Background & Context
While working with a leading technology partner to implement an Azure Integration platform, we completed a DR failover test of the platform and application. During this test, we identified a DNS resolution change that required the client to manually complete, in order to move the domain from the active uksouth public IP address to the passive ukwest public IP address.
Following the successful DR test, the client asked us to investigate including either Front Door or Traffic Manager into the Architecture so that the Integration platform support team would be able to self-manage a full DR if needed.
Challenges & Pain Points
During the course of the Proof of Concept (PoC) , we encountered some issues with the use of Front Door, two of which ultimately were raised with Microsoft:
- Front Door did not appear to implement alongside a hardened Key Vault
- Front Door’s health probes did not run under SSL
We were able to resolve and work around the issues encountered, and our recommendation was to implement Front Door and then fall back to Traffic Manager if required. We created production strength Terraform scripts to deploy both Front Door and Traffic Manager in order for the client to decide which to proceed with.
Solutions & Strategies
We worked to an agile methodology, utilising collaborative sprint planning and demos to relay our progress to the client and keep aligned with their expectations. We use Azure DevOps on the client’s subscription, giving them a real-time transparent view of everything we are doing.
Results & Outcomes
The client was walked through a detailed PoC outcome report that provided an updated platform design for DR, Solution Architecture for both Azure Front Door and Azure Traffic Manager, along with the Terraform deployment scripts and failover test evidence. We then updated the existing Integration Architecture with their choice of Traffic Manager.
They were left with a secure application landing zone for Azure Integration Services, providing them with all the cloud infrastructure to build further integration applications in Azure. We handed over a detailed Azure DevOps wiki and completed a full DR test ready for them to accept into Production.
Feedback & Learnings
As the PoC was relatively siloed from the rest of the client’s team, we operated within our subscription in order to not expose their Production instance to any breaking changes. The client was supportive of the costs incurred and actively involved with the Demo and discussions around next steps.
Future Prospects
Off the back of this work, we have continued to work with the client to onboard further integrations and provide additional Architecture, Solution, Development, and Test support for their progressing projects.
Final Thoughts
The client recognised the opportunity to extend their services during the first DR Failover and supported our recommendations to investigate which option was best suited to their needs. They understood the time and technical requirements needed and were supportive and engaging. It was a great opportunity to develop the initial Azure Integration Platform they requested further to support their real day requirements.
Further reading
How we’re taking flight data up into the cloud
Evanta’s 2022 CIO Leadership Perspectives Study found that data and analytics, meeting digital business priorities, and implementing application strategies, architecture and integration platforms were the
Accelerating cloud-based enterprise service bus (ESB) implementation
The challenge A global energy company relies upon thousands of integrations to provide services to every part of the organisation. To transition from this resource-hungry
Replacing the workflow logic engine for a bank’s account transfer system
We recently did some work with one of the UK’s largest building societies and banks. Their challenge was to update a legacy system that was