[atlas] Proposal: Measure well-known CDNs,[CDN-HTTP]

Hi RIPE Atlas users,

I stumbled upon this thread and I wanted to chime in. I work for Cloudflare
and we of course use a wide set of external visibility and performance
management tools to track our visibility.

But on occasions I also use the RIPE Atlas network to perform some specific
reachability tests. I believe the distribution of probes and the fact that
they're als in eyeball ASN is a really valuable factor that RIPE Atlas has.

At the moment I'm running into a limitation around HTTP tests. I would like
to use the RIPE Atlas network to curl some pages to retrieve and evaluate
the HTTP headers. This is currently not possible with Atlas as it is only
allowed to probe Anchors for safety reasons, which I fully understand and
support (you don't want to turn Atlas into a botnet).

The HTTP targets that I would like to probe would be all on our IP space,
so I don't see a direct risk from the outbound part of the request. I can
see there is some risk with the return part, where the content from an
"unknown" system gets returned to a probe in someone's private network,
with possible ramifications.

So I was pondering about two possible options that can potentially help me,
but not risk the Atlas project or demonstrate favoritism towards particular
companies.


1) One of the options would be to open up some of our IP space to the
probes for HTTP probes.
2) The other option would be for us to host an Anchor and put our CDN in
front of it. That way RIPE has complete control over the returned data.

I would love to hear the general consensus on such approach and the
viability of it

Best Regards,

-- 

Dirk-Jan van Helmond


On 17/12/2022 00:19, Barry Raveendran Greene wrote:
>
>
>> On Dec 16, 2022, at 23:29, Stephane Bortzmeyer <bortzmeyer at nic.fr>
wrote:
>>
>> There is a larger problem here, a more strategic one: such a feature
>> would contribute to the centralisation of the Internet, which is
>> already too important. Tagging some targets are "important" and
>> "worthy of measurements" would mean that we consider some HTTP servers
>> to be more useful than others. That would be a bad message from RIPE.
>
> We’ve come full circle - we started with centralized PTTs - moved to a
decentralized ASN/Paul Baran model - now Re-centralized based on marketing
domination.
>
> +1 with Stephane’s observation. The selection of who to measure is a
statement.

+1

Also, while the data would be useful, I don't think the role of the ripe
ncc is to grade commercial services. Let other companies or individual
researchers do that.

-- 

Dirk-Jan van Helmond
Leader Solutions Engineering EMEA
mobile: +31611645622
email: dirkjan at cloudflare.com
www.cloudflare.com
<https://www.cloudflare.com/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ripe.net/ripe/mail/archives/ripe-atlas/attachments/20221220/3f0e0739/attachment.html>