Revert "YARN-11709. NodeManager should be shut down or blacklisted when it ca…"

[brumi1024]

Reverts #6960

[hadoop-yetus]

🎊 +1 overall

Vote	Subsystem	Runtime	Logfile	Comment
+0 🆗	reexec	17m 20s		Docker mode activated.
			_ Prechecks _
+1 💚	dupname	0m 0s		No case conflicting files found.
+0 🆗	codespell	0m 1s		codespell was not available.
+0 🆗	detsecrets	0m 1s		detect-secrets was not available.
+1 💚	@author	0m 0s		The patch does not contain any @author tags.
+1 💚	test4tests	0m 0s		The patch appears to include 1 new or modified test files.
			_ trunk Compile Tests _
+1 💚	mvninstall	44m 41s		trunk passed
+1 💚	compile	1m 29s		trunk passed with JDK Ubuntu-11.0.24+8-post-Ubuntu-1ubuntu320.04
+1 💚	compile	1m 26s		trunk passed with JDK Private Build-1.8.0_422-8u422-b05-1~20.04-b05
+1 💚	checkstyle	0m 39s		trunk passed
+1 💚	mvnsite	0m 46s		trunk passed
+1 💚	javadoc	0m 49s		trunk passed with JDK Ubuntu-11.0.24+8-post-Ubuntu-1ubuntu320.04
+1 💚	javadoc	0m 41s		trunk passed with JDK Private Build-1.8.0_422-8u422-b05-1~20.04-b05
+1 💚	spotbugs	1m 33s		trunk passed
+1 💚	shadedclient	35m 38s		branch has no errors when building and testing our client artifacts.
			_ Patch Compile Tests _
+1 💚	mvninstall	0m 34s		the patch passed
+1 💚	compile	1m 19s		the patch passed with JDK Ubuntu-11.0.24+8-post-Ubuntu-1ubuntu320.04
+1 💚	javac	1m 19s		the patch passed
+1 💚	compile	1m 16s		the patch passed with JDK Private Build-1.8.0_422-8u422-b05-1~20.04-b05
+1 💚	javac	1m 16s		the patch passed
+1 💚	blanks	0m 1s		The patch has no blanks issues.
+1 💚	checkstyle	0m 27s		the patch passed
+1 💚	mvnsite	0m 36s		the patch passed
+1 💚	javadoc	0m 35s		the patch passed with JDK Ubuntu-11.0.24+8-post-Ubuntu-1ubuntu320.04
+1 💚	javadoc	0m 33s		the patch passed with JDK Private Build-1.8.0_422-8u422-b05-1~20.04-b05
+1 💚	spotbugs	1m 28s		the patch passed
+1 💚	shadedclient	34m 56s		patch has no errors when building and testing our client artifacts.
			_ Other Tests _
+1 💚	unit	25m 7s		hadoop-yarn-server-nodemanager in the patch passed.
+1 💚	asflicense	0m 36s		The patch does not generate ASF License warnings.
		173m 9s

Subsystem	Report/Notes
Docker	ClientAPI=1.47 ServerAPI=1.47 base: https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-7028/1/artifact/out/Dockerfile
GITHUB PR	#7028
Optional Tests	dupname asflicense compile javac javadoc mvninstall mvnsite unit shadedclient spotbugs checkstyle codespell detsecrets
uname	Linux 90faf6d5a35c 5.15.0-117-generic #127-Ubuntu SMP Fri Jul 5 20:13:28 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux
Build tool	maven
Personality	dev-support/bin/hadoop.sh
git revision	trunk / 5648501
Default Java	Private Build-1.8.0_422-8u422-b05-1~20.04-b05
Multi-JDK versions	/usr/lib/jvm/java-11-openjdk-amd64:Ubuntu-11.0.24+8-post-Ubuntu-1ubuntu320.04 /usr/lib/jvm/java-8-openjdk-amd64:Private Build-1.8.0_422-8u422-b05-1~20.04-b05
Test Results	https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-7028/1/testReport/
Max. process+thread count	552 (vs. ulimit of 5500)
modules	C: hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager U: hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager
Console output	https://ci-hadoop.apache.org/job/hadoop-multibranch/job/PR-7028/1/console
versions	git=2.25.1 maven=3.6.3 spotbugs=4.2.2
Powered by	Apache Yetus 0.14.0 https://yetus.apache.org

This message was automatically generated.

[slfan1989]

@brumi1024 Why do we need to revert this PR? Is it because we encountered some issues?

[brumi1024]

@slfan1989 there is an issue with the current implementation: we catch every PrivilegedOperationException - including the ones caused by a user-requested application kill - and then proceed to mark the NM unhealthy. This should not happen. Actually bit down in this class there is a separate exit code handling method for the container launch, which throws a config related exception in cases where the error is truly unrecoverable without admin input, I plan to reuse here as well.

But I'll only have time to work on that next week, until then I think this state is harmful, as after a few applications kills most of the NMs will be marked unhealthy, requiring a restart.