. Not critical but it looks like a logger got misconfigured.