gmane.ietf.tools

IETF tools list issues (update)

Hi, an updated list of issues from my POV:

* the RFC-I problem was fixed (= ietf.org unlisted),
  an update sent via GMaNe might be trapped in the
  moderation queue or in a meeting with Dave Null:
  <http://article.gmane.org/gmane.discuss/11839>

* various errata are still AWOL, for a complete list
  of 667 RFCs with reported errata as of 2008-07-08
  see <g502oa$btu$1 <at> ger.gmane.org>, available in
  <http://article.gmane.org/gmane.ietf.tools/1109> 
  <http://www.ietf.org/mail-archive/web/tools-discuss/current/msg01563.html>

* the "challenge" wasn't identified as challenge by
  GMaNe, I'm not sure if that is a bug or a feature
  in practice.  Here is the "raw" challenge message:
  <http://article.gmane.org/gmane.ietf.tools/1114/raw>

  RFC 3834 issues in this "challenge":
  1 - The envelope sender address was the ordinary list
      address, the list bounce address might be better,
      see RFC 3834 chapter 3.3
  2 - There was no OPTIONAL (RFC 3834 chapter 3.1.8)
      Precedence: list (or similar) header field
  3 - There was no RECOMMENDED (3.1.7) Auto-Submitted
      header field 
  4 - There were no RECOMMENDED (3.1.6) References and
      In-Reply-To header fields with the Message-ID of
      the challenged article, IMO that would be good
  5 - There was no OPTIONAL (3.1.5) Auto: subject tag

(Continue reading)

headers

Frank Ellermann | 19 Jul 22:02

Re: IETF tools list issues (update)

> Likely the C/R system is still waiting for a response
> via GMaNe, I'll test that later.

Done - it was still waiting, only for the message that
triggered the original "challenge".  Because I reposted
it later you got that twice - please ignore this dupe:
<http://article.gmane.org/gmane.ietf.tools/1135/raw>

Apparently a later article with the same sender address
was dropped.  This message might make it, because the
sender address is now confirmed.  Or it triggers a new
challenge, because the sender address is unsubscribed.

When a spambot answers any "challenge" the C/R system
is defeated, is it really that simple ?

 Frank

headers

Henrik Levkowetz | 31 Jul 11:39

Re: IETF tools list issues (update)

Hi Frank,

On 2008-07-19 19:47 Frank Ellermann said the following:
...
> * the "challenge" wasn't identified as challenge by
>   GMaNe, I'm not sure if that is a bug or a feature
>   in practice.  Here is the "raw" challenge message:
>   <http://article.gmane.org/gmane.ietf.tools/1114/raw>
> 
>   RFC 3834 issues in this "challenge":
>   1 - The envelope sender address was the ordinary list
>       address, the list bounce address might be better,
>       see RFC 3834 chapter 3.3
>   2 - There was no OPTIONAL (RFC 3834 chapter 3.1.8)
>       Precedence: list (or similar) header field
>   3 - There was no RECOMMENDED (3.1.7) Auto-Submitted
>       header field 
>   4 - There were no RECOMMENDED (3.1.6) References and
>       In-Reply-To header fields with the Message-ID of
>       the challenged article, IMO that would be good
>   5 - There was no OPTIONAL (3.1.5) Auto: subject tag
>   6 - There was no RECOMMENDED (3.1.2) Reply-To header
>       field, that's bad in conjunction with the next
>       points
>   7 - The 2822-From was not a list owner address or say
>       tools-discuss-request <at> ietf.org as RECOMMENDED in
>       3.1.1, it was the list address, IMO that's wrong.

I'll be releasing 2 new versions of postconfirm (the challenge
issuing tool) over the next week or so, and plan to address at

(Continue reading)

headers

Frank Ellermann | 31 Jul 13:41

Re: IETF tools list issues (update)

Henrik Levkowetz wrote:

> I'll be releasing 2 new versions of postconfirm (the challenge
> issuing tool) over the next week or so, and plan to address at
> least issues 3, 4, and 6, and possibly also 1 and 7 above in the
> second of those releases.

Thanks.  This is a rather weird case, GMaNe got three further
challenges, where GMaNe apparently changed its addresses for
unlear reasons unilaterally from local <at> gmane to local <at> m.gmane
style on different IETF lists.

Because I knew the drill I simply confirmed this using the same
route via GMaNe, not via another mail provider, and that worked.

 Frank