[ https://issues.apache.org/jira/browse/FLINK-6633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16057176#comment-16057176 ]
Stefan Richter commented on FLINK-6633: --------------------------------------- [~cre...@gmail.com], I think this log was already enough to figure out the problem. Good news is, the basic incremental checkpointing works as expected. Bad news is the combination of {{StandaloneCompletedCheckpointStore}} and externalized checkpoints is broken. {{StandaloneCompletedCheckpointStore}} does not recover completed checkpoints on restore and therefore does not re-register them. The fix will first add the checkpoint from which we restored to the {{StandaloneCompletedCheckpointStore}}, so that it can be registered as shared state again, and references from future checkpoints work. > Register with shared state registry before adding to CompletedCheckpointStore > ----------------------------------------------------------------------------- > > Key: FLINK-6633 > URL: https://issues.apache.org/jira/browse/FLINK-6633 > Project: Flink > Issue Type: Sub-task > Components: State Backends, Checkpointing > Affects Versions: 1.3.0 > Reporter: Stefan Richter > Assignee: Stefan Richter > Priority: Blocker > Fix For: 1.3.0 > > > Introducing placeholders for previously existing shared state requires a > change that shared state is first registering with {{SharedStateregistry}} > (thereby being consolidated) and only after that added to a > {{CompletedCheckpointStore}}, so that the consolidated checkpoint is written > to stable storage. -- This message was sent by Atlassian JIRA (v6.4.14#64029)