Does anyone know of a method to restore recon on a Quake 1.30 server that runs under Windows?
My server supplier, NFO seems to be getting annoyed of me asking for a machine reboot, even though I've been doing business with him for 10 years. The reboot is required about every 3 weeks due to the recon bug. It might be annoying for him since the machine my server is on also accommodates several other servers. I had noticed that the machine had not been restarted for 267 days.
Also, is this only a problem in Windows, or does Linux suffer the same bug?

Thanks in advance for your help,
Jus'Gibs

Do you mean "rcon" ?
The case where rcon commands stop working from remote clients such as a game client console or Rcon Commander ?

If that is so, then yes, we have the same thing on Linux.
I use putty to login to the server, open each server console and stop the game Ctrl+C, then restart each server.

You shouldn't need to restart the operating system, and certainly not require a machine reboot to resolve it in my experience.

Of course Windows has it's own memory management issues, but a Quake server has such low memory requirements relative to modern system specifications, this shouldn't be an issue.

We used to support three Q3 servers on NT4 with 32MB of RAM quite comfortably.
(note that's Megabytes, not Gigabytes)

Thanks Chim, you are always so willing to help.

Yes, I did mean rcon. I did'nt proofread my message, and my spell checker sabotaged me.

The thing is, NFO restarts my server every day, but rcon is not reestablished. NFO has always stated it requires a machine reboot to reestablish rcon.

Many times I myself stop and restart my server, but it has no effect on rcon for me.

Thank you again for your attention.

Jus'Gibs

I've had Q3 services lose RCON access and their on 132c, the Q3 instance is online however when sending commands it does not perform them nor does it provide any informational error, even deliberately mistyping the password to see if it says "Bad rcon", it doesn't do that either, and nothing shows up in the logs.

The only fix is what chimaera500 said, log onto the server by SSH or RDC and restart the Q3 instance, but the OS won't need to be restarted. if your host has TCAdmin you can restart it via that, or maybe try and find a remote PID restarter.