[
https://issues.apache.org/jira/browse/FLINK-30845?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gyula Fora reassigned FLINK-30845:
----------------------------------
Assignee: Fabio Wanner
> Params in jarURI end up in file name
> ------------------------------------
>
> Key: FLINK-30845
> URL: https://issues.apache.org/jira/browse/FLINK-30845
> Project: Flink
> Issue Type: Bug
> Components: Kubernetes Operator
> Affects Versions: kubernetes-operator-1.3.0
> Reporter: Fabio Wanner
> Assignee: Fabio Wanner
> Priority: Minor
> Labels: pull-request-available, starter
>
> *Context*
> Jar files for jobs are submitted to the operator by supplying a URI to the
> .jar file. This URI can be a file system path or a URI to some HTTP resource.
> If a HTTP URI is given, the file will be fetched using the
> {{{}HttpArtifactFetcher{}}}.
> There are cases where the supplied URI will contain additional params. For
> example if pre-signed S3 URLs are used.
> Example:
> {code:java}
> https://some-domain.example.com/some.jar?some=params{code}
> *Problem*
> When the HttpArtifactFetcher determines the name of the .jar file it does
> also use the params as part of the file name. In the example from above the
> resulting file name would be: {{some.jar?some=params}}
> Submitting this job to Flink will result in an error as it will be checked
> for the file name to end with {{.jar}}
> *Possible Solution*
> In the {{HttpArtifactFetcher}} it would be enough to replace:
> {code:java}
> String fileName = FilenameUtils.getName(url.getFile());{code}
> with
> {code:java}
> String fileName = FilenameUtils.getName(url.getPath());{code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
