AT times, when I restart a Region Server, the RPS and Num. Regions won't catch up at all until and unless I do a rolling restart of all the Region Servers. Why does this happen?
HDP version: 3.0.1
Hbase Version: 2.0.2
I could not see any ERROR snippet in the RS logs post restart. However, there are 2 WARNs:BTW, Major Compaction was triggered for a table and during that time, I restarted a Region Server.
2019-05-13 22:42:44,298 WARN [regionserver/prd-hdpcyclops015:16020] throttle.CompactionThroughputControllerFactory: org.apache.hadoop.hbase.regionserver.compactions.PressureAwareCompactionThroughputController is deprecated, please use org.apache.hadoop.hbase.regionserver.throttle.PressureAwareCompactionThroughputController instead
2019-05-13 22:42:44,332 WARN [regionserver/prd-hdpcyclops015:16020] regionserver.SecureBulkLoadManager: hbase.bulkload.staging.dir is deprecated. Bulkload staging directory is hdfs://cyclops/apps/hbase/data/staging
2019-05-13 22:42:44,404 INFO [SplitLogWorker-prd-hdpcyclops015:16020] regionserver.SplitLogWorker: SplitLogWorker prd-hdpcyclops015.XXXXXX.nm1,16020,1557767558749 starting
2019-05-13 22:42:44,413 INFO [regionserver/prd-hdpcyclops015:16020] regionserver.HeapMemoryManager: Starting, tuneOn=false
2019-05-13 22:42:44,421 INFO [regionserver/prd-hdpcyclops015:16020] regionserver.ChunkCreator: Allocating data MemStoreChunkPool with chunk size 2 MB, max count 11796, initial count 0
2019-05-13 22:42:44,423 INFO [regionserver/prd-hdpcyclops015:16020] regionserver.ChunkCreator: Allocating index MemStoreChunkPool with chunk size 204.80 KB, max count 13107, initial count 0
Can anyone please assist me here in resolving this issue?