Yarn Client always connects to rm1 when RM is in HA

[#1]

ENVIRONMENT: HDP 2.5 with RM in HA.

SYMPTOMS:

gaurav@g1:~> hadoop jar /usr/hdp/2.5.0.0-1245/hadoop-mapreduce/hadoop-mapreduce-examples.jar pi 5 8 
Number of Maps = 5
amples per Map = 8 
Wrote input for Map #0 
Wrote input for Map #1 
Wrote input for Map #2 
Wrote input for Map #3 
Starting Job 
11/02/16 12:20:01 INFO impl.TimelineClientImpl: Timeline service address: http://g1.openstacklocal:8188/ws/v1/timeline
/11/02/16 12:20:02 INFO client.AHSProxy: Connecting to Application History server at /0.0.0.0:10200 
11/02/16 12:20:05 WARN ipc.Client: Failed to connect to server: g2.openstacklocal:8032: retries get failed due to exceeded maximum allowed retries number: 0 java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)

ROOT CAUSE: This a known behavior reported in internal jira BUG-65968.This is not a problem but an expected behavior.No matter which RM is currently active, the application would always try to connect to the rm1, then rm2. Please note that this message is just a warning and wont affect the job run in any way.

[#2]

Is this bug resolved:BUG-65968, what the current status?

[#3]

Hi,

Why this bug is coming only in HDP 2.5.If we have both 2.5 and 2.4 version.But this error is coming only in hdp 2.5.3 Is this a bug by yarn or HDP. Is there a workaround.

[#4]

Is there any config parameter to make it shorter failing over to rm2?

Is below config parameter effective on this procedure?

yarn.resourcemanager.connect.retry-interval.ms=30000