Check whether the outcome improved
Replay healthy and backlog flow
Use the same work classes, arrival mix, item sizes, consumer settings, and dependency conditions as the baseline in a controlled environment. Record depth, oldest age, arrival, completion, service time, retries, and final outcomes. Keep the observation window long enough to include both buildup and drain.
Check ordering, loss, and isolation
Verify required ordering, no missing items, no duplicate effects, and correct handling of a failing or slow item beside unrelated work. Confirm new arrivals remain visible and critical work is not silently starved. Compare identifiers and final state across the whole sample.
Observe customer-visible recovery
Follow representative items from queue entry through the action they support. Check stale or partial status, retry messaging, final records, and recoverability after a worker or dependency failure. Continue after depth normalizes to catch residual retries, delayed notifications, or incorrect completion.
What to carry forward
Classify backlog behavior as recovered, regressed, unchanged, or inconclusive with flow, age, isolation, and final-state evidence attached. It passes only when work drains without loss or duplication, required ordering holds, unrelated classes proceed, and delayed outcomes become trustworthy.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow