Skip to main content

Active-Passive


Overview

One active JFrog Platform deployment and a passive deployment that is ready to take over.

Why

  • The passive site is already in place when the active site must fail over
  • The organization keeps a single URL in normal operation and during failover
  • Platform configuration and repository data are already on the passive site
  • This is a common disaster recovery pattern

How

  • Run the active deployment at jfrog-a.my.org and a passive one, usually in another data center, at jfrog-b.my.org
  • Point a global DNS CNAME, jfrog.my.org, at the active site
  • On failover, update that DNS name to jfrog-b.my.org
  • Keep the platform configuration in sync with Platform Federation, and the data in sync with repository federation. Self-managed Platform Federation is in early access. See Disaster Recovery
  • Review Limitations

Architecture

The following is a schematic high-level architecture of this topology.

Active-Passive Active-Passive