Create a backup
A backup says what to capture, when to run, and where to store the result. You create it once and assign it to one or more clients. Editing it later updates every client it's assigned to. Before you start, you need at least one storage destination and one client with the agent installed. Start the guided setup There are two ways to start: On Backup Schedules, choose + New backup. On a client's page (Clients, then the client), choose + New backup. The client is already selected, and you can pick files by browsing that host. The setup has six steps: Client, What, Options, Schedule, Destination and Review. Use Back and Next to move between them. When editing, you can click any completed step to go straight to it. 1. Client Pick the host this backup runs on. Online clients can be browsed in the next step. For clients that are offline, you type the paths instead. 2. What Choose what to back up: A Docker Compose stack: the whole application. That means the compose file, its environment files (the stack's .env and every env_file: its services use), its named volumes and, optionally, its container images. Tick the stack's docker-compose.yml. Also tick any bind-mounted folders, such as configuration or uploads, so data that lives outside named volumes is included. Files & folders: any files or folders, for example /etc, application data or home directories. A folder includes everything inside it. Tick items in the file browser. Paths are those on the client. If you'd rather type them, or the client is offline, choose Type paths instead and enter one absolute path per line. The console warns you about paths that don't look absolute. For Files & folders, you can also turn on Also capture Docker Compose stacks found inside these folders. The agent then finds compose files under the selected folders and backs up each stack's volumes and images in the same job. For example, select /srv to capture the files and Docker data of every stack under it. 3. Options The container options appear only when the backup captures Docker stacks. Advanced applies to every backup. While the backup runs: Choice Effect Stop containers (recommended) The stack is stopped while its data is copied, so databases and other busy data are captured consistently. The stack is started again afterwards. Keep running (live) No downtime, but a busy database may be captured in the middle of a write and restore in an inconsistent state. With Stop containers, the stack normally stays stopped until the upload finishes. Turn on Restart containers as soon as the local copy finishes to shorten the downtime. The archive is then staged in a temporary file on the host, which needs roughly one archive's worth of free disk space. Option Default Effect Back up named volumes On The stack's data. Only turn it off if you back up that data another way. Include container images On Restores work without access to a registry, for example on an air-gapped host. Archives are larger. Pull fresh images on restore instead of using the saved ones Off Pull images from their registry at restore time. Under Advanced: Skip files matching (regular expression): matched against each path relative to the folder being backed up. Matching folders are skipped entirely. Example: (^|/)(node_modules|\.git|\.cache)(/|$). Skip volumes matching (regular expression): volumes to leave out by name, such as caches, or a database volume you dump separately. Example: _cache$|^tmp_. Compose project name: defaults to the name of the compose file's folder. Patterns are checked when you save, so an invalid expression is reported straight away. 4. Schedule Pick a schedule type: Type Runs Daily Every day at a time you choose Every few hours Every 1, 2, 3, 4, 6, 8 or 12 hours, at a chosen minute past the hour Weekly On the days you pick, at a time you choose Monthly On day 1 to 28 of each month, at a time you choose Custom (cron) A five-field cron expression (minute, hour, day of month, month, day of week). Descriptors such as…
BackupMunk Enterprise documentation