The context · June 2025
Cloudflare’s June 2025 postmortem described an outage affecting Workers KV and dependent services, triggered by a failure in underlying storage infrastructure involving a third-party provider. [1]
A proposed project brief inspired by the period’s developments. This is an idea for exploration, not an announced Sansa product, an approved roadmap or a claim of completed client work.
Map one journey end to end
Select a critical portal journey and record its identity, storage, network and third-party dependencies. Include the services used by support staff during an incident. Agree which operations may be delayed and which must fail safely.
Rehearse in a controlled environment
Simulate a timeout, an unavailable data store and a delayed callback. Observe retry behaviour, user messages and recovery after the dependency returns. Check for duplicate submissions and stale information. Keep the exercise outside production until its boundaries and recovery procedure are understood.
Leave behind an operating tool
The output would be a dependency map, a prioritised set of fixes and a short incident playbook. Repeat the rehearsal after significant architectural changes. The aim is a portal that fails predictably and recovers cleanly, with support staff able to explain what is happening.
Source & context
Cloudflare · Service outage postmortem, 12 June 2025This retrospective was written for the archive in September 2026. The linked primary source documents the announcement or event; the practical interpretation and proposed approach are Sansa’s editorial perspective. Public examples do not imply a client relationship. Product capabilities and guidance may have changed since the period discussed.
Another perspective · June 2025
What the June outage teaches about hidden dependencies
Continue readingWorking through a similar question?
Talk it through with Sansa