Hi,
Observium is not designed to operate as two synchronised peer servers. There is no native mechanism for database synchronisation or active/active clustering between independent Observium instances. High availability at the database layer (e.g., clustered MySQL/MariaDB) is technically possible, but it introduces significant operational complexity and is generally not recommended unless you already operate a mature database HA platform.
In most environments, Observium is a support and monitoring system rather than a customer-facing service. Because of this, the most effective way to ensure continuity is not clustering, but a robust and well-tested recovery strategy.
Our recommended approach is:
• A single primary Observium server • Automated daily database backups • Backup of RRD data and configuration • Documented restore procedure • Optional cold standby server prepared to assume service if required
With this model:
*
Device additions and configuration remain consistent (single source of truth)
*
Failure modes are predictable
*
Recovery is controlled and verifiable
*
Operational complexity remains low
In practice, this approach provides more reliable outcomes than attempting to build active/active redundancy around a monitoring platform.
Thanks, adam.
Ashraf Alkheroo via observium wrote on 02/03/2026 12:08:
We are currently planning a deployment involving two Observium servers and would appreciate clarification regarding the appropriate edition and recommended architecture.
Our objective is to implement a high-availability (HA) setup with the following requirements:
• Two Observium servers operating together.
• If a device is added for monitoring on one server, it should automatically appear on the other server.
• The database should be synchronized between the two servers.
• The overall solution should provide redundancy and ensure service continuity if one server fails.
Could you please confirm:
• Which Observium edition supports this type of multi-server / HA deployment?
• Whether native database synchronization between two servers is supported.
• Any recommended architecture or best practices for achieving full HA with shared data consistency.
We would appreciate your guidance in selecting the correct version and architecture for our needs.
Kind regards,
Ashraf ALkheroo
Network and System administrator
Support : +41 (0)21 544 29 29
Internet · VoIP · Network · Hosting
Consulting · IT Management
observium mailing list -- observium@lists.observium.org To unsubscribe send an email to observium-leave@lists.observium.org
