dbtrail

Point-in-time recovery for MySQL — no locks, no schema changes, no waiting for a restore.

SELECT * FROM orders WHERE id = 123 AS OF '2026-05-20 14:00:00'

— against production MySQL. That's the experience bintrail makes possible.

Bintrail tails the binary log, keeps every row change with full
before/after images in a searchable index, and turns "someone ran the wrong
UPDATE" from a restore-from-backup incident into a two-minute fix:

• See every change — what changed and when, for every row, with before → after diffs
• Undo precisely — generate exact reversal SQL for just the damaged rows
• Time-travel — query any row (or table) as it was at any moment
• From a web console — browse, recover, and add servers to monitor, all in the UI

It also ships an MCP server, so Claude (or any MCP
client) can search your change history and draft recoveries.

Works with MySQL, Percona Server for MySQL, Amazon RDS for
MySQL, Amazon Aurora MySQL, and Google Cloud SQL for MySQL —
bintrail connects over the replication protocol, so it never needs access
to the binlog files on disk (that's what makes managed cloud databases
work). Requires MySQL 8.0+ with binlog_format=ROW and
binlog_row_image=FULL — bintrail doctor checks both and prints the
exact fix for anything missing.

Get started

curl -fsSLO https://raw.githubusercontent.com/dbtrail/bintrail/main/docker-compose.yml
docker compose up -d
docker compose logs -f bintrail

Nothing to configure. The logs end with your console URL:

Console is running — open it and add the MySQL servers to watch:

    http://127.0.0.1:8090/?token=ab12cd34…

Open it and click + Add server: paste the MySQL you want to watch —
host, user, password — and bintrail runs the preflight checks (failures come
back as fix-this cards), provisions an index for it, and starts streaming.
Watching events within the minute, and the terminal is already behind you.
The console binds to your machine only (127.0.0.1) and every request
requires the token from the URL.

The compose stack ships a pinned MySQL 8.4 container as the index store.
That index holds the forensic record — it is your system of record, so
back up its volumes (lose them and you re-index from scratch). bintrail
ships that MySQL but does not operate it: disk, backups, and
upgrades are yours. Prefer your own? Set INDEX_DSN in .env to a MySQL
8.0+ you run and remove the bundled index services — bintrail installs only its
tables there, never operating the server. Either way, what's supported on
each side is spelled out in SUPPORT.md; for an index that's
operated for you, that's dbtrail.

Other ways to install — plain Docker, .deb/.rpm, go install,
source builds, and the binary quickstart: see docs/install.md.

Just curious? One container, zero setup, time-travel SQL in 30 seconds:
docker run --rm -p 6033:6033 ghcr.io/dbtrail/bintrail-demo — see
the demo image.

How it works

The index is self-contained: recovery never needs the original binlog files,
and old partitions rotate out to Parquet files — on local disk or in any
S3-compatible bucket (S3, MinIO, …); no cloud account needed. Queries read
them transparently either way. Time-travel SQL (AS OF) is served by a
MySQL-protocol shim behind ProxySQL — your clients keep speaking plain MySQL.

Documentation

License

Apache-2.0 — free for any use, including commercial and production.
Contributions welcome: see CONTRIBUTING.md (CLA required,
prompted automatically on your first PR).