Vikas Vishwakarma created HBASE-18633:
-----------------------------------------
Summary: Add more info to understand the source/scenario of large
batch requests exceeding threshold
Key: HBASE-18633
URL: https://issues.apache.org/jira/browse/HBASE-18633
Project: HBase
Issue Type: Bug
Affects Versions: 1.3.1
Reporter: Vikas Vishwakarma
Assignee: Vikas Vishwakarma
In our controlled test env, we are seeing frequent Large batch operation
detected warnings (as implemented in HBASE-18023).
We are not running any client with large batch sizes on this test env, but we
start seeing these warnings after some runtime. Maybe it is caused due to some
error / retry scenario. Could also be related to Phoenix index updates based on
surrounding activity in the logs. Need to add more info like table/region name
and anything else that will enable debugging the source or the scenario in
which these warnings occur.
2017-08-12 03:40:33,919 WARN [7,queue=0,port=16020] regionserver.RSRpcServices
- Large batch operation detected (greater than 5000) (HBASE-18023). Requested
Number of Rows: 7108 Client: xxx
2017-08-12 03:40:34,476 WARN [7,queue=0,port=16020] regionserver.RSRpcServices
- Large batch operation detected (greater than 5000) (HBASE-18023). Requested
Number of Rows: 7096 Client: xxx
2017-08-12 03:40:34,483 WARN [4,queue=0,port=16020] regionserver.RSRpcServices
- Large batch operation detected (greater than 5000) (HBASE-18023). Requested
Number of Rows: 7091 Client: xxx
2017-08-12 03:40:35,728 WARN [3,queue=0,port=16020] regionserver.RSRpcServices
- Large batch operation detected (greater than 5000) (HBASE-18023). Requested
Number of Rows: 7102 Client: xxx
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
