Skip to content

Conversation

kfindeisen
Copy link
Member

This PR adds a daily chained collection to the central repo, and refactors some other collection-management code.

@kfindeisen kfindeisen requested a review from kherner March 14, 2024 22:05
self._count_datasets(central_butler, ["raw", "calexp"], f"{instname}/defaults"),
0)

def test_export_outputs_retry(self):
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just verifying that is was intentional to eliminate the test_export_outputs_retry function entirely and move the new line into test_export_outputs. Nothing actually called the retry function anyway, right?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See description of dde8987; test_export_outputs and test_export_outputs_retry used to test different circumstances, but they became identical on #64.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@kherner, does that answer your question, or are there outstanding issues with the review?

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good to go.

This test became obsolete with the removal of raw export on DM-39188.
Some of the code for defining an output run has been factored into the
chain definition.
The chain will make it easier for people to do real-time analysis of
the pipeline outputs, since it depends only on the instrument and
processing date. Each chain should be small enough (a few collections
per pipeline per deployment) not to cause performance problems.
Currently, the two collections are identically named, but this will be
important for forward-compatibility.
This run is not used yet (except as a mock for the preload metrics),
but will be needed once we have preload tasks/pipelines.
In addition to setting up access to the buckets, Butler operations need
access to the registry DB.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants