[
https://issues.apache.org/jira/browse/FLINK-26484?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17504128#comment-17504128
]
Matthias Pohl edited comment on FLINK-26484 at 3/10/22, 9:30 AM:
-----------------------------------------------------------------
Further discussion happened in the PR. No need to backport this change because
we don't do retries in older versions of Flink.
master: 4907f5fb7033b1f918c4e721a1ccf95878f980a2
was (Author: mapohl):
master: 4907f5fb7033b1f918c4e721a1ccf95878f980a2
> FileSystem.delete is not implemented consistently
> -------------------------------------------------
>
> Key: FLINK-26484
> URL: https://issues.apache.org/jira/browse/FLINK-26484
> Project: Flink
> Issue Type: Bug
> Components: Connectors / FileSystem, Runtime / Coordination
> Affects Versions: 1.15.0
> Reporter: Matthias Pohl
> Assignee: Matthias Pohl
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.15.0
>
>
> The BlobServer cleanup does not work for the Presto S3 filesystem in case of
> failure due to some bug in the recursive delete implementation. The {{false}}
> return value is not processed which leads to an error case being "swallowed",
> i.e. recursive cleanups do not work in this case (see
> [PrestoS3FileSystem:496|https://github.com/prestodb/presto/blob/master/presto-hive/src/main/java/com/facebook/presto/hive/s3/PrestoS3FileSystem.java#L496])
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
