The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

is there a k8s_executor configuration where you can set the max concurrent pods? There's runs on the deployment settings, but there may be multiple pods in one run, so a finer-grained control would be nice

we don't have a global limit for this right now unfortunately - you kind of have to approximate with a run limit and then potentially an ops-per-run limit

thanks <@U016C4E5CP8> -- if I tag a large group of assets with the same tag, I should be able to add limit their concurrency right?

I'm trying that and I think it's working but hard to see what's going on

we only support concurrency limits across runs with tags that are applied at the job level

so if I tag every asset in question with {"foo": "bar"}, this won't work unless they're in the same run?
```run_queue:
  tag_concurrency_limits:
    - key: foo
      value: bar
      limit: 40```

those limits won't check tags on assets at all - they'll check tags on jobs