[FLINK-17796] Respect user specified classpath for application mode

[wangyang0918]

What is the purpose of the change

Currently, when we deploy a Flink application cluster(e.g. Yarn, K8s), C/-classpath could not work properly. Since we override the PipelineOptions.CLASSPATHS in YarnApplicationClusterEntrypoint and KubernetesApplicationClusterEntrypoint. Then client updated configuration will not take effect.

./bin/flink run-application -p 5 -t yarn-application\
--classpath file:https:///flink/shared/dependency.jar ... \
-Dtaskmanager.memory.process.size=4096m \
examples/streaming/StateMachineExample.jar

Brief change log

• Add getClasspath to combine the user specified classpath and program classpath
• Wire the getClasspath to all deployment

Verifying this change

• Manually test on K8s/Yarn cluster

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): (yes / no)
• The public API, i.e., is any changed class annotated with @Public(Evolving): (yes / no)
• The serializers: (yes / no / don't know)
• The runtime per-record code paths (performance sensitive): (yes / no / don't know)
• Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn/Mesos, ZooKeeper: (yes / no / don't know)
• The S3 file system connector: (yes / no / don't know)

Documentation

• Does this pull request introduce a new feature? (yes / no)
• If yes, how is the feature documented? (not applicable / docs / JavaDocs / not documented)

[flinkbot]

Thanks a lot for your contribution to the Apache Flink project. I'm the @flinkbot. I help the community
to review your pull request. We will use this comment to track the progress of the review.

Automated Checks

Last check on commit fd1a7f5 (Fri Oct 16 10:48:45 UTC 2020)

Warnings:

• No documentation files were touched! Remember to keep the Flink docs up to date!

_{Mention the bot in a comment to re-run the automated checks.}

Review Progress

• ❓ 1. The [description] looks good.
• ❓ 2. There is [consensus] that the contribution should go into to Flink.
• ❓ 3. Needs [attention] from.
• ❓ 4. The change fits into the overall [architecture].
• ❓ 5. Overall code [quality] is good.

Please see the Pull Request Review Guide for a full explanation of the review process.

The Bot is tracking the review progress through labels. Labels are applied according to the order of the review items. For consensus, approval by a Flink committer of PMC member is required

Bot commands

The @flinkbot bot supports the following commands:

• @flinkbot approve description to approve one or more aspects (aspects: description, consensus, architecture and quality)
• @flinkbot approve all to approve all aspects
• @flinkbot approve-until architecture to approve everything until architecture
• @flinkbot attention @username1 [@username2 ..] to require somebody's attention
• @flinkbot disapprove architecture to remove an approval you gave earlier

[flinkbot]

CI report:

• 16460c8 Azure: PENDING
• 3edcf14 UNKNOWN
• fd1a7f5 Azure: PENDING

Bot commands

The @flinkbot bot supports the following commands:

• @flinkbot run travis re-run the last Travis build
• @flinkbot run azure re-run the last Azure build

[kl0u]

Instead of having this code in different places (see ExecutionConfigAccessor), wouldn't it make sense to change the ConfigUtils.decodeListFromConfig() to sth like:

public static <IN, OUT, E extends Throwable> List<OUT> decodeListFromConfig(
			final ReadableConfig configuration,
			final ConfigOption<List<IN>> key,
			final FunctionWithException<IN, OUT, E> mapper) throws E {

		checkNotNull(configuration);
		checkNotNull(key);
		checkNotNull(mapper);

		final List<IN> encodedString = configuration.get(key);
		if (encodedString == null) {
			return Collections.emptyList();
		}

		final List<OUT> result = new ArrayList<>(encodedString.size());
		for (IN input : encodedString) {
			result.add(mapper.apply(input));
		}
		return result;
	}

And have this method become simply:

protected static List<URL> getClasspath(final Configuration configuration, final PackagedProgram program) throws MalformedURLException {
		final List<URL> classpath = ConfigUtils.decodeListFromConfig(
				configuration,
				PipelineOptions.CLASSPATHS,
				URL::new);
		classpath.addAll(program.getClasspaths());
		return Collections.unmodifiableList(classpath.stream().distinct().collect(Collectors.toList()));
	}

Of course this will require some more changes in the ExecutionConfigAccessor and all the classes that are affected by the thrown exception, e.g. the executors who do not throw exception, they will now have to bubble up the exception.

[wangyang0918]

@kl0u I have addressed your comments. Please have another look.

[kl0u]

Changes look good @wangyang0918 ! I will merge as soon as my Azure gives green.