Hi, I posted this question on Stack Overflow a few days ago and hoping I can get some eyes on it: https://stackoverflow.com/questions/79600594/ignitedatastreamer-resulting-in-partition-states-validation-has-failed-partition. Essentially, if there's a partition map exchange running while I stream data in parallel into a cache (partitioned with 1 backup), I see WARN logging about partition update counters being inconsistent *after *the final streaming job has finished.
Is this a bug? It seems to contradict the IgniteDataStreamer API where "data streamer doesn’t guarantee [...] data consistency until successfully finished." Full details with some additional observations and sample code to reproduce are included in the original post. Thanks in advance! John
