We had an incident in production caused by this error. I know that this can happen and we have fixed that issue. But what I can't understand is why we for several hours got the same exception for exact same key. We
are using AtomicMapLookup.getFineGrainedAtomicMap, I think this should give us a new map when the old has been removed. What could be the cause of this behavior, we are using 5.2.1.Final.