akalash commented on a change in pull request #18765:
URL: https://github.com/apache/flink/pull/18765#discussion_r813113242
##########
File path: docs/content/docs/ops/state/checkpoints.md
##########
@@ -175,6 +175,35 @@ above).
$ bin/flink run -s :checkpointMetaDataPath [:runArgs]
```
+### Checkpoint/Savepoint guarantees
+Not always the snapshot can be successfully restored.
+The following list contains operations which can be done without recovery lose
depends on the type of the snapshot.
+
Review comment:
Discussable. But I personally don't think that grouping makes sense here
since this grouping is not natural. I mean having groups doesn't give us or
users more understanding or advantages. for example, if we could say that we
support all types of snapshot for group 'Application changes' but only native
savepoint for group 'Configuration changes' it would be useful. But in the
current implementation, it will give us only more lines in the table
--
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]
