Hbase read replica performing Compaction - Strange issue

Question

We enabled Hbase read replica - However we bump into this situtation we observed that read replica cluster is still doing compaction and Primary cluster is reporting a FileNotFound Exception for multiple regions.

What could be the reason for this ? Is there any specific configuration parameter that is enabled ?

shubham_sharma · Answer

Hi @csguna

We can check below for read replica properties which can be configured -

https://docs.cloudera.com/cdp-private-cloud-base/7.1.9/hbase-high-availability/topics/hbase-read-replica-properties.html

Compaction would be carried out for primary region according to design. Flushes, compactions and bulk loads can change the list of files. This means that the secondaries will need to discover the file changes, but until all secondaries are in sync, the old store files should not be deleted.

Sign up

Please use your Cloudera account.

Login to the community

Please use your Cloudera account.