[Community-Discuss] Current violations to be investigated by AFRINIC

Owen DeLong owen at delong.com
Fri Dec 21 05:09:56 UTC 2018


Why do you assume that unadvertised automatically means not properly utilized? I don’t know what the situation is with the blocks in question, but there are valid uses for IP addresses that would not be visible in any routing table you can access. 

Owen


> On Dec 20, 2018, at 07:36, Dabu Sifiso <dabu.sifiso at yandex.com> wrote:
> 
> 
> It is very sad to that we are continuing this discussion.
> 
> If people want to really fight, just have a look at the statistics here:
> http://bgp.potaroo.net/ipv4-stats/prefixes_unadv_pool.txt
> 
> There are at least 2 /14s issued 10 and 18 months ago respectively in the list from the same country but to different organizations.
> The oldest one has never been advertised, not even as smaller blocks.
> The newer one is only advertised as a total of /18 /24 (4X /20 and 1X /24)
> 
> Proponents of the review policy say AFRINIC already has the rights and means to do review and de-register.
> 
> By policy and RSA anything in the list from that link that is not currently being announced and was requested for public internet services should simply be de-registered and returned to AFRINIC, no discussion about it.
> AFRINIC doesn't need a review policy and already can de-register any clear violations, this was repeated several times.
> 
> Please AFRINIC, do apply the Policy and RSA as it currently is and review the 1+ millions of IP address from 2017 and 2018  that are in the list of unadvertised prefixes and older than 8 months from Potaroo,
> That is around 140 prefixes to review, just for 2017/2018.
> 
> This is what a small group of people want and say is already possible to be done anyways, let's have it done now no need for new policy.
> Consequences can already be tested.
> 
> Thank you.
> 
> /Dabu
> 
> 
> 
> 
> 
> 20.12.2018, 09:10, "Benjamin Eshun" <benjamin.eshun at gmail.com>:
>> Owen,
>> 
>> The reported class was the most debated part of the proposal and it has evolved significantly. Anything to add as improvement? See new thread on rpd. - https://lists.afrinic.net/pipermail/rpd/2018/008952.html
>> 
>> Review policy is in the purview of the PDP as it falls under  "proper and responsible usage and management of INRs" as defined by bylaws as scope for community policy development under the framework of the PDP.
>> 
>> But it will be interesting to see a review operational plan from staff, exercising their exclusive and discretionary  rights here and see how different  it can be from what is being discussed.
>> 
>> Benjamin
>> 
>> On Dec 20, 2018, at 1:17 AM, Owen DeLong <owen at delong.com> wrote:
>> 
>>>> On Dec 19, 2018, at 13:41 , Komi Elitcha <kmw.elitcha at gmail.com> wrote:
>>>> 
>>>> Owen,
>>>> 
>>>> RSA  reiterates AFRINIC mandate "to fulfill a stewardship role for those number resources allocated to it, for the use of the African Internet community as constituted by its members collectively" and  grants Afrinic the rights to review members anytime and at its discretion and makes obligations on members to cooperate .
>>> 
>>> If you were under the impression that I disagreed with this, you are mistaken.
>>> 
>>>> So RSA does not prescribe or prohibite a way of doing the review.
>>> 
>>> Correct. I leaves it to staff discretion to determine if a complaint warrants any form of review.
>>> 
>>>> This proposal  tries to define how review may be conducted.
>>> 
>>> It goes further than that. If it merely defined how AfriNIC staff would conduct a review, it would be an operational procedure and out of scope for the PDP, frankly.
>>> 
>>> However, it goes further than that and it attempts to set policy on things that can trigger a review. One of those things is a report of potential abuse. It’s quite clear
>>> from the text of the proposal and from the arguments offered by the proponents that this clause is intended to create an environment supportive of witch hunts against
>>> larger resource holders in the region.
>>> 
>>>> The "reported" class has gone through intensive  discussions and improvements. Repeating old  discussion points without putting them into context does not  bring any value.
>>> 
>>> And yet you provide no additional context for it in this message whatsoever.
>>> 
>>> The point is that the potential for abuse of this clause has never yet once been addressed by the proposal authors.
>>> 
>>>> As for the rest of your points below,  they have been discussed and concluded. I am sorry, can't help you more. You need to make time to revisit  the archives.
>>> 
>>> What additional points? The only point I made below was the potential for abuse of the “reported” class of review. Everything below is supporting details and additional context for why I believe that clause would be abused and how the abuse would be costly to a subclass of resource members in the region, thus making the proposed policy discriminatory in addition to its other flaws.
>>> 
>>> Owen
>>> 
>>>> -Komi
>>>> 
>>>>> Le mer. 19 déc. 2018 à 3:21 AM, Owen DeLong <owen at delong.com> a écrit :
