Re: [Duplicity-talk] Anyone using duplicity with box.net (webdavs)?

[edgar . soldin]

Just got fresh, because i read that the listbody is not mandatory. 
http://www.ietf.org/rfc/rfc4918.txt
"  A client may choose not to submit a request body.  An empty PROPFIND
   request body MUST be treated as if it were an 'allprop' request.  "
Works for me with box.net and german mediencenter.telekom.de .

Interestingly the automatic creation of the backup folder on the first access 
leads to a

  File "/usr/lib/python2.6/httplib.py", line 960, in getresponse
    raise ResponseNotReady()
ResponseNotReady

a second run succeeds then. my guess it that the backend shouldn't fail here 
but retry because obviously the folder was created successfully.

@Ken & all: Could you please check how your backends react on empty listbodies. 
My take here is, if it is empty there is less to misinterpret for each and 
every implementation.

..ede/duply.net

On 01.03.2012 17:37, Kenneth Loafman wrote:
> No objections here.  I'll test against rsync.net <http://rsync.net> and if 
> that works too, we have a fix.
> 
> ...Ken
> 
> On Thu, Mar 1, 2012 at 10:29 AM, Edgar Soldin <address@hidden 
> <mailto:address@hidden>> wrote:
> 
>     ken, any objections trying it without allprop? ..ede
> 
>     -------- Original Message --------
>     Subject: Re: [Duplicity-talk] Anyone using duplicity with box.net 
> <http://box.net> (webdavs)?
>     Date: Wed, 29 Feb 2012 16:35:47 +0100
>     From: address@hidden <mailto:address@hidden>
>     Reply-To: Discussion of the backup program duplicity <address@hidden 
> <mailto:address@hidden>>
>     To: Discussion of the backup program duplicity <address@hidden 
> <mailto:address@hidden>>
> 
>     On 29.02.2012 15:52, Kenneth Loafman wrote:
>     > On Wed, Feb 29, 2012 at 8:25 AM, SanskritFritz <address@hidden 
> <mailto:address@hidden> <mailto:address@hidden <mailto:address@hidden>>> 
> wrote:
>     >
>     >     2012/2/29 Kenneth Loafman <address@hidden <mailto:address@hidden> 
> <mailto:address@hidden <mailto:address@hidden>>>:
>     >     > Too bad there is not a client for Webdav like NcFTP that has gone 
> to this
>     >     > trouble already.
>     >
>     >     Well, there are davfs [1] working and wdfs [2] not working whith
>     >     box.net <http://box.net> <http://box.net>, I just tested it.
>     >
>     >     [1] http://savannah.nongnu.org/projects/davfs2
>     >     [2] http://noedler.de/projekte/wdfs/
>     >
>     >
>     > [1] may be the route to go.  We could use the FTP backend as a model.  
> I'm not inclined to fight issues related to non-standard implementations of 
> the servers, thus NcFTP was chosen, so we'll need a volunteer to tackle this 
> problem.
>     >
> 
>     disagreed, these are fully fledged fuse based filesystems, which can 
> easily be set up today and used via file:// backend already.
> 
>     the guy who found the solution for box.net <http://box.net> claims that 
> it chokes on <D:allprop/> .. if that's the case we could easily set up a 
> --webdav-options switch that allows switching it off.
>     on the other hand, we currently don't need it anyway (maybe for the 
> enhanced upload plausability) but as far as i can see is <D:allprop/> not 
> really necessary as servers should at least return the list of files (urls) 
> and more properties at will, if no <D:allprop/> was defined.
> 
>     shouldn't we try the route without <D:allprop/> first?
> 
>     ..ede
> 
>     _______________________________________________
>     Duplicity-talk mailing list
>     address@hidden <mailto:address@hidden>
>     https://lists.nongnu.org/mailman/listinfo/duplicity-talk
> 
>