gyfora commented on code in PR #375:
URL:
https://github.com/apache/flink-kubernetes-operator/pull/375#discussion_r976892681
##########
flink-kubernetes-operator/src/main/java/org/apache/flink/kubernetes/operator/service/AbstractFlinkService.java:
##########
@@ -182,7 +184,13 @@ public JobID submitJobToSessionCluster(
throws Exception {
// we generate jobID in advance to help deduplicate job submission.
var jobID = FlinkUtils.generateSessionJobFixedJobID(meta);
- runJar(spec.getJob(), jobID, uploadJar(meta, spec, conf), conf,
savepoint);
+ Configuration runtimeConfig = removeOperatorConfigs(conf);
+ runJar(
+ spec.getJob(),
+ jobID,
+ uploadJar(meta, spec, runtimeConfig),
+ runtimeConfig,
+ savepoint);
Review Comment:
I looked at the code and I think for sessionJobs we should not remove any
configs. When we submit the jar through the rest api the configuration is not
passed.
Removing operator configs here can actually break some artifact fetcher
mechanisms within the operator here.
We don't have very good test coverage it seems that would catch this
problem. I will open some tickets
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
