Search RPD Archives
Limit search to: Subject & Body Subject Author
Sort by:

[rpd] Last Call for "AFPUB-2016-GEN-001-DRAFT-04 - Internet Number Resources Review by AFRINIC"

David Hilario d.hilario at outsideheaven.com
Sat Jul 1 00:38:32 UTC 2017


On Jul 1, 2017 1:49 AM, "ALAIN AINA" <aalain at trstech.net> wrote:


> On 29 Jun 2017, at 08:32, Nishal Goburdhan <nishal at controlfreak.co.za>
wrote:
>
>
>> On 29 Jun 2017, at 09:22, ALAIN AINA <aalain at trstech.net> wrote:
>>> On 28 Jun 2017, at 07:04, Nishal Goburdhan <nishal at controlfreak.co.za>
wrote:
>>> On 27 Jun 2017, at 18:56, Tutu Ngcaba <pan.afrikhan at gmail.com> wrote:
>>>> On 27 Jun 2017 6:10 p.m., "Nishal Goburdhan" <nishal at controlfreak.co.za>
wrote:
>>>>
>>>> or,  said earlier, audit everyone.  equally.  no member is more
special than any other.
>>>> and small members are just as guilty of resource abuse as large ones.
 something that's been said before but the author's haven't ack'd.    :-/
>>>>
>>>> Woooza  bra,
>>>>
>>>> but this must be the case for sure. no one shall be special at all and
for i know its review for all who it applied at the time and shall be for
all the people who the afrinic given ip address from the big one to the
small one. I agree like this.
>>>
>>> i'm glad you agree.
>>> except, that's not what the policy text reads.
>>>
>>> i don't have a problem with audits;  but do it across the board (if at
all).
>>> -then- determine, if, resource wise, this is useful, or not.
>>
>>
>> The proposal  defines 3 classes:
>>
>> - random : limited to certain level of membership
>> - reported
>>
>> But  also
>> ***
>> 13.3.2 Selected
>> A member is selected because of an internal report or due to a lack of
contact between the AFRINIC and the member.
>> *****
>> This class allows the review of  any member which according to AFRINIC
does not comply
>
>
> thanks, i saw that.
> however, i did not understand the reason for "- random : limited to
certain level of membership”

In case of non-compliance, higher impact.


This shows that this proposal is not impartial, but targeted and not a
review policy but large LIR resource reclamation policy proposal.

> when you could simply write:
> “All resource members would be subjected to an audit, at least once every
Z period”

By default all ressources are subjected  to an audit as specified by  RSA
and the allocation/assignment policies.

The review policy proposal defines how AFRINIC could best conduct these
reviews:

- Randomly select ressources
- Attend to reported  ressources
- Select ressources  for internal reasons

All at staff discretion...


And this is where danger lives.



>
> i think my text is cleaner, but i might be biased  :-)
>
> (and to give you some context, my bias would also ask afrinic to start
with the resources referenced by the AIRRS report, if that’s still running!)

I don't know if AIRRS  is still running, but we do have a similar tool
here: http://196.200.57.137/africa-bgp/

AFRINIC could this tool or any other to feed  the   "selected  class” and
start immediately


Previous proposals for the de-registration of unannounced space failed and
had long list of reasons for being rejected.

I agree that space could be "better" used sometimes, but that is highly
debatable and personal as what it means, but there can be many reasons for
an LIR to have unannounced space.

In any case, these LIRs cannot ask for more space as long as it is like
that, so there is no real problem.

With the transfer policy, hunting down unannounced space would be rather
dangerous, you would literally reclaim what will be perceived as assets in
the various finance departments of the companies, "assets" from these
organisations will not be allowed to be stripped away just like that

Just imagine the scenario, you ask to transfer away your resources as they
are unannounced and you do not need them.

Transfer is rejected, and the same person rejecting your transfer starts a
review at his own discretion...

The conflict of interest for AFRINIC would be far too great.

This proposal is a reclamation from large LIRs policy proposal, not a
review proposal intended to further improve the registry's data accuracy
and educate the members on best practice.

It will also jeopardise the incoming IPv4 transfers, potentially scaring
people from actually doing the right thing and having the space registered
correctly.

As such I still object to it.

Regards,
David Hilario


—Alain

>
>
>
>>>> if small have idle they must return it. if big have idle they must
also returned it. they the afrinic shall give the one who also need also
instead of big or small keeping idle for so so long.
>>>
>>> that's not going to happen;   the "transfer" policy removed the
incentive to do the "Right Thing”.
>>
>>
>> The intra-RIR transfer policy allows a ressource holder to transfer idle
space to meet another member IPv4 need  approved by AFRINIC. Otherwise  you
should return idle  to AFRINIC or be forced by an review.
>
>
> and again, i ask - why would the resource holder not simply have returned
this to afrinic (which, is what tutu thought would be the case, and, what
should have been the natural terms of the *contract* (aka RSA) under which
these resources were issued), and have the requestor go through the
otherwise standardised "needs based process" at afrinic (which, would be a
lot less work).
>

> i stand by my initial assertion above, even though i recognise that it’s
not worth debating this any more.
> i was simply telling tutu that what he thought would happen, would not be
the case.
>
> —n.


_______________________________________________
RPD mailing list
RPD at afrinic.net
https://lists.afrinic.net/mailman/listinfo/rpd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20170701/6129536c/attachment.html>


More information about the RPD mailing list