Compare PostgreSQL and MySQL Schema Snapshots

The capture and comparison workflow

An organization administrator with Database Pulse access chooses an enabled PostgreSQL or MySQL target. Capture a baseline, then another snapshot after a change approved outside Spekra. Select the baseline and comparison to view added, removed or changed supported objects.

PostgreSQL captures a selected user schema; MySQL captures the database configured on that target. Register another MySQL target for a different database. Comparing PostgreSQL against MySQL is not supported.

• Read supported table and column metadata without selecting business rows.
• Compare selected ordinary-index attributes.
• Retain snapshots, complete comparison reports and content hashes.
• Download evidence or remove comparisons before deleting their referenced snapshots.

What differences are covered?

Supported fields include column names/types/nullability and selected identity, generated or default-presence markers. Ordinary indexes include selected method, uniqueness and ordered-column attributes. MySQL also reports index prefix length, direction and visibility.

Default/generated expressions, constraints, enum/set labels, triggers/routines, partition details and expression-index definitions are excluded. PostgreSQL partial/expression indexes may be recorded as unsupported inventory. No differences means only that the collected subset matches.

Consistency and retry behavior

PostgreSQL uses a repeatable-read, read-only transaction. MySQL requires two matching consecutive metadata scans and rejects unstable collection. Two matching scans do not create a global DDL lock or guarantee an atomic catalog view.

A completed capture can be replayed with its idempotency key. Changing target or schema under the same key rejects the request. Captures have bounded timeouts/concurrency; failures return errors instead of substituting a demo catalog.

Evidence limits and responsibilities

Each organization permits 100 snapshots and 200 comparisons. Snapshot/report JSON is bounded to 2 MB; collection also limits tables, columns and indexes. The UI displays the first 200 changes; complete retained report JSON is downloadable.

A SHA-256 hash is a content checksum, not an author's signature. Downloads can include object names and operator email addresses. Review them before sharing. There is no scheduled continuous drift capture, evidence re-import, automatic DDL, DDL dry-run validation or production AI diagnosis.

Frequently asked questions

Evaluate your own workflow

Start with supported sample or staging infrastructure and agree on a success measure. Contact the operator to arrange a pilot. This static website does not create accounts or send invitations automatically.