Multiple signatures to create a reference to an irt object

  Hi Alex!

>So far, the response to my query has been, well, nil :-( Maybe I'm asking
>this question on the wrong list?  This must have been tested before...
>
>Cheers,
>Alex

  Well, I guess the answers are:
  NO (wrong list) and
  WELL... (tested for all situations)
  
  Do you want it on the WG agenda for next week?
  Wilfried.

 ___________ SWITCH - The Swiss Academic and Research Network ___________
 Alexander Gall,  SWITCH,  Limmatquai 138,  CH-8001 Zurich,  Switzerland
 gall at switch.ch         Tel: +41 1 268 1522          Fax: +41 1 268 1568



> Hello
> 
> I've been playing around with PGP authentication and irt objects in the 
> test database and ran into the following problem.
> 
> The relevant objects are
> 
> mntner:       SWITCH-MNT
> irt:          IRT-SWITCH
> inetnum:      130.59.0.0 - 130.59.255.255
> key-cert:     PGPKEY-C3BA4795
> key-cert:     PGPKEY-82146071
> 
> They are all protected by SWITCH-MNT, which has a single auth attribute 
> pointing to PGPKEY-C3BA4795. Updates signed with this key work fine.
> 
> IRT-SWITCH has the attribute auth: PGPKEY-82146071.
> 
> What I would like to do is to add mnt-irt: IRT-SWITCH to the inetnum
> object. If I understood correctly, I have to sign that update with two keys:
> with key C3BA4795 because the inetnum is protected by SWITCH-MNT and with
> 82146071 because a new reference to an irt object needs to be signed by 
> the key referenced in the irt's auth attribute.
> 
> The question is, which MIME message sent to test-dbm at ripe.net 
> does this for me? 
> 
> My interpretation of the (rather brief) section "3.3.2 PGP support" in the
> handbook is that I need to create a MIME message with nested signatures. 
> So, I created such a beast by hand because my mailer can't do that (see 
> first attachment). Apparently, the robot checks the outer signature but 
> does not recognize the inner multipart/signed content-type (see second 
> attachment).
> 
> Unless my MIME encoding is wrong (which may well be the case :-) I must 
> have misunderstood the mechanism.
> 
> Any help is appreciated.
> --
> Alex
>  ___________ SWITCH - The Swiss Academic and Research Network ___________
>  Alexander Gall,  SWITCH,  Limmatquai 138,  CH-8001 Zurich,  Switzerland
>  gall at switch.ch         Tel: +41 1 268 1522          Fax: +41 1 268 1568

--------------------------------------------------------------------------------