A deployed public API uses a longer finite deadline for an ordinary membership feed than for a private account feed. How can a client verify successful public reads, exact publication replay and automatic background completion while ensuring private authority and revoked credentials retain their existing boundaries?
A public mixed activity feed can require more than six seconds of database work even when it returns only twenty-five cards. The SQL routine and its application adapter have separate finite deadlines. How should those bounds be coordinated so an authorized read can finish while preserving visibility, cancellation, pag…
A background worker indexes a published problem successfully, then PostgreSQL cancels recommendation completion at a five-second statement limit. A single identical retry can hit the same limit again. How should the SQL, HTTP transport and whole-turn deadlines be aligned so the current-source and version checks can fi…
A public API returns a new discussion identifier before a background worker finishes computing its recommendations. The worker updates only derived context and recommendation rows, but it holds an exclusive lock on the public discussion while performing a second candidate scan. A normal read requests a shared lock and…
A public API returns a new discussion identifier before a background worker finishes computing its recommendations. The worker updates only derived context and recommendation rows, but it holds an exclusive lock on the public discussion while performing a second candidate scan. A normal read requests a shared lock and…
A public API returns a new discussion identifier before a background worker finishes computing its recommendations. The worker updates only derived context and recommendation rows, but it holds an exclusive lock on the public discussion while performing a second candidate scan. A normal read requests a shared lock and…
Shows how to add optimistic concurrency to an existing EF Core entity with a SQL Server rowversion column in one additive migration, deploy it expand-first, and handle DbUpdateConcurrencyException. Use when existing rows must keep working, old code must run against the new schema, and lost updates must stop. Not for P…
Shows how to deploy EF Core migrations to production only as a reviewed, immutable artifact: generate idempotent SQL in CI, allowlist and SHA256-hash it, require explicit opt-in and approval, rehearse on staging, back up, and apply under transaction and advisory lock. Use for gated production EF Core releases. Not for…
Shows how to build a public audit history projection for document approvals that cannot leak private review notes: filter at projection time on both document status and event visibility, store the public read model in a separate table with no private columns, and handle unpublish, rebuilds and counts. Use when exposin…
Expect the Monday failure and design for it: weekend backfills rewrite partitions, so the schema the check validated on Friday is replaced by the backfill output shape, and the fix is aligning the backfill writer schema with the serving schema. Use when a table maintained by both a daily writer and a weekend backfill …
A schema version mismatch means producer and consumer disagree on the contract version: read both version identifiers, diff the two schemas field by field, and either upgrade the consumer to the producer's version or roll the producer back until the consumer catches up. Use when a producer-consumer pair failing on sch…
Makes workflow rollbacks safe by versioning the full bundle of workflow, prompts, and tool contracts, and testing the rollback path in staging. Use when decisions break right after a rollback. Not for bugs in the current version.
Repairs schema-migration backfill failures with versioned schemas, per-version migrators, and resumable idempotent backfill jobs. Use when backfills die on renamed fields or leave half-converted history. Not for volume-driven backfill failures.
Unblocks legitimate schema changes past dbt contracts for analytics engineers supervising repair agents. Use when a contract blocks a schema change the business actually wants. Not for contract violations from bugs.
How to fix a rollback script that references columns that no longer exist because the down migration was never tested: repairing the down migration and adding up/down testing to CI. Use when rollback fails on a stale down migration. Not for rollbacks that fail for data reasons.
Recovery when an agent deployed the app before the migration job finished and new code queries a column that does not exist yet: stabilizing the deploy, finishing the migration, and wiring the ordering so it cannot recur. Use when deploy and migration raced. Not for code bugs unrelated to schema timing.
How to do expand/contract properly after removing a column during the expand phase: restoring the column, finishing the reader migration, and only dropping in the contract phase. Use when a column was dropped while readers still needed it. Not for columns with no remaining readers.
Recovery when a column rename broke microservices that were still reading the old name: restoring compatibility immediately, then doing the rename properly with dual-write and phased reader migration. Use when a rename broke unknown readers. Not for renames with no live readers.
How to fix a zero-downtime deploy broken by adding a NOT NULL column without a default in one step: recovering the deploy, and the expand/contract pattern that makes it safe. Use when old code and new schema collide during a rolling deploy. Not for NOT NULL additions done safely in phases.
Recovery and prevention for an agent that assumed DDL was transactional on mysql: the half-applied migration cannot be rolled back because mysql commits DDL implicitly. Use when a mysql migration failed midway and left a partial schema. Not for postgres, where transactional DDL usually allows rollback.
How to reconcile when an agent squashed migrations locally but production still runs the old files: why squashing applied history breaks, and the safe paths (keep old files, or stamp after proving schema equivalence). Use when a squash rewrote history that production already recorded. Not for squashes of never-applied…
How to recover when an agent applied migrations out of order, skipping a revision: comparing the state table against the file chain, applying the skipped revision safely, and reconciling when its work is already partially done. Use when the schema is inconsistent because a revision was skipped. Not for branching heads…
How to retry a cockroachdb schema change that failed due to concurrent DML: understanding why the declarative schema changer aborts, retrying with backoff, and reducing contention. Use when a DDL statement fails because writes raced it. Not for schema changes that fail on syntax or validation.
Vectle found the following relevant skills: - agent retried a failed migration without checking state first - the retry applied the same DDL twice and corrupted... - [gh-ost error: FATAL cannot acquire lock on table during online schema migratio…