[{"data":1,"prerenderedAt":23},["ShallowReactive",2],{"content:work-infrastructure-renewal:en":3},{"slug":4,"title":5,"sector":6,"scope":7,"summary":8,"challenge":9,"solution":10,"technologies":11,"outcome":18,"details":19},"infrastructure-renewal","Infrastructure you can rebuild in an afternoon","E-commerce","Infrastructure & operations","Moving a hand-built hosting setup to reproducible, monitored infrastructure.","An online shop ran on servers configured by hand years earlier. Deployments happened at night, backups had never been restored, and peak season meant anxiety.","All environments described as code, a deployment pipeline with automated tests and rollback, centralised logging and alerting, and a restore procedure that is exercised regularly.",[12,13,14,15,16,17],"OpenTofu","Docker","PostgreSQL","CI\u002FCD","Prometheus","Grafana","Illustrative example. The goal of this kind of engagement is routine daytime deployments and a documented, tested recovery path.",[20,21,22],"Staging environment identical to production, created from the same code.","Alerts tied to customer-facing symptoms such as checkout errors, not just CPU load.","Runbooks for the most likely incidents, written with the client’s team.",1794663560424]