On 2019 Mar 16 10:50:00, you wrote to Nick Andre:

 TL> But nodelist data maintenance is still concentrated in the hands of NCs,
 TL> and aggregated by RCs and ZCs.

since they are tasked by policy for this job, this is as it should be... they
are responsible for their segments...

 TL> Some of these roles could be replaced by a web based front end, where
 TL> individual sysops actually maintain their own nodelist entries.

this is already possible and has been since the early days... no web thing
needed... it just takes operators that know and understand the format... the
worst part is knowing what to use when and where but that's easily sussed by
reading the documentation of the nodelist format...

at one time, my makenl was configured to allow operators to submit their own
individual nodelist entries... it worked quite well when the proper format and
data was submitted...

 TL> The interface would validate all fields, so that the nodelist data at
 TL> least conformed to standards (of course, there's still a degree of
 TL> "GIGO", but that goes for any human entered content ;) ).

already done and handled since eons... just no web interface where you have to
employ more security than necessary to prevent bots and humans from attacking,
changing others' entries, submitting invalid/false data, etc... a NC/HUB should
know who is in their segment and not just rubber stamp what is sent to them for
processing... the RCs/ZCs have to have some trust in their NCs but they should
also still check the segments they generate to send upstream...

 NA>> The MakeNL source I believe is open, but the rest is all custom to
 NA>> this system, just as ZC2's software is specific for him, ZC3 and ZC4
 NA>> etc.

 TL> Yes, I have the MakeNL source here that I compiled on my Pi.  I use
 TL> MakeNL on the Pi and x86_64 to generate my VLRadio nodelists.  Further
 TL> processing by another script generates my DNS zone file.  Along the
 TL> way, scripts hatch the nodelist and rebuild and hatch the infopack
 TL> every week, including the latest nodelist.  For various reasons, I
 TL> require 100% automation to keep everything up to date.

how do you know if/when invalid data gets in? i'm speaking of data that passes
the tests but is still invalid/incorrect...

)\/(ark

Always Mount a Scratch Monkey
Do you manage your own servers? If you are not running an IDS/IPS yer doin' it
wrong...
... A single fact can ruin a good argument everytime!
---