[
https://issues.apache.org/jira/browse/FLINK-29913?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17633657#comment-17633657
]
Yanfei Lei commented on FLINK-29913:
------------------------------------
[~roman] Great proposal! I think that maintaining the mapping of <state handle
ID, file name> is necessary, because RocksDB needs the file name(xxx.sst) to
rebuild its instance.
And I have a question about "always generate unique state handle IDs":
if the confirmation notification was missing, TM re-upload some state handles,
how do we delete those duplicate state handles?
> Shared state would be discarded by mistake when maxConcurrentCheckpoint>1
> -------------------------------------------------------------------------
>
> Key: FLINK-29913
> URL: https://issues.apache.org/jira/browse/FLINK-29913
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Checkpointing
> Affects Versions: 1.15.0, 1.16.0
> Reporter: Yanfei Lei
> Priority: Minor
>
> When maxConcurrentCheckpoint>1, the shared state of Incremental rocksdb state
> backend would be discarded by registering the same name handle. See
> [https://github.com/apache/flink/pull/21050#discussion_r1011061072]
> cc [~roman]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
