Only create parallel test databases on schema changes #36826
Closed
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This introduces a new key in the
ar_internal_metadata
table,:schema_sha1
, which tracks the SHA1 of the schema file being loaded into the database.This can be used to keep a test database exactly in sync with a schema file. This is similar to the maintain_test_schema functionality non-parallel tests currently use, but is more accurate (migrations can be changed or rolled back). I plan to update
maintain_test_schema
to use this instead in a future PR.This adjusts parallelization test databases hook to only reload the schema when it has changed, and to no longer drop the database at exit.
Refs #36807