[AMBARI-25570] Move DataNode RPC metrics graphs to HDFS DataNode dashboard and fix the metric name change

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.7.4
• Fix Version/s: 2.7.6
• Component/s: ambari-metrics
• Labels: grafana pull-request-available
• Assignee: Unassigned
• Reporter: huangkunlun
• Created: 13/Oct/20 08:22
• Updated: 24/Oct/22 09:01
• Resolved: 23/Oct/20 08:25

Description

Some RPC Service Point metrics are lost in the Grafana NameNode page, shows: 'No Data Points', as the attachment shows.

Ambari Version:2.7.4

Grafana URL: http://${Grafana_HOST}:3000/dashboard/db/hdfs-namenodes

The abnormal metrics are shown as follows:

metric	widget name
rpc.rpc.datanode.RpcQueueTimeAvgTime	RPC Service Port Queue Time
rpc.rpc.datanode.RpcQueueTimeNumOps	RPC Service Port Queue Num Ops
rpc.rpc.datanode.RpcProcessingTimeAvgTime	RPC Service Port Processing Time
rpc.rpc.datanode.RpcProcessingTimeNumOps	RPC Service Port Processing Num Ops
rpc.rpc.datanode.CallQueueLength	RPC Service Port Call Queue Length
rpc.rpc.datanode.RpcSlowCalls	RPC Service Port Slow Calls

Issue Links

• links to: Pull Request