Upgrades
This section covers how to upgrade Memgraph to a newer version. The process depends on your deployment type — standalone or high-availability (HA) cluster.
General recommendations
- Recompile query modules after every upgrade. While an old build might still work, skipping recompilation could lead to unexpected crashes.
- Always check the release notes and version-specific guides before upgrading, as some releases require additional migration steps.
- Back up your data before starting the upgrade process.
Standalone instances
Create a backup
Before upgrading, back up your data by creating a snapshot and copying durability files to a safe location.
Stop the instance
Shut down the running Memgraph instance gracefully to ensure all data is flushed to disk.
Install the new version
Replace the Memgraph binary or Docker image with the target version. For Docker deployments, pull the new image tag. For native installs, use your package manager to upgrade the package.
Start the instance
Start Memgraph with the same data directory and configuration as before. Memgraph will automatically recover from the existing durability files.
Always check the version-specific migration guides before upgrading, as some releases require additional steps.
High-availability clusters
Memgraph’s HA setup supports In-Service Software Upgrades (ISSU), allowing you to upgrade cluster nodes one at a time with minimal or no downtime. The upgrade follows a rolling strategy — replicas are upgraded first, then leadership is transferred, and the former main instance is upgraded last.
For a detailed, step-by-step ISSU procedure using Helm charts, see In-Service Software Upgrade (ISSU).
Version-specific guides
Certain Memgraph releases introduce breaking changes or require additional migration steps. Check the version-specific guides before upgrading.