Adapted from template-code-location's docker-publish.yml. The repository ships two images, so the workflow builds the code location from the root Dockerfile and the payload from payload/Dockerfile, and tags both with the same short SHA - that shared tag is what keeps a code location and the payload it dispatches on the same version.
Each image gets its own gate before it is pushed. The code location image must load its Dagster definitions and carry payload/work.py for the subprocess transport. The payload image must contain dagster_pipes and must not contain dagster, which is readiness checklist L11; running that assertion against the code location image fails as expected, so the check discriminates rather than passing vacuously.
Changelog: added
Both jobs were launched from the Dagster UI, so the numbers in the new section 5.6 come from the daemon and run launcher path rather than from execute_in_process. The tightly coupled run reports one contributing worker across seven steps; the loosely coupled subprocess run reports four, which is the difference the two patterns are supposed to make visible. Hostnames are redacted, process IDs are not.
Also ignores a local .dagster_home, which the README's dagster dev instructions will create for anyone who sets DAGSTER_HOME alongside the repo.
Changelog: added
The rename of process_work_units and _results_from_pipes during the fan-out refactor left three references behind, in the guide, the checklist and the loosely coupled module docstring. The guide also showed loosely_coupled_k8s_job without its executor_def and claimed the pipes client replaced the executor, which stopped being true once the graph fanned out.
Both images were built and smoke tested: the code location image loads its definitions, and the payload image ships dagster_pipes without dagster. That check is now recorded as L11 in the readiness checklist, so the isolation argument is verified at image level and not only in the source.
Changelog: fixed
The reference graph processed every work unit inside a single op, so exactly one worker was ever reported regardless of executor. That made the guide's claim that k8s_job_executor yields several distinct contributing_hosts false, and left the reference implementations unable to demonstrate the executor choice at all. generate_work_units is now a DynamicOut and both graphs map over it, so one step is created per unit and the loosely coupled pattern dispatches one external workload per unit.
Evidence is split into contributing_workers (host and pid, differs per process) and contributing_hosts (differs only across machines), because the previous single field could not distinguish multiprocess fan-out from no fan-out. Tests now assert the mapped step keys rather than a host count, since execute_in_process ignores executor_def and cannot prove executor behaviour on its own. Adds a Windows note: multiprocess_executor did not complete during authoring and left orphaned processes.
Changelog: fixed
Canonical location for documentation, example workflows and reference service implementations covering distributed execution patterns. Covers AC1-AC4: execution-target selection, decision support, readiness checks and code-level linkage. Tightly coupled jobs and the loosely coupled subprocess transport are verified by the test suite; the Kubernetes pipes transport is implemented but not yet cluster-run and is marked as such in the guide.
Changelog: added
