We set the oplog space size to 50GB, and our data size is about 9000W. We keep the resumeToken in memory, we also use hasNext() and next() to get the change stream data. When a large amount of data is updated, it blocks in the hasNext() method and cannot return true or false, so next() could not execute. But when mongo client is restarted, hasNext() will not block with the same resumeToken. Why are we blocking in the hasNext() method?