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

[rpd] Last Call - RPKI ROAs for Unallocated and Unassigned AFRINIC Address Space AFPUB-2019-GEN-006-DRAFT03.

Paschal Ochang pascosoft at gmail.com
Tue Jun 15 15:56:00 UTC 2021


Hello, Find my comments inline.

On Tuesday, June 15, 2021, Noah <noah at neo.co.tz> wrote:


> Hi Job,

>

> On Mon, Jun 14, 2021 at 2:12 PM Job Snijders <job at fastly.com> wrote:

>

>> On Sat, Jun 12, 2021 at 09:01:03PM +0300, Noah wrote:

>> > While this was resolved within 10mins, were there measures put in place

>> by

>> > the NCC as mitigation process to curb a repeat of the incident?

>>

>> This problem lasted more than 70 minutes!

>>

>> The point I am trying to make: once an RIR automatically creates AS0

>> ROAs for unallocated/unassigned space, any database incidents where the

>> registration status of a resource accidentally (and temporarily) lapses,

>> can result in blocklisting via AS0.

>>

>> The risk of things going wrong at a future point in time versus the

>> perceived benefits of such ROAs don't align well.

>>

>> The best measure to put in place to curb incidents is to not implement

>> this type of policy.

>>

>

> There are several of us who support this policy for valid reasons, some of

> which are founded on the premise [1] that you shared with the APNIC region

> back in 2019.

>

>

> [1] https://mailman.apnic.net/mailing-lists/sig-policy/

> archive/2019/08/msg00065.html

>

>

> And if you look at the bogons [2] currently seen in the global routing

> table, one can easily find the prefixes from AFRINIC pool, some being

> advertised by unassigned ASNs.

>

>

> [2] http://thyme.rand.apnic.net/current/data-add-IANA

>

>

> It is unfortunate that the problem statement of the proposal is not well

> defined to allow a better understanding of the problem being addressed and

> the assessment of the benefits vs risks.

>


Well if the problem statement of the proposal is not Well defined as you
said, then the proposal is not well drafted and that's a problem. In
research if your problem statement is not well defined your research
justification is also not tenable.



> Yes.....Things can go wrong at many layers and at different levels,

> however, the entitled parties shall play their role and we all learn

> from incidents and keep improving.

>

Well there are some lessons you don't wonna experience or learn from you
might come out too charred to survive.


>

> The issuance of AS0 ROAs should follow the conservative approach RIRs use

> in managing changes in number ressources status, the addition of INRs to

> bogon list, subject to AS0 ROAs.

>

>

> In my humble opinion.

>

>

> Cheers,

>

> Noah

>

>



--
Kind regards,

Paschal.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20210615/9654e4ad/attachment-0001.html>


More information about the RPD mailing list