Feature/hdfs 16303

[KevinWikant]

Description of PR

A rare edge case was noticed in DatanodeAdminDefaultMonitor which causes a DatanodeDescriptor to be added twice to the pendingNodes queue.

• a datanode is unhealthy so it gets added to "unhealthyDns"
• an exception is thrown which causes this catch block to execute
• the datanode is added to "pendingNodes"
• under certain conditions the datanode can be added again from "unhealthyDns" to "pendingNodes" here

How was this patch tested?

All "TestDecommission" & "DatanodeAdminManager" tests pass when run locally

Same manual testing from #3675 applies.

For code changes:

• Does the title or this PR starts with the corresponding JIRA issue id (e.g. 'HADOOP-17799. Your PR title ...')?
• [n/a ] Object storage: have the integration tests been executed and the endpoint declared according to the connector-specific documentation?
• [n/a] If adding new dependencies to the code, are these dependencies licensed in a way that is compatible for inclusion under ASF 2.0?
• [n/a] If applicable, have you updated the LICENSE, LICENSE-binary, NOTICE-binary files?

[hadoop-yetus]

💔 -1 overall

Vote	Subsystem	Runtime	Logfile	Comment
+0 🆗	reexec	0m 0s		Docker mode activated.
-1 ❌	patch	0m 19s		#3922 does not apply to trunk. Rebase required? Wrong Branch? See https://cwiki.apache.org/confluence/display/HADOOP/How+To+Contribute for help.

Subsystem	Report/Notes
GITHUB PR	#3922
Console output	https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-3922/1/console
versions	git=2.17.1
Powered by	Apache Yetus 0.14.0-SNAPSHOT https://yetus.apache.org

This message was automatically generated.