We have been running the hbase cluster for more than 2 years with replication between 2 peers say (ZK1, ZK2). Suddenly some of the rows are not replicated to ZK2 from ZK1. But the value present in the ZK1. Those data were written before 5 months. During that time, we have an issue in connecting ZK2 from ZK1 as there is a problem with the running ZK2 cluster master.
After that, we have identified the issue and fix it. During that time, replication lag status in ZK1 is in the range of 1 billion rows for each region server. All the replication lag has been cleared in 2 days.
But now, on checking some of the rows during that time are missing in ZK2. How to find out what are the rows are missing in the ZK2.
Also, how to find the issue behind the replication.
Please help
与恶龙缠斗过久,自身亦成为恶龙；凝视深渊过久,深渊将回以凝视…
