> @Ivan > I wonder if some tests on Jepsen with bookie restarts may find this kind of > issues, given that it is not a network/SO problem If jepsen can catch then normal integration test can. The readers in question, are they tailing with long poll, or just calling readLastAddConfirmed in a loop? What is the configuration in terms of ensemble/write/ack?
I can try to put together a repro too, using the integ test framework. -Ivan
