Thank-you for this. I wondered how great it would be to "reach into the stream," tweak the state of some events, and replay data updates.
As if the bug never happened, with data in a happy state, but with the exact repro for future test cases.
With logging as its own domain of expertise, microservices, and (justifiably) opaque access controls, getting to the root cause requires frustrating persistence.
As if the bug never happened, with data in a happy state, but with the exact repro for future test cases.
With logging as its own domain of expertise, microservices, and (justifiably) opaque access controls, getting to the root cause requires frustrating persistence.