Hi all,as a guy that spent a lot of time operating C News and INN I find this interesting.
I couple of weeks ago I posted something similar to this on news.admin.peering because I needed some peers for testing. The project is now at a point that I would like to share it with a few more people.
I've run my own server off and on for a few years. I think 2022 was the last time I ran one. I've always wanted to see more alternatives to INN so I built my own.
About the server:
While INN is more about the classic Unix philosophy. My server, OpenUsenetServer, is closer to the cloud-native philosophy. It is easy to set up and easy to configure. All user and admin functions are configured through a web gui. Before any asks, yes it was AI-assisted development. I took the source code for INN and I used its functionality and way of doing things as the Gold Standard and from there built an entirely new application with backwards compatibility for peers.
The project is Docker-based and can be found here: https://github.com/tgeek77/OpenUsenetServer
as a guy that spent a lot of time operating C News and INN I find this interesting.Shoot me an email if you'd like a user account.
doug eastick <eastick@mcd.on.ca.invalid> wrote:
as a guy that spent a lot of time operating C News and INN I find
this interesting.
Shoot me an email if you'd like a user account.
You get to see some of the stats that I'm collecting and get a feel
of the GUI. You can also spin up your own server at any time. I can't
say that it's production ready, but at least it hasn't crashed yet.
I like the idea yet it does not want to compile on Debian (package conflicts).
Also trying docker and podman exposes package conflicts to get either working.
- Bugs found, each with a fix branch:
- the head is parsed into a map and written back, so header order
changes and folded lines are joined (RFC 5537 lets a relay change
only Path and Xref);
- a bare CR in an article becomes a line break;
- articles that are not valid UTF-8 (Latin-1, yEnc, a NUL byte) are
refused, because the PostgreSQL columns are TEXT; on IHAVE that is
436, so the peer retries them forever. The fix stores them as bytea.
- Design points for later: one feed connection per article, and no
MODE STREAM fallback when a peer doesn't answer CAPABILITIES
(Diablo's feeder doesn't).
This is awesome. I've been out sick for the last week but I will address these
this afternoon.
Jason
| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 76 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 59:34:29 |
| Calls: | 1,265 |
| Calls today: | 1 |
| Files: | 1,362 |
| D/L today: |
4 files (9,632K bytes) |
| Messages: | 286,017 |