[
https://issues.apache.org/jira/browse/FLINK-10963?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16701724#comment-16701724
]
ASF GitHub Bot commented on FLINK-10963:
----------------------------------------
igalshilman commented on a change in pull request #7161:
[FLINK-10963][fs-connector, s3] Cleanup tmp S3 objects uploaded as backups of
in-progress files.
URL: https://github.com/apache/flink/pull/7161#discussion_r237008022
##########
File path:
flink-core/src/main/java/org/apache/flink/core/fs/RecoverableWriter.java
##########
@@ -121,6 +121,32 @@
*/
RecoverableFsDataOutputStream recover(ResumeRecoverable resumable)
throws IOException;
+ /**
+ * Marks if the writer requires to do any additional cleanup/freeing of
resources occupied
+ * as part of a {@link ResumeRecoverable}, e.g. temporarily files
created or objects uploaded
+ * to external systems.
+ *
+ * <p>In case cleanup is required, then {@link
#cleanupRecoverableState(ResumeRecoverable)} should
+ * be called.
+ *
+ * @return {@code true} if cleanup is required, {@code false} otherwise.
+ */
+ boolean requiresCleanupOfRecoverableState();
+
+ /**
+ * Frees up any resources that were were previously occupied in order
to be able to
+ * recover from a (potential) failure. This can be temporary files that
we written or
+ * objects that were uploaded (e.g. S3).
+ *
+ * <p><b>NOTE:</b> This operation should not through an exception if
the resumable has already
+ * been cleaned up and the resources have been freed.
+ *
+ * @param resumable The {@link ResumeRecoverable} whose state we want
to clean-up.
+ * @return {@code true} if the resources were successfully freed,
{@code false} otherwise
+ * (e.g. the file to be deleted was not there).
+ */
+ boolean cleanupRecoverableState(ResumeRecoverable resumable) throws
IOException;
Review comment:
Would it make sense to have just this method (and omit
`requiresCleanupOfRecoverableState()`) with a default no-op implementation?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Cleanup small objects uploaded to S3 as independent objects
> -----------------------------------------------------------
>
> Key: FLINK-10963
> URL: https://issues.apache.org/jira/browse/FLINK-10963
> Project: Flink
> Issue Type: Sub-task
> Components: filesystem-connector
> Affects Versions: 1.7.0
> Reporter: Kostas Kloudas
> Assignee: Kostas Kloudas
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.7.1
>
>
> The S3 {{RecoverableWriter}} uses the Multipart Upload (MPU) Feature of S3 in
> order to upload the different part files. This means that a large part is
> split in chunks of at least 5MB which are uploaded independently, whenever
> each one of them is ready.
> This 5MB minimum size requires special handling of parts that are less than
> 5MB when a checkpoint barrier arrives. These small files are uploaded as
> independent objects (not associated with an active MPU). This way, when Flink
> needs to restore, it simply downloads them and resumes writing to them.
> These small objects are currently not cleaned up, thus leading to wasted
> space on S3.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
