Hi All, I am starting a job which is present in first project (repos) from second project (API) using graphql. I am using

mutation LaunchRunMutation(
  $repositoryLocationName: String!
  $repositoryName: String!
  $jobName: String!
  $runConfigData: RunConfigData!
) {
  launchRun(
    executionParams: {
      selector: {
        repositoryLocationName: $repositoryLocationName
        repositoryName: $repositoryName
        jobName: $jobName
      }
      runConfigData: $runConfigData
    }
  ) {
    __typename
    ... on LaunchRunSuccess {
      run {
        runId
      }
    }
    ... on RunConfigValidationInvalid {
      errors {
        message
        reason
      }
    }
    ... on PythonError {
      message
    }
  }
}
 How can I pass custom parameter to the job ?

@prha are you able to help with this?

Hi @Gowtham Manne … I’m not sure what you mean by custom parameter here. Are you running into some configuration issue?

Hi @prha Thanks for the reply. Not Really. I have job That takes a input of List of string. How can we pass the parameter using above graphql query ??

I think jobs themselves don’t take inputs, but they can compose a bunch of ops that can take inputs. Jobs that contain ops with unconnected inputs may be launched by providing those values in a number of different ways. You can provide those values using an input manager or by providing the values in the run config. You can read more about it here: https://docs.dagster.io/concepts/io-management/unconnected-inputs I’d recommend first making sure you can launch runs in the first project with all the unconnected inputs getting satisfied (maybe using launchpad), and then later using the run config you need to launch via the graphql API.

@prha But here I am starting/launch the job/op using an graphql. Can I pass input parameter for the ops/jobs from using graphql. Like context for example from above example you provided?

In the example “loading-a-built-in-dagster-type-from-config”, the value was provided via run config. The launch run graphql mutation allows you to pass run config in (the 4th argument).

Just wanted to make sure this is the one you are talking about right ?

runConfigData: $runConfigData

Here’s the schema description of

RunConfigData

:

"""
This type is used when passing in a configuration object
        for pipeline configuration. Can either be passed in as a string (the
        YAML configuration object) or as the configuration object itself. In
        either case, the object must conform to the constraints of the dagster config type system.
"""
scalar RunConfigData

It depends on how you’re passing it in (probably via Python?), but you either pass a run config dictionary

# example run config dictionary, from the docs example
{"ops": {"my_op": {"inputs": {"input_string": {"value": "marmot"}}}}}

or the yaml string of the run config dictionary:

# example run config yaml, from the docs example
ops:
  my_op:
    inputs:
      input_string: 
        value: "marmot"

I would recommend you first familiarize yourself with the required run config needed to launch your job in repository A (using launchpad) before then trying to launch using the GraphQL API.