We are seeing a scenario where HBase regions are not distributed evenly across all HBase region servers, even though balancer_swicth is set to true. This was working fine without any issues until recently. No config changes are made on the HBase cluster or restarts for past 5-6 months.
I manually triggered balancer using balancer command from HBase shell, but still no change in region allocation. No errors or warn messages in Master logs.
Please let me know what could go wrong and cause this scenario with Hbase.
Thanks in advance for the help.
how to resolve size of data per Region Server which is not evenly distributed?