Case study
Disaster Recovery Never Paused During an Upgrade for a Global Investment Firm
How SquareShift upgraded a global investment firm's 16-node ELK Stack from RHEL 7 to RHEL 8 tier by tier, keeping replication and disaster recovery running throughout.
Book a session →
Client
A global investment management firm that offers retirement services, advisory solutions, and financial insights to individuals and institutions worldwide.
Industry
Financial Services
Services
Stack
RHEL (Red Hat Enterprise Linux) · Elasticsearch · Kibana · Logstash · Fleet
How SquareShift delivered it.
The challenge
The client is a global investment management firm offering retirement services, advisory solutions, and financial insights worldwide. Its 16-node ELK Stack environment ran on RHEL 7 and needed an operating-system upgrade, but replication and disaster recovery had to keep running throughout.
What we delivered
SquareShift helped the firm choose RHEL 8, then upgraded all 16 nodes tier by tier — hot, then cold, then warm, then frozen — coordinating the rolling process across production and disaster-recovery clusters so neither stopped.
Along the way, the team fixed shard allocation, LDAP, and alerting problems that surfaced during the move, and delivered a licensing and sizing analysis for the firm's next step. Planning an ELK Stack upgrade? See our Elasticsearch upgrade services.
The outcome
All 16 nodes now run on RHEL 8, with cross-cluster replication and disaster recovery having stayed live through the entire upgrade. The firm also has a licensing and sizing analysis in hand to plan its Elastic footprint going forward.
A rolling operating-system upgrade on a live replication topology only works if disaster recovery is treated as a constraint from the first node, not an afterthought once production is done.
Platform & Software Engineering Practice Lead, SquareShift
