One of these days I am going to take this old code from nntpchan to map the rocksolid groups and their distribution:
https://github.com/nilesr/nntpchan-mapper/blob/master/
Seeing all these different paths that messages take while playing with bbx really made me curious how far this stretches by now...

--
Posted on def4

trw wrote:

> One of these days I am going to take this old code from nntpchan to map the rocksolid groups and their distribution:
> https://github.com/nilesr/nntpchan-mapper/blob/master/
> Seeing all these different paths that messages take while playing with bbx really made me curious how far this stretches by now...

The link does not work for me (404).

Is this a graphical representation of paths? Some have done this (for usenet in general) and the data is gathered from the Path: header. Someone made some impressive graphics for this a few months ago, but I can't find it right now.

I would be curious to know who does carry rocksolid.* I can pull some headers and make a list, I'll just write a simple script to do that. Not right away, but I also find this interesting.

--
Posted on Rocksolid Light
news.novabbs.org

Retro Guy wrote:

> I would be curious to know who does carry rocksolid.* I can pull some headers and make a list, I'll just write a simple script to do that. Not right away, but I also find this interesting.

Done, lol :)

Surprised to see supernews.com in there. Mostly just our nodes of course. I can send you the output if you like (it's just a file full of Path: headers.

Retro Guy

--
Posted on Rocksolid Light
news.novabbs.org

This is the working link:
https://github.com/nilesr/nntpchan-mapper/
And yes, it is just graphs made from the path infos of each message.
If you have something else (or also the aggregated paths data), let's have a look at it.

--
Posted on def4

Yes, I would like to see the results, and also the script.
If small enough you could just attach an archive to a message, or otherwise use boxs:
http://ana5icnp3au4qy5i.onion
http://boxs.i2p

cheers

trw

--
Posted on def4

Here a raw readout of the results (attached). These are the first host hit in posting. Of course almost exclusively our nodes which is no surprise. Lots of duplication, it's just a long list. If this info is useable, I can format it better than this, include full path, whatever helps:

Retro Guy

Ok, in that data I find 20 different nodes where posts have been inserted. These:

alphanet.ch
def
def2
def3
def5
i2pn2.org
neodomea5yrhcabc.onion
networks
newnode
news.neodome.net
news.novabbs.com
news.rocksolidbbs.com
nntp.supernews.com
retrobbs.novabbs.com
retrobbs2
rocksolid0
rocksolid2
rocksolid3
rocksolidbbs
somenode2

Mind you, some of those do not exists anymore (def, def2, def3 for example).

When I do
find /var/spool/news/articles/rocksolid/ -type f ! -wholename "*feeds*" -exec grep 'Path:' {} \; >> rs_raw_paths.txt
on def5 (my only new server), I get after some refining to 98 different nodes that have been posting or forwarding articles (see attached file).

Next I would be interested to look at the connections and generate a graph.

--
Posted on def4

Server message: cut out attachment carrying_nodes.txt, not a recognized image (allowed: jpg, gif, png), maximum size= bytes

damn, I have to work on the attachments some more. I will just paste the results here below:

10.128.12.155
10.13.0.7
127.117.190.215
127.140.234.14
127.147.52.208
127.190.160.170
127.215.126.48
127.223.110.70
127.76.124.96
127.99.134.157
192.168.0.80
192.241.184.77
3.eu.feeder.erje.net
4.us.feeder.erje.net
alphanet.ch
border1.nntp.dca1.giganews.com
buffer1.nntp.dca1.giganews.com
buffer2.nntp.dca1.giganews.com
d14
def
def2
def2.lan
def3
def5
defi2p2
desktop14.dt
feeder2.usenet.farm
feeder3.usenet.farm
feeder6.news.weretis.net
feeder7.news.weretis.net
feeder8.news.weretis.net
feeder.erje.net
feeder.usenetexpress.com
feed.usenet.farm
i2pn2.org
i2pn.org
i2pwall.i2p
inn
ip68-2-45-86.ph.ph.cox.net
kevin
leaf.retrobbs.ddns.net
local
localhost
mmn
n8
neodomea5yrhcabc.onion
networks
newnode
news.alphanet.ch
news.alt.net
news.dizum.net
news.etla.org
newsfeed9.news.xs4all.nl
newsfeed.xs4all.nl
news.mixmin.net
news.neodome.net
news.novabbs.com
news.rocksolidbbs.com
news.samoylyk.net
news.supernews.com
news.unit0.net
news.usenet.farm
news.uzoreto.com
nntp.giganews.com
nntprss
nntp.supernews.com
novabbs
novabbs.com
novabbs-internal
paganini.bofh.team
posting.uzoreto.com
replybot
retrobbs
retrobbs2
retrobbs.novabbs.com
retrobbs.rocksolidbbs.com
retrobbs.synchro.net
rocksolid0
rocksolid2
rocksolid3
rocksolidbackup
rocksolidbbs
rocksolidbbs.com
rocksolidnntp1
rocksolid.rslight.i2p
rs
rs4
sewer
somenode2
sono
testnetwork
testvm
tr1.eu1.usenetexpress.com
tr3.eu1.usenetexpress.com
tr3.iad1.usenetexpress.com
v2
weretis.net

