[HDFS-12611] Ozone: SCM Cli: Use the hostname of SCM server instead of the bind address

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: HDFS-7240
• Component/s: ozone
• Assignee: Marton Elek
• Reporter: Marton Elek
• Created: 06/Oct/17 21:43
• Updated: 26/Apr/18 21:54
• Resolved: 09/Oct/17 18:33

Description

During tests I found that I can't use SCM CLI in multi-node environment:

2017-10-06 21:00:25 INFO  Client:939 - Retrying connect to server: 0.0.0.0/0.0.0.0:9860. Already tried 0 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1000 MILLISECONDS)
2017-10-06 21:00:26 INFO  Client:939 - Retrying connect to server: 0.0.0.0/0.0.0.0:9860. Already tried 1 time(s); retry policy is RetryUpToMaximumCountWithFixedSleep(maxRetries=10, sleepTime=1000 MILLISECONDS)

The reason is that we should use the OZONE_SCM_CLIENT_ADDRESS_KEY instead of OZONE_SCM_BIND_HOST_KEY as the bind host could be 0.0.0.0 and the SCM_CLIENT_ADDRESS contains the external address of the SCM server (in my case scm was on a different host)