-
Notifications
You must be signed in to change notification settings - Fork 96
Migration progress: include DFSA records in the history log #3039
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…lled. Work in progress.
This field is encoded as a Spark SQL LONG, which has a (signed) range of 64-bits.
THe history will be maintained adjacent to the crawler framework.
…ion, updated to use the Historical record type.
… as default values.
❌ 49/50 passed, 1 failed, 2 skipped, 1h7m16s total ❌ test_running_real_migration_progress_job: AssertionError: Workflow failed: migration-progress-experimental (31m40.612s)
Running from acceptance #7013 |
…ables and UDF refresh. Previously it was implicit: there was a transitive dependency via the grants refresh.
This is currently failing due to a bug in the crawlers that means the snapshots cannot be loaded when the Spark-based runtime is being used; fixed in #3046. |
## Changes This PR fixes an issue with the DFSA and used-table crawlers that could prevent loading of the snapshots. When loading they convert the rows to dictionaries using `.as_dict()` which isn't available on rows provided by the spark-based lsql backend. Instead `.asDict()` needs to be used. Incidental changes: - An existing integration test was updated to also test snapshot loading for these crawlers. - Another test was renamed to fix a typo in the name. ### Linked issues Relates to #3036, #3039. ### Tests - existing unit tests - existing integration tests
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm
…tances instead of attaching the timezone.
Following a discussion, we've decided not to include DFSA records in their current form in the history table. Each DFSA record corresponds to a problem with another resource (eg. notebook, jobs). As such the intent is to aggregate these records and include them in the list of failures on the resource-specific record. |
…ons in addition to the normal type-based ones (#3068) ## Changes This PR cherry-picks some changes from #3039 that updated the `HistoryEncoder` to work correctly with databases that are declared with `__future__.__annotations__` in effect. When this annotation is in effect, python converts all type-hints during import/declaration into strings and then performs deferred resolution at a later stage. (This is why forward references work.) Unfortunately the dataclass mechanism captures field types prior to deferred resolution. This PR ensures that our type checking works anyway. ### Linked issues Cherry-picks from #3039. ### Tests - updated unit tests
Changes
This PR follows on from #2743 by extending the set of updates that we capture to include updated
DirectFsAccess
snapshots for dashboards and jobs.Linked issues
Follows #2743.
Functionality
migration-progress-experimental
Tests