--
Posted on def4

On Mon, 21 Dec 2020 13:04:55+0000
trw <trw@anon.com> wrote:

> When I do
> find /var/spool/news/articles/rocksolid/ -type f ! -wholename
> "*feeds*" -exec grep 'Path:' {} \; >> rs_raw_paths.txt on def5 (my
> only new server), I get after some refining to 98 different nodes
> that have been posting or forwarding articles (see attached file).

I did the same also (just using php instead) and see all the 'hops'.

Some peers may not carry rocksolid, but forward the messages anyway.
Not sure if it can be determined that way who carries the groups and
who does not other than the first (last) in the line that shows where
posted.

I'll produce the same as you did from rocksolid2 (and run uniq this
time) and post the results.
>
> Next I would be interested to look at the connections and generate a
> graph.

I've seen it done and it's quite interesting. Looks complicated, lol.

Here's what I get:

3.eu.feeder.erje.net
4.us.feeder.erje.net
Path: rocksolid2
alphanet.ch
border1.nntp.dca1.giganews.com
buffer1.nntp.dca1.giganews.com
buffer2.nntp.dca1.giganews.com
def2
def3
def5
feed.usenet.farm
feeder.erje.net
feeder.usenetexpress.com
feeder2.usenet.farm
feeder3.usenet.farm
feeder6.news.weretis.net
feeder7.news.weretis.net
feeder8.news.weretis.net
i2pn.org
i2pn2.org
neodomea5yrhcabc.onion
news.alphanet.ch
news.alt.net
news.dizum.net
news.etla.org
news.mixmin.net
news.neodome.net
news.novabbs.com
news.samoylyk.net
news.unit0.net
news.usenet.farm^M
news.uzoreto.com
newsfeed.xs4all.nl
newsfeed9.news.xs4all.nl
nntp.giganews.com
nntp.supernews.com
paganini.bofh.team
posting.uzoreto.com
retrobbs.rocksolidbbs.com
rocksolid0
rocksolid2
rocksolid3
rocksolidbackup
rs
sewer
testvm
tr1.eu1.usenetexpress.com
tr3.eu1.usenetexpress.com
tr3.iad1.usenetexpress.com
weretis.net

Retro Guy <retro_guy@retrobbs.rocksolidbbs.com> wrote:
>> Next I would be interested to look at the connections and generate a
>> graph.
>
> I've seen it done and it's quite interesting. Looks complicated, lol.

This is what inpath2dot.pl does.

You can see an example of result here (for all, not just rocksolid.*)
http://shakotay.alphanet.ch/news/inpath.png

On Tue, 22 Dec 2020 08:38:17 +0100 (CET)
Marc SCHAEFER <schaefer@alphanet.ch> wrote:

> Retro Guy <retro_guy@retrobbs.rocksolidbbs.com> wrote:
> >> Next I would be interested to look at the connections and generate
> >> a graph.
> >
> > I've seen it done and it's quite interesting. Looks complicated,
> > lol.
>
> This is what inpath2dot.pl does.
>
> You can see an example of result here (for all, not just rocksolid.*)
> http://shakotay.alphanet.ch/news/inpath.png

Very nice. It's similar to one I've seen but much less congested and
easier to read.

Thanks for posting the link.

Retro Guy <retro_guy@retrobbs.rocksolidbbs.com> wrote:
> Very nice. It's similar to one I've seen but much less congested and
> easier to read.

Actually, it depends on the options you give to the script; here I use
-f 2 -s \.POSTED

-f n factor (defines how verbose the graphic will be. Defaults to 1.)
-s pattern skip (perl-regexp-pattern to ignore in the result)

https://www.novabbs.com/computers/article-flat.php?id=550&group=news.software.nntp#550

Looks like with control-archive 1.9.0 being released, rocksolid has been added to the active file at isc.org:

Hierarchy updates:

* Remove the dictator.* hierarchy (shuttered by its maintainer)
* Add rocksolid.*
* Fix line endings in nntpworld.* key
* Replace the fr.* key and URLs and add a key fingerprint

ftp://ftp.isc.org/pub/usenet/CONFIG/active

--
Posted on Rocksolid Light
news.novabbs.org