Look inside and download
$BOAT_API_BASE is https://boat.dev/api/v1.
What pull downloads
boat snapshot pull writes two folders:
They hold the files exactly as they were at that snapshot.
--path extracts only the files under that path. It still downloads the whole snapshot.
pull recreates symlinks that stay inside home_user/ and docker/. On every platform it skips a link that points outside that tree, such as /etc, or one that resolves through another symlink. On Windows it also skips links unless Developer Mode is on or the shell runs as administrator. It prints how many entries it skipped.
pull is for your home folder and your volumes. A snapshot also saves your changes in /etc, /usr, /opt, /root, /srv and /var, other home folders, and folders that you create at /. To get one of these files, ask the file endpoint for its absolute path, as in the curl example above.
Download a saved snapshot
A saved snapshot downloads the same way. Use its snapshot id, not its name.- Get the id:
GET $BOAT_API_BASE/saved-snapshots/<name>returns it insnapshot.snapshotId. - Download:
boat snapshot pull <snapshotId> -o ./restore.
pull. Use getSnapshotDownload, then put the files together on your computer.
You can also see the files, and download them, in the Snapshots tab of the dashboard.
For all calls, read the Snapshots API.
Which snapshots you can download
The newest snapshot of a sandbox stays while it runs and for at least 14 days after it stops. After that, automated retention cleanup can delete it once enabled (details). Older snapshots stay in the list only for a short time, usually less than a day. Resume and fork always use the newest one. Read How long boat keeps data and Restore an older snapshot.- After the deletion, the sandbox stays in your list as expired. Resume and fork fail with
409 snapshot_data_expired. - To keep the data longer, save a saved snapshot before the deletion:
boat snapshot <id> <name>. This works on a stopped sandbox. A saved snapshot does not expire.