DumpCat vs BackupSheep

BackupSheep was a paid SaaS from 2017 to 2023 and is now open source (GPLv3), self-hosted with Docker Compose and in beta. DumpCat is an Apache-2.0 backup manager shipped as one binary. Both keep your data on your own infrastructure; here is where they differ.

At a glance

DumpCatBackupSheep
Price and licenseFree and open source (Apache-2.0). No account, no license key.Free and open source (GPLv3). Released as a beta.
How it runsSelf-hosted: one binary with a web UI, running as a systemd or Windows service next to your database.Self-hosted with Docker Compose: web app, queue worker, scheduler, PostgreSQL and RabbitMQ. Connects to databases directly or through an SSH tunnel.
PlatformsLinux (amd64, arm64) and Windows (64-bit).Wherever Docker Compose runs.
DatabasesMySQL 5.7, 8.0, 8.4, Percona Server, MariaDB 10.x and 11.x.MySQL (bundled 8.4 client), MariaDB, PostgreSQL 14 to 18; also cloud provider snapshots and WordPress sites.
Backup methodsLogical with mysqldump / mariadb-dump or parallel mydumper; hot physical backups with XtraBackup / MariaBackup, including incrementals. No locks on InnoDB.Native dumps. Physical backups are not mentioned.
Point-in-time recoveryYes. Continuous binlog archiving; restore to a timestamp, a binlog position or just before a GTID.Binlog backups and point-in-time recovery are not mentioned.
StorageLocal disk / NFS, S3 and compatibles, Google Drive, OneDrive, SFTP, FTP / FTPS, WebDAV, Backblaze B2, Azure Blob, Dropbox and every rclone backend; several per job.26 destinations, including local / NFS, S3 with Object Lock, Google Drive, OneDrive, Dropbox and pCloud.
Encryptionage encryption on your server before anything is uploaded; zstd compression; SHA-256 for every file.Backup files are encrypted with installation-bound keys that the installer creates.
Scheduling and retentionCron-style schedules; retention per storage by count, age or GFS (daily / weekly / monthly).Daily, weekly, monthly or cron schedules, with keep-last retention.
RestoreFrom the web UI: restore under a new database name, test-decrypt every file before dropping anything, integrity checks after each run.One-click restore from the console; every backup is a standalone archive.
Where your data goesStraight from your server to the storage you choose. Nothing passes through a third party.Your own infrastructure: from the BackupSheep instance you run to your storage.
NotificationsEmail, Telegram, Slack, Discord and HMAC-signed webhooks.Email, Slack and Telegram; team permissions and an audit log.

Choose BackupSheep if

  • You also want PostgreSQL, cloud provider snapshots or WordPress sites in the same tool.
  • You already run everything with Docker Compose, and want a REST API for every action.
  • You want S3 Object Lock retention for immutable backups.

Choose DumpCat if

  • You need point-in-time recovery from binlogs, to a second or just before a GTID.
  • Your databases are large: parallel mydumper and hot physical backups with XtraBackup / MariaBackup, including incrementals.
  • You want a native Windows service, or one binary with an embedded SQLite database instead of a Docker stack with PostgreSQL and RabbitMQ.
  • You want a stable 1.x release rather than a beta.
  • You prefer the Apache-2.0 license.

Sources

Facts about BackupSheep were checked on October 18, 2026 against the sources below. Products change: if something here is out of date, please open an issue on GitHub.

  1. BackupSheep on GitHub
  2. BackupSheep README
  3. BackupSheep website

BackupSheep is a trademark of its owner. DumpCat is an independent open-source project and is not affiliated with it.

Other comparisons