mynameborat commented on a change in pull request #938: SAMZA-1531: Support
run.id in standalone for batch processing.
URL: https://github.com/apache/samza/pull/938#discussion_r264797657
##########
File path:
samza-core/src/main/java/org/apache/samza/execution/LocalJobPlanner.java
##########
@@ -107,10 +111,11 @@ private void createStreams(String planId,
List<StreamSpec> intStreams, StreamMan
LOG.info("A single processor must create the intermediate streams.
Processor {} will attempt to acquire the lock.", uid);
// Move the scope of coordination utils within stream creation to address
long idle connection problem.
// Refer SAMZA-1385 for more details
- JobCoordinatorConfig jcConfig = new JobCoordinatorConfig(userConfig);
- String coordinationId = new ApplicationConfig(userConfig).getGlobalAppId()
+ APPLICATION_RUNNER_PATH_SUFFIX;
- CoordinationUtils coordinationUtils =
-
jcConfig.getCoordinationUtilsFactory().getCoordinationUtils(coordinationId,
uid, userConfig);
+ if(coordinationUtils == null) {
+ JobCoordinatorConfig jcConfig = new
JobCoordinatorConfig(appDesc.getConfig());
+ String coordinationId = new
ApplicationConfig(appDesc.getConfig()).getGlobalAppId() +
CoordinationConstants.APPLICATION_RUNNER_PATH_SUFFIX;
Review comment:
To me it sounds like the `coordinationId` that is constructed here is very
Zk specific i.e. `<app_id>/ApplicationRunnerData`. Can we rename the
`coordinationId` to something more general along namespace that way we can
leave the namespace to be interpreted in the way implementers want to? Having
`/` as part of the application runner path suffix doesn't right here.
Additionally, this would also help us align with the namespacing for
metadata work.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
