[
https://issues.apache.org/jira/browse/FLINK-5823?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16220211#comment-16220211
]
ASF GitHub Bot commented on FLINK-5823:
---------------------------------------
Github user StephanEwen commented on a diff in the pull request:
https://github.com/apache/flink/pull/4907#discussion_r147089150
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/checkpoint/CheckpointCoordinator.java
---
@@ -116,6 +117,10 @@
* accessing this don't block the job manager actor and run
asynchronously. */
private final CompletedCheckpointStore completedCheckpointStore;
+ /** The root checkpoint state backend, which is responsible for
initializing the
+ * checkpoint, storing the metadata, and cleaning up the checkpoint */
+ private final StateBackend checkpointStateBackend;
--- End diff --
No, not really. Wanted to pick a name here that allows us later to possibly
have different "backends" for the data structure, meaning heap and RocksDB in
different operators. One state backend needs to be the root that defines how
metadata is checkpointed and recovered.
> Store Checkpoint Root Metadata in StateBackend (not in HA custom store)
> -----------------------------------------------------------------------
>
> Key: FLINK-5823
> URL: https://issues.apache.org/jira/browse/FLINK-5823
> Project: Flink
> Issue Type: Sub-task
> Components: State Backends, Checkpointing
> Reporter: Stephan Ewen
> Assignee: Stephan Ewen
> Priority: Blocker
> Fix For: 1.4.0
>
>
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