>>>>> It is the reported class which is the main sticking point here. Anyone who has worked for a large organization with a complex network will see the obvious opportunity for abuse here.
>>>>> 
>>>>> Without that category in the proposed policy, AfriNIC has the discretion to follow up on a report if the find it credible. The proposed policy would shift that to a burden for AfriNIC to disprove the credibility of any report they declined to investigate.
>>>>> 
>>>>> That’s a major change and one which the proponents have made it quite clear they intend to use as a tool to attack certain resource members. (I’ll note at least one opponent has expressed the possibility of using that same tool against a proponent should this proposal become policy).
>>>>> 
>>>>> When I initially expressed some support for this proposal, either it lacked this provision or I had not fully understood the implications of it. Coming to understand that was a major factor in my switch to strong opposition. The blatant expression of the proponents of their intent to use this as a tool to force AfriNIC to attack certain organizations on their behalf was another major factor.
>>>>> 
>>>>> Now the proponents will claim that an audit is not an attack, but the reality is that absent substantial evidence of wrongdoing, such a review of a network of any significant size and/or complexity is a costly and time-consuming process.
>>>>> 
>>>>> I’ve done these types of reviews for organizations internally as a consultant, so I know first hand what kind of resources and how long they can take. Please believe me when I say that it is non-trivial and that the cost scales non-linearly with the size and complexity of the network.
>>>>> 
>>>>> Owen
>>>>> 
>>>>>> On Dec 18, 2018, at 12:37, Benjamin Eshun <benjamin.eshun at gmail.com> wrote:
>>>>>> 
>>>>>> Jackson,
>>>>>> 
>>>>>> Just a simple question.
>>>>>> 
>>>>>>> On Tue, Dec 18, 2018 at 11:16 AM Jackson Muthili <jacksonmuthi at gmail.com> wrote:
>>>>>>> 
>>>>>>> A policy that incites or stimulates divisiveness, targeting and
>>>>>>> victimizing is very, very bad.
>>>>>>> 
>>>>>> 
>>>>>> How does a proposed policy that gives guidelines to AFRINIC to carry
>>>>>> out audit and review of INR that it already has the mandate to carry
>>>>>> out “very very bad”?
>>>>>> 
>>>>>> For the avoidance of doubt, I have repeated the classes of review in
>>>>>> the proposed policy.
>>>>>> 
>>>>>> **  extract **
>>>>>> 
>>>>>> 13.3 Classes of review: Members to be reviewed shall be selected
>>>>>> according to the following classes:
>>>>>> 
>>>>>> 13.3.1 Random
>>>>>> 
>>>>>> The member is chosen by AFRINIC at random between the membership.
>>>>>> 
>>>>>> 13.3.2 Selected
>>>>>> 
>>>>>> Member is selected because of an internal report or due to a lack of
>>>>>> contact between the AFRINIC and the member.
>>>>>> 
>>>>>> 13.3.3 Reported: Here, members are reviewed either because:
>>>>>> 
>>>>>> A) They have requested the review themselves or
>>>>>> 
>>>>>> B) There has been a community complaint made against them that
>>>>>> warrants investigation. Complaints shall be backed by evidence and
>>>>>> AFRINIC staff shall evaluate the facts as appropriate to conduct the
>>>>>> review. However this review is not applicable to a member  with the
>>>>>> same resources portfolio on which a full review has been completed in
>>>>>> the preceding 24 months.
>>>>>> 
>>>>>> AFRINIC staff may, at its sole discretion, after having assessed the
>>>>>> nature of the evidence found in the community complaint, require that
>>>>>> such evidence be (i) submitted in the form of a sworn affidavit or
>>>>>> (ii) declared to be true before a Commissioner of Oath.
>>>>>> 
>>>>>> *end of extract ***
>>>>>> 
>>>>>> Unless if you want to suggest that AFRINIC should be stripped of its
>>>>>> responsibility to randomly review any memberships, stripped off its
>>>>>> powers to investigate a member that has stopped communicating, and
>>>>>> stripped off its powers to listen to complaints from  the community
>>>>>> and its members in general.
>>>>>> 
>>>>>> I know that we are all noble people and that we have the very best
>>>>>> intentions for the community, but simply put AFRINIC cannot be just
>>>>>> left to the good intensions of the community and members.
>>>>>> 
>>>>>> Benjamin
>>>>>> 
>>>>>> _______________________________________________
>>>>>> Community-Discuss mailing list
>>>>>> Community-Discuss at afrinic.net
>>>>>> https://lists.afrinic.net/mailman/listinfo/community-discuss
>>>>> 
>>>>> _______________________________________________
>>>>> Community-Discuss mailing list
>>>>> Community-Discuss at afrinic.net
>>>>> https://lists.afrinic.net/mailman/listinfo/community-discuss
>> ,
>> 
>> _______________________________________________
>> Community-Discuss mailing list
>> Community-Discuss at afrinic.net
>> https://lists.afrinic.net/mailman/listinfo/community-discuss
> 
> _______________________________________________
> Community-Discuss mailing list
> Community-Discuss at afrinic.net
> https://lists.afrinic.net/mailman/listinfo/community-discuss




More information about the Community-Discuss mailing list