I am using Dagster to run meltano, via the excellent extension plugin. However I am discovering an issue when using Dagster timeouts and retries. I have a timeout of 900s applied to a job, and then 2x more retries on failure. However, after the first job timing out, I see the two retry jobs failing with this message (see image)

2023-09-25T13:19:57.906790Z [error    ] Another 'prod:tap-dynamics-to-target-postgres-small-batch' pipeline is already running which started at 2023-09-25 13:00:22.623619. To ignore this check use the '--force' option.

I've seen the message before, but normally when I am manually interrupting/ restarting pipelines when testing locally, not in the automated context. Is this due to how the job is being killed by dagster? How does meltano know to 'close' a run that did not complete due to error?