saintstack commented on issue #739: HBASE-23157 WAL unflushed seqId tracking may wrong when Durability.AS… URL: https://github.com/apache/hbase/pull/739#issuecomment-545720071 Thanks for bringing up the " indicates a last flushed sequence id" message on the master. Good one. It is related. At least for the HBASE-23181 case, it makes sense because an entry goes into SequenceIdAccounting with a sequenceid that is one less than the flush sequenceid (somehow it is getting in after the flush the flush). Once it goes in, the Master starts complaining. You added a whole new flushcache method to HRegion in TestMaxFlushedSeqId.patch. How is it different from the original? The test looks good. What do you see as the output? You can make it so a flush gets a sequenceid that is less than the last edit's sequenceid? The last edit does not make it into the flush? Its out of order? Nice work.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
