Hello! I am trying to run a job with the following custom k8s configuration.

{    "dagster-k8s/config": {
        'job_config': {
            "volumes": [
                {"name": "volume1", "secret": {"secret_name": "volume1"}},
                {"name": "volume2", "secret": {"secret_name": "volume2"}},
                {"name": "volume3", "secret": {"secret_name": "volume3"}},
            ],
            "volumes_mounts": [
                {"name": "volume1", "mountPath": "volumemount1"},
                {"name": "volume2", "mountPath": "volumemount2"},
                {"name": "volume3", "mountPath": "volumemount3"},
        ],}
        }

}

I am not able to mount volume even though apparently in the source code it is possible to mount volumes

Hi Javier - I see a “volumes_mounts” key there - should that be “volume_mounts”?

yeah sorry I correct this bit and I still have the same error

what is the error?

it returns the following

Python Error: "volumes" key not found

Yeah, so this gets a bit confusing, but in a k8s job the volumes go on the pod spec and the volume mounts go on the container (neither of them go on the k8s job, which is what job_config affects). I think what you want is something like:

{    "dagster-k8s/config": {
        'pod_spec_config': {
            "volumes": [
                {"name": "volume1", "secret": {"secretName": "volume1"}},
                {"name": "volume2", "secret": {"secretName": "volume2"}},
                {"name": "volume3", "secret": {"secretName": "volume3"}},
            ],
        },
        "container_config": {
            "volume_mounts": [
                {"name": "volume1", "mountPath": "volumemount1"},
                {"name": "volume2", "mountPath": "volumemount2"},
                {"name": "volume3", "mountPath": "volumemount3"},
        ],}
        }

}

(I also changed secret_name to secretName)

Created issue at: https://github.com/dagster-io/dagster/issues/6503

I like this bot Daniel

Created issue at: https://github.com/dagster-io/dagster/issues/6505