[WIP] [Postgres] Resumable replication per-table

[rkistner]

Some history:
#150 added support to resume initial replication on a per-table level (skipping tables already replicated).
#152 improved table snapshots to be faster when retrying. However, with a 5-minute snapshot timeout, we're still limited to around 2.5 million rows per table.
#160 helped to increase the number of rows slightly, to around 3 million rows per table. This is likely due to results being buffered before the query times out.

This attempts to chunk initial snapshot queries using the primary key, allowing us to resume initial replication, extending the number of rows we can replicate significantly. Right now, this is limited to tables with a single primary key column.

TODO:

• Persist last seen key, allowing us to actually resume the replication.
• Test with every possible key type, or limit supported types.
• Test edge cases with TOAST values.

[changeset-bot]

⚠️ No Changeset found

Latest commit: f1ff6db

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR