Backfill without overwriting a newer edit
A migration worker and a live request can touch the same row. Make the update conditional on the state the worker actually read.
Read articleCobnex blogs / Australia
Cobnex blog
AI implementation, software architecture, cloud operations and Australian technology policy.
100 articles in Software architecture
Page 2 of 6
Reject an edit based on an old revision
Carry the version a user actually edited into the save request. Compare it atomically with the current record before replacing data.
Read articleTry to bypass a module's business rule
Test the paths most likely to avoid the public contract: direct repository imports, background jobs and administrative updates. The same invariant should hold for each supported caller.
Read articleCrash the publisher after delivery but before acknowledgement
The outbox's duplicate window is predictable. Stop the worker after broker acceptance and verify that replay does not repeat the consumer's business effect.
Read articleSend identical commands at exactly the same time
Sequential retries can pass while concurrent requests still duplicate an effect. Use a barrier to make two callers compete for the same operation identity.
Read articleReplay the error response an older client expects
Error contracts influence retries, forms and recovery. Test them with supported older clients, not only with a server-side schema validator.
Read articleFail the compensation as well as the original step
A recovery path that only works when every corrective call succeeds is incomplete. Test the state left behind when compensation is rejected or its response is lost.
Read articleCreate matching record identifiers in two tenants
Similar records expose missing scope in queries and caches. Use two isolated test tenants and exercise the same operation through every supported path.
Read articleReplay an older event into the new consumer
Historical messages exercise assumptions that fresh test fixtures miss. Test their final effects before relying on replay as a recovery tool.
Read articlePause a reader between its database read and cache write
A controlled interleaving can reproduce stale repopulation reliably. Ordinary load tests often miss this short but important race.
Read articleRehearse the migration while a long transaction is open
A fast command on a quiet database can wait behind live work. Test lock contention and cancellation before the release window.
Read articleMake two requests compete for the last place
A deterministic race test proves the capacity rule more clearly than a large burst of requests that may never overlap at the critical point.
Read articleMeasure the changes that cross module boundaries
Dependency counts are useful signals, but the practical question is whether ordinary business changes remain understandable and local. Review actual change patterns.
Read articleMeasure the oldest unpublished business event
Queue depth shows volume, but age shows how long a business change has been waiting. Track delivery and consumer completion as separate stages.
Read articleCount duplicate effects, not duplicate requests
Repeated requests are normal in a retrying system. The important measure is whether they create additional business actions or leave the caller without a reliable outcome.
Read articleMeasure use of the contract you want to retire
Request counts are a starting point. Identify active callers, their business cycles and the operations still depending on the old contract before setting a retirement decision.
Read articleMeasure unresolved business states in the saga
A coordinator's success rate can hide workflows that remain partly committed. Track the business states and their age, including failed or uncertain compensation.
Read articleMeasure isolation failures outside the main API
Include jobs, exports, caches and support tools in the evidence. A clean set of controller tests does not establish isolation across the application.
Read article