@johann yeah I might have went about this a goofy way - it's probably more simpler to just expose a new field postgres_db.db_params in the storage config, and then just append the urlencoded values to the generated postgresql connection string

More complicated than expected - will have to update the k8s run launchers to take in postgres_connection_string_secret rather than postgres_password_secret

clarify localhost

fix fixture import

This script looks better placed in python_modules/automation since it's generating some persisted artifacts (the query snapshots) - it should probably be invoked as dagster-graphql-client.

Add to graphql documentation as well? there are directive you can use to make the classes and associated methods show up

Dropping in favor of inspecting the pipeline run's step_keys_to_execute

god bless alex

comment

fix tests

add tests

^ flake

rebase

I think from the introductory line in https://docs.dagster.io/concepts/repositories-workspaces/workspaces#workspaces, it should make sense that workspace lives under dagit.

clarifying comment on why dagit needs a workspace

@johann I'm cautious around putting this under dagster-user-deployments since that would generate a manifest used by the parent chart, using values from the subchart. That's a bit confusing. We already do that with dagster-user-deployments.enabled, which is not good.

up

+johann

+yuhan

back to !=

update layout

developing -> contributing

test 8

test 7

test 6

test 5

test 4

test 3

test 2

update rabbitmq hostname

+alex