Skip to content

docs(deploy): note set_replica_identity partial-failure semantics#193

Open
dimitri-yatsenko wants to merge 1 commit into
mainfrom
docs/deploy-partial-failure-note
Open

docs(deploy): note set_replica_identity partial-failure semantics#193
dimitri-yatsenko wants to merge 1 commit into
mainfrom
docs/deploy-partial-failure-note

Conversation

@dimitri-yatsenko

Copy link
Copy Markdown
Member

One-line addition to the set_replica_identity spec (deploy-operations.md), following the #1466 implementation landing on datajoint-python master.

The spec's Behavior section documented idempotency but omitted the failure mode: the operation is not atomic — tables are altered one at a time, so if execution raises on table N of M, the first _N_−1 ALTERs have already committed and the exception propagates without returning a summary. This matches the behavior already documented in deploy.py's docstring; re-running converges since each ALTER is idempotent.

Surfaced during a spec-vs-implementation consistency check of datajoint/datajoint-python#1466.

The spec omitted the failure mode: tables are altered one at a time, so a
mid-run error leaves earlier ALTERs committed and returns no summary. Matches
the behavior documented in deploy.py's docstring. Re-run to converge.

@MilagrosMarin MilagrosMarin left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Matches deploy.py's "Partial-failure semantics" docstring on master exactly, and closes the spec-vs-implementation gap from my #1466 review point #7. Approving.

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