[EMAIL PROTECTED] wrote on 17/02/2003 22:46:21: > > most of the ghostkill code has already done by me and is available for a > > long time on the sf.net gnuworld patches list. You just did the addition > > to be able to turn it on/off. > I just looked at your code(I assume that it is the one under the sf login > key2peace), and it seems that you did this differently, by creating a new > command called GHOST. The way I did it was, if the user has enabled the > ghostkill flag, if they log in again, they get killed. Which of these were > actually approved? What was the original wording given to cservice? >
To the best of my knowledge, approval was only given by CSC literally a few days ago. There is no specific wording. OUTSiders patch was submitted a few weeks before that, but it hadn't been looked at yet, as to be honest - I thought the chances of it being accepted by CSC for Undernet use where slim. I don't think that CSC fully understand what the patch entails and I suspect that crap from other 'parties' on the Undernet will hit the fan when it goes live, which is why I'm waiting for them to discuss it amongst themselves first. :) > Also, does anyone have any preference for server kills or kills by the nick? My > code uses server kills, while OUTsider used client kills. It should be easy to > change either of them across to use the other type. > The preference is for either a seperate command (To avoid accidentally killing a client you forgot you had logged in by using LOGIN), or for a switch to be added to the LOGIN command. The KILL should be issued from the server numeric. (So they can be filtered out a little easier by moaning opers). > Also, is this going to be committed, because I see that it isn't in current > cvs(I checked the cvs tree before I wrote my patch). I wrote a security critical > patch a while ago, and I gave it to a few people and I think sent it to bugs@ > (not sure exactly, it was a while ago). It has been about 8 months since I first > wrote a patch, and it looks like your patches are having the same problems with > no one committing it. Firstly, give me the exact date, time & subject matter of your security critical patch. To suggest it has been 8 months is quite frankly a joke! (unless I never actually received it or was made aware of it in the first place). To the best of my knowledge, I've committed every security related patch to the tree as soon as I've been available to do it. If it is a critical, exploitable problem, then there are three other people who have access and can commit the patch in my abscence. Asking in the usual place will reveal these people :) You have to understand that CSC very rarely approve anything that introduces new features, it makes their life harder supporting the thing in #cservice, and it takes time to get all the people dishing out the advice up to speed with new stuff. It takes an awful lot of convincing to get stuff accepted, and Iso has spent a great deal of time trying to get people to understand the benefits of the GHOST command over the past few months. Cheers, Greg.
