A large financial services organization running Octopus Server on an outdated single-instance setup migrated to a High Availability cluster across a different Active Directory domain. The move involved database migration, shared storage relocation, authentication changes, and DNS cutover. By working through the customer's detailed questions and breaking the migration into small, reversible, checkpointed stages, Octopus's customer engineering team helped the cautious enterprise complete the upgrade safely. The customer now follows a regular six-monthly upgrade cadence and rarely needs support.
Table of contents
A cautious, multinational customerAnswering the questions, one at a timeSmall, safe stepsThe outcomeWhat this means for other cautious teamsTags:Questions this post answers
What steps should I follow to migrate Octopus Server from a single instance to a High Availability cluster with minimal risk?
Break the migration into small, reversible stages with checkpoints rather than one cutover. Steps include notifying users in advance, adding new HA nodes as UI-only nodes behind the load balancer, building HA node VMs ahead of time, putting the existing server into maintenance mode and draining tasks, backing up and restoring the database into the new environment, moving shared storage separately, installing Octopus Server on one HA node first to test local authentication and deployments, then repeating for remaining nodes, enabling Active Directory only after basics are proven, and pointing DNS at the load balancer last. Teams planning a similar high-availability migration can find staged upgrade playbooks like this on daily.dev.
Why would a company move Octopus Server to a different Active Directory domain during a High Availability upgrade?
A financial services organization needed to relocate its Octopus Server HA cluster into a different Active Directory domain as part of consolidating infrastructure, which affected the database, home directory, shared storage, authentication, and DNS. Testing local Octopus user authentication before enabling Active Directory sign-in reduced the risk of locking out teams during the switch. Engineers weighing similar domain and authentication migration risks can track real-world examples on daily.dev.