What it does
The engine library can mint a snapshot token naming a committed position and re-pin it later for reads that see exactly that state. If the position has aged out of a retention ring, the engine has rewound, or the process restarted, the read fails with a typed error and never returns newer data. No route or command exposes it.
How it works
A token records a log position, the project it was minted for and the engine's rewind generation. Re-pinning through snapshot_at yields a snapshot that reads exactly the state as of that position, so later commits and partial batches are invisible. The retention ring keeps a configurable number of recent roots, 64 by default and capped by a hard ceiling so it cannot grow without bound.
Every failure is closed: a compacted position, a token for a different project, a malformed or forged token and a recovery-uncertain engine each return a distinct typed error. Tokens are authenticated with a standard HMAC and a random per-instance key. The HTTP read-consistency tokens are a separate feature.
Where you reach it
- Library
Engine::snapshot_token, Engine::snapshot_at, SnapshotToken (engine library, not wired to the server)
Limits
- Not reachable from the server or the command line. The admin time-travel route reports a position and a time window and does not read historical state.
- Tokens do not survive a restart, a rewind or retention aging.
Evidence
- TestsCovered in the engine repository by 28 temporal-snapshot tests and 9 snapshot-anomaly tests
Test coverage is listed as present in the engine repository at this edition. This page does not report test results.
KynetraDB is in private preview
This is not available in the preview. Request a preview to discuss the current scope.