gyfora commented on code in PR #670:
URL:
https://github.com/apache/flink-kubernetes-operator/pull/670#discussion_r1325961310
##########
flink-kubernetes-operator/src/main/java/org/apache/flink/kubernetes/operator/config/FlinkConfigBuilder.java:
##########
@@ -420,13 +421,24 @@ private void setResource(Resource resource, Configuration
effectiveConfig, boole
? JobManagerOptions.TOTAL_PROCESS_MEMORY
: TaskManagerOptions.TOTAL_PROCESS_MEMORY;
if (resource.getMemory() != null) {
- effectiveConfig.setString(memoryConfigOption.key(),
resource.getMemory());
+ effectiveConfig.setString(
+ memoryConfigOption.key(),
parseResourceMemoryString(resource.getMemory()));
}
configureCpu(resource, effectiveConfig, isJM);
}
}
+ // Using the K8s units specification for the JM and TM memory settings
+ private String parseResourceMemoryString(String memory) {
+ try {
+ return MemorySize.parse(memory).toString();
Review Comment:
Even if Flink memoryutils interpret the decimal bytes incorrectly, we should
not silently change the interpretation as it may cause failure in currently
deployed prod jobs.
We can fix this when we change "major version" ie bump the crd to v1. For
now I think it is fine to interpret the current configs as before (even if
technically incorrect at times) and open a follow up JIRA to fix this for v1.
Wdyt?
--
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: issues-unsubscr...@flink.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
