SYMPTOMS:
Cannot access Tez View. On accessing Tez View, we get the below error:
error code: 500, message: Internal Server Error{"message":"Failed to fetch results by the proxy from url: http://<hostname>:8188/ws/v1/timeline/TEZ_DAG_ID?limit=11&_=1470838899327&...Error.","status":500,"trace":"Connection refused"}
ROOT CAUSE:
The issue is due to "YARN ResourceManager URL" pointing out to the standby RM node when you have a RM HA configured. This is usually a configuration change under "Cluster Configuration" while setting up a TEZ view where the user updates opts for the "Custom" option and includes only a single Resource Manager address in "YARN ResourceManager URL" configuration.
SOLUTION:
In case of RM HA configuration , the "YARN ResourceManager URL" should point out to a both the Resource Manager addresses which should be in a semi-colon separated format (e.g. sumeshnode1.openstacklocal:8088;sumeshnode3.openstacklocal:8088)
