Jia Fan

I'm not sure about we need `authority management`, `task alert` cause seatunnel is a framework.

Please provide your environment, like Spark deployment mode (yarn or standalone), and your submit shell command.

And please make sure your SeaTunenl package permission is can be access with your start shell command's user.

> cluster mode, why use shell to tar resource ? Don't be so different between versions. Don't refactor stable modules and focus on other directions Compress resource is in Java...

> > > cluster mode, why use shell to tar resource ? Don't be so different between versions. Don't refactor stable modules and focus on other directions > > >...

> I like to work on this issue. I think these links [source of tunnel](https://seatunnel.apache.org/docs/2.1.2/category/source) & [supported links](https://seatunnel.apache.org/docs/2.1.2/category/sink) should be replaced. Great, I assgined to you

Maybe you can create a new page about all transform like this: https://seatunnel.apache.org/docs/2.1.2/category/source. Or just use transform common options page.

> The links are correct in github just it is not pointed to specific file. Instead of that it is pointed to folder and here https://seatunnel.apache.org/docs/2.1.2/category/source path has category folder...

You can use `customUpdateStmt ` in seatunnel spark jdbc sink to fulfill this requirement. Check this https://seatunnel.apache.org/docs/2.1.2/connector/sink/Jdbc#customupdatestmt-string

> @Hisoka-X Can we split file sink into `HdfsFileSink`, `LocalFileSink` and `S3FileSink` or other? I implemented `HdfsFileSink` and `LocalFileSink`, and I hope some people can implement `S3FileSink`. Got it