I'm now using HBase-0.98.1-cdh5.1.2 for data storage. In my project, I need to use the feature of Endpoint provided by HBase. Here is my problem.
After I created a table and uploaded some data, the Endpoint was working right. At this moment, split had not happened yet. However, after I uploaded more data, the first split took place and splited the fomer data into two regions. Then the result returned by Endpoint was not complete about the former data. It seemed that part of the fomer data which was splited into second region was missing. But actually I could query all the data in hbase shell. Except the first split, split happened afterwards did not result in more mistakes.