Magnus Holmgren <[EMAIL PROTECTED]> writes: > Boy, what an old bug!
> This has been discussed some times[1], but no conclusion reached. I'd > like to suggest (again, probably) the following: As near as I can tell, the reason why no conclusion has been reached is because no one has written the code for any of the various reasonable solutions that have been proposed. Everyone seems to be waiting for the inetd maintainer to write the code for their proposed solution, which he has said repeatedly he doesn't have time to do. It's good that you asked here about what people think of your solution. Personally, I dislike xinetd and don't want to use it, but I don't have a problem using its format as the input format (it's better than what update-inetd uses now). However, any replacement update-inetd also has to support the old format and ideally should be able to write xinetd configurations based on it for a transition period. However, even more importantly than the discussion of possible solutions, we need an implementation. If you have a solution, please *implement* it (and in a way that doesn't make large parts of Debian buggy, which means being backwardly compatible) and I bet you'll be able to build a consensus around your solution as long as it's halfway reasonable. -- Russ Allbery ([EMAIL PROTECTED]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
