-
Notifications
You must be signed in to change notification settings - Fork 87
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
Update table-migration workflows to also capture updated migration progress into the history log #3239
base: main
Are you sure you want to change the base?
Conversation
…-status index: this is supposed to be handled explicitly.
…ry log when refreshing the migration status at the end.
❌ 51/55 passed, 4 flaky, 4 failed, 4 skipped, 3h48m9s total ❌ test_hiveserde_table_in_place_migration_job[hiveserde]: TimeoutError: timed out after 0:20:00: (22m53.305s)
❌ test_table_migration_job_refreshes_migration_status[hiveserde-migrate-external-tables-ctas]: AssertionError: Workflow failed: assessment (22m16.883s)
❌ test_hiveserde_table_ctas_migration_job[hiveserde]: TimeoutError: timed out after 0:20:00: (22m54.925s)
❌ test_table_migration_job_publishes_remaining_tables[regular]: AssertionError: Workflow failed: assessment (5m58.772s)
Flaky tests:
Running from acceptance #7725 |
It should return all the tables as a snapshot.
…ng the migration index is available during encoding.
There was a marginal benefit to ensuring the migration progress singleton could be initialized prior to loading the snapshot, but it wasn't really worth the eye-catching local.
@job_task(job_cluster="user_isolation") | ||
def verify_progress_tracking_prerequisites(self, ctx: RuntimeContext) -> None: | ||
"""Verify the prerequisites for running this job on the table migration cluster are fulfilled.""" | ||
ctx.verify_progress_tracking.verify(timeout=dt.timedelta(hours=1)) |
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.
This forces the UCX catalog to be created before table migration, while it was not a pre-requiste before
## Changes Exclude TACL migration in table migration integration tests because these were not asserted, and to speed up the tests and reduce flakiness ### Linked issues Attempt to reduce flakiness blocking CI in #3239 Similar to #3437 in the sense that both PR scope integration tests to a smaller set of resources ### Tests - [x] modified integration tests
@asnare : When you are back, rebase with main and check if the CI still fails, the linked PR make the tests more robust |
Changes
The table-migration workflows already contained tasks at the end that log information about tables that still need to be migrated. The primary purpose of this PR is to update these workflows so they also capture updated progress information into the history log.
Other changes include:
Updating the (singleton) encoder for table-history so that initialisation doesn't trigger an implicit refresh of theMoved to Refactor refreshing of migration-status information for tables, eliminate another redundant refresh. #3270.TableMigrationStatus
data. Instead this is controlled at the workflow level, as intended.Linked issues
Conflicts with #3200 (will need rebasing).(Resolved.)Functionality
updated documentation
modified existing workflows:
migrate-tables
migrate-external-hiveserde-tables-in-place-experimental
migrate-external-tables-ctas
scan-tables-in-mounts-experimental
migrate-tables-in-mounts-experimental
Tests