WebDAV client RFC4437 Redirection vs other Hosts

Hello,
my name is Silvio, I'm working on a research institute in Italy.
For our experiment we use WebDAV endpoint that aggregate multiple WebDAV storage in a single namespace. Would be nice use WinSCP to access to those data, however I see that WinSCP does not follow redirection, maybe because the system use to redirect the client to endpoints which are different from the starting one.
Would be possible change this behaviour, allowing the client to follow redirection even to different endpoint as browser does? Maybe the user can select it as an option, if this approach is considered bad from a security point of view.
This possibility should be allowed by RFC4437.
Thank you for your attention and congratulation for your nice client.
Best Regards,
Silvio

Thanks for your post.
So does it mean that your server redirects different folders to different servers?

Something like that. The typical storage system work as follow, there is one frontend server (head node) that store the full namespace of the filesystem, then when I try to access to access to a specific file, the head node redirect me to one of the disk server (disk node) that store really the file.
For example in my case I logged with WinSCP on the machine dpm-head.exp.infn.it using WebDAV via x509 certificate, and I can browse the filesystem properly, then when I try to download a file I receive the following error:

Error in file transfer
Resource WebDAV moved in 'http://dpm-disk03.exp.infn.it:9001/SE02e/belle/2018-07-21/ciao026.txt.27390045.0?token=kSMC9ykzdnIMhYrbSZClU7PRVcc%3D%401532678152%400&dav_sfn=%2Fdpm%2Fna.infn.it%2Fhome%2Fbelle%2FTMP%2Fbelle%2Fuser%2Fspardi

The browser can work properly with this kind of redirection.
Thank you again!
Regards
Silvio

OK, I see.

This issue has been added to the tracker:
Issue 1667 – Support redirected WebDAV downloads (even to other hosts)

Great! Thank you so much!!!
Best Regards
Silvio

I thumbs up'ed this issue yesterday :)
My use case is as follows:

• central (WebDAV) system in AWS for incoming data (large media files)
  
• after metadata extracted and thumbnails created, we move the original data to an S3 bucket
  
• the system will pull data from S3 back to a local cache to serve GET requests, if needed
  
• to keep the local cache small, we would rather send the client a redirect for GET and download the file with a (presigned) AWS URL.
  
• it works with a few other WebDAV clients, but WinSCP is my favorite, so I would love it if this worked here too :)
  

@turchinc: OK, thanks for sharing your use case. I'll contact you, once/if I have anything to test.

@turchinc: I have sent you an email asking for a session log.