Github user StefanRRichter commented on a diff in the pull request:
https://github.com/apache/flink/pull/3770#discussion_r113898244
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/jobgraph/JobVertex.java ---
@@ -125,6 +132,8 @@ public JobVertex(String name) {
public JobVertex(String name, JobVertexID id) {
this.name = name == null ? DEFAULT_NAME : name;
this.id = id == null ? new JobVertexID() : id;
+ this.operatorIDs.add(new OperatorID(this.id.getLowerPart(),
this.id.getUpperPart()));
+ this.operatorIdsAlternatives.add(null);
--- End diff --
Ok, in this cases it really seems better to make this explicit as you
suggested. Also I was wondering if `operatorIdsAlternatives` should be a
List<List<OperatorID>> -- just want to make sure that only at most one
alternative ID must be maintained per operator. But I think that we can always
determine the savepoint version and only need compatibility to the hasher
version that was valid under that savepoint version.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
