Data migration

Coming from an older Playwright Reports Server instance? A one-time importer registers the reports and results already sitting in your storage backend, so a fresh instance picks up the whole back-catalogue without re-uploading anything.

What it does

• Scans the configured storage layout (fs / s3 / azure) for the original server's {project}/{id} structure and registers the reports and results it finds.
• Serves those reports in place from their original location. It doesn't move, copy, or re-pack anything, so nothing is duplicated.
• Backfills the database (run history, stats) so analytics and flakiness treat the imported reports the same as freshly uploaded ones.

How to run it

The Data Migration panel shows up on the init page - the landing screen of a fresh instance - as an "Import legacy data?" action. It only appears while the reports table is empty, so it's a bootstrap step rather than an ongoing sync; once you have reports the option is gone. The import runs on the server, so you can leave the page and it keeps going; the widget reports progress and completion.

Point the new instance at the same storage (DATA_STORAGE plus the matching S3/Azure or fs volume) as the old one first - the importer reads whatever that backend contains. See Storage and Configuration.

When to use something else

For ongoing ingestion from CI, use the reporter or CLI upload. To pull artifacts out of GitHub Actions on a schedule, that's GitHub Sync. This importer is only for the initial move of an old server's data.