A Leading Global Banking Group Re-Platforms Production Hosts With Zero Data Loss

Project context

The client wanted to modernize its entire Elastic Cloud Enterprise estate and its infrastructure. SquareShift ran the upgrade in eight staged phases, targeting zero data loss and no unplanned downtime.

Modernize the entire Elastic Cloud Enterprise estate while re-platforming the underlying infrastructure, with a zero data-loss target and no unplanned downtime.

Re-platform 20 production hosts from RHEL 7 / Docker to RHEL 8 / Podman without service interruption.

Advance the ECE control plane along a safe, staged path (3.1 → 3.3.0 → 3.8.2 → 4.0.x) and upgrade every managed customer deployment from Stack 7.17.5 to 8.x.

Project objectives

Upgrade the ECE control plane from 3.1 to 4.0.x through a validated, version-by-version staging path.

Re-platform all 20 hosts to RHEL 8 / Podman role by role, keeping the platform continuously available.

Upgrade 100% of managed deployments to Elastic Stack 8.x ahead of the ECE 4.0 hard gate.

Retire all legacy RHEL 7 infrastructure and validate the modernized end state.

Challenges

Hosts ran legacy RHEL 7 and Docker, and the platform ran ECE 3.1.

ECE 4.0 required every deployment on Stack 8.x, with zero data loss and no unplanned downtime.

Solution delivered

SquareShift started all eight phases with platform configuration backups and verified system and customer snapshots before any change.

Moved hosts to RHEL 8 and Podman one by one.

Migrated proxies, allocators, coordinators and directors by role, keeping ZooKeeper quorum.

Upgraded all deployments to Stack 8.x before ECE 4.0.

Health-checked each phase in Cloud UI, then retired RHEL 7.

The outcome

20 production hosts were re-platformed to RHEL 8. The work ran in 8 phases, each starting with verified backups. No data was lost during the platform migration.