Re: [Sidrops] trying to limit RP processing variability

[Job Snijders <[e-mail]>]

On Thu, Apr 16, 2020, at 15:30, Randy Bush wrote:
> >>> An attacker who can delete files can as easily replace them with
> >>> something else with the same result.  
> >> not really.
> > I cannot think of a case where an attack can manipulate the rsync
> > exchange, impersonate the rsync server, or manipulate the file system
> > of the legitimate rsync server in such a way that they can signal
> > to an rsync client to delete existing files but not to replace them
> > partially or completely. 
> > 
> > What am I missing?
> 
> 6486 4.2
> 
>    fileList:
>       This field is a sequence of FileAndHash objects.  There is one
>       FileAndHash entry for each currently valid signed object that has
>       been published by the authority (at this publication point).  Each
>       FileAndHash is an ordered pair consisting of the name of the file
>       in the repository publication point (directory) that contains the
>       object in question and a hash of the file's contents.
>                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

I think I see an instance of miscommunication:

I think Martin's perspective is that whether the files are deleted or modified by MITM, the RP outcome is the same: the PP has to be tossed. I don't think Martin suggested that files can trivially be replaced, just that deletion or tampering have the same effect. 

Right?

Kind regards,

Job