[Coder-Com] Found bug about servers linking in ircu2.10.11.beta.24.9 : Origin of Bug
Hello, I found when this bug happens Imagine, I have hub Aserv , leaf Bserv I did by error an U line for Aserv into ircd.conf of Bserv. When I tried to link Bserv, it crashed because there was this U line. I know it is was error to put this U line but I tested on ircu 2.10.10 and they was no crash. I hope I helped you. Regards, Chojin ---Original Message--- Hello, I don't know if this bug has been reported but I'll describe there. I have a hub and a leaf server. When leaf server try to link to hub server, leaf crashes. Debug from leaf said: [2002-3-18 14:24:23] DEBUG [DEBUG]: Reading data from ChoEU [2002-3-18 14:24:23] DEBUG [DEBUG]: Client Parsing: CONNECT = TEST.tnw.info [2002-3-18 14:24:23] DEBUG [DEBUG]: Connect to = TEST.tnw.info[@62.4.22.98] [2002-3-18 14:24:23] DEBUG [DEBUG]: Sending [0x08289ae0] to ChoEU [2002-3-18 14:24:23] DEBUG [DEBUG]: Adding buffer 0x08289ae0 = [:TESTEU.tnw.info NOTICE ChoEU :*** Connecting to TEST.tnw.info.] to = normal queue [2002-3-18 14:24:23] DEBUG [DEBUG]: poll: delay: 1016457893 (1016457863) = 3 [2002-3-18 14:24:23] DEBUG [DEBUG]: Sending queued data to ChoEU [2002-3-18 14:24:23] DEBUG [DEBUG]: poll: delay: 1016457893 (1016457863) = 3 [2002-3-18 14:24:23] DEBUG [DEBUG]: Sending [0x08289ae0] to = TEST.tnw.info [2002-3-18 14:24:23] DEBUG [DEBUG]: Adding buffer 0x08289ae0 [PASS = :test] to normal queue [2002-3-18 14:24:23] DEBUG [DEBUG]: Sending [0x082898a0] to = TEST.tnw.info [2002-3-18 14:24:23] DEBUG [DEBUG]: Adding buffer 0x082898a0 [SERVER = TESTEU.tnw.info 1 1016457653 1016457863 J10 AFAP] +h :TNW Test Server] = to normal queue [2002-3-18 14:24:23] DEBUG [DEBUG]: poll: delay: 1016457893 (1016457863) = 3 [2002-3-18 14:24:23] DEBUG [DEBUG]: Sending queued data to TEST.tnw.info [2002-3-18 14:24:23] DEBUG [DEBUG]: poll: delay: 1016457893 (1016457863) = 3 [2002-3-18 14:24:23] DEBUG [DEBUG]: Reading data from TEST.tnw.info [2002-3-18 14:24:23] DEBUG [DEBUG]: Client Parsing: PASS :test [2002-3-18 14:24:23] DEBUG [DEBUG]: Client Parsing: SERVER TEST.tnw.info = 1 1016453905 1016457884 J10 AEAP] +h :TNW Test Server [2002-3-18 14:24:23] DEBUG [DEBUG]: Got SERVER TEST.tnw.info with = timestamp [1016457884] age 1016453905 (1016457653) [2002-3-18 14:24:23] DEBUG [DEBUG]: sv_cl: check access for = TEST.tnw.info[62.4.22.98] TESTEU: s_conf.c:488: attach_confs_byname: Assertion `0 !=3D tmp->name' = failed. Aborted TEST.tnw.info is my hub. How fix that ? thank you for helping. Regards, Chojin
Re: [Coder-Com] Ideas
- Original Message - From: "Alexander Maassen" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Tuesday, April 02, 2002 5:03 PM Subject: Re: [Coder-Com] Ideas > Dave, ya right, but there is still an issue within the +x model undernet > will use, issue is a users hosts gets 'hidden' *AFTER* he successfully > authed with X, so that system is actually kinda useless in my opinion. > > Reason: > Problem is that users will still be 'nukable' since there IP does not get > hidden upon connect and joining channels thus still showing there TRUE ip. > > Solution: > Forced +x mode upon connect using a default hostmask (for example : > [EMAIL PROTECTED]) if a user doesn't want this feature then he can > still do a -x afterwards, then it's up to the user to take care of his > protection. > lol You forgot /quote who %i bug :) fake host or not, people can still see your IP :( > The other fact is that ircu sets the new host by emulating a quit/rejoin of > the user, thus showing quits all over the channels. Kinda irritating if you > got 70k users constantly reauthing etc. I think it will put a hell lot more > stress on the server handling then if you would handle the hostchange > internally, also imagine that clients also have event handlers, some of them > might return output after a join/quit, thus increasing the bandwidth usage > dramatically as well. Handling it internally like other ircd's already do > releases that pressure. > I'm working for my network for a solution to this. Chojin
[Coder-Com] gline uworld command doesn't require to be authed
Hello, I noticed an ircop can do /msg uworld gline ... with authed to uworld, but he can't remgline or anything as not authed. Is that a bug ? -- Chojin
[Coder-Com] G-Line with UWorld
Hello, when G-Lines are done with UWorld. when irc server is restarted and cservices too, UWorld doesn't put back G-Lines even when G-Lined people come on IRC... It's a bug ? Or may I forgot something ? -- Chojin
Re: [Coder-Com] Ban/Kill with fingerprint
- Original Message - From: "Chris Crowther" <[EMAIL PROTECTED]> To: "Chojin" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Wednesday, April 17, 2002 8:02 PM Subject: Re: [Coder-Com] Ban/Kill with fingerprint > On Wed, 17 Apr 2002, Chojin wrote: > > > what about ban/kill using fingerprint associated with a provider ? (I want > > to ban someone with *.aol.com associated with his fingerprint). > > What exactly do you mean by fingerprint? You can set klines and > bans on masks such as *@*.aol.com already. > I mean fingerprint is a sort a computer id and if I gline a users with *@*.aol.com with his fingeprint, even if user changes his IP (with isp reconnection) he is still glined because his fingerprint is glined. Other users from aol can join. -- Chojin
Re: [Coder-Com] Ban/Kill with fingerprint
- Original Message - From: "Valcor" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Thursday, April 18, 2002 1:25 PM Subject: Re: [Coder-Com] Ban/Kill with fingerprint > On Wed, 17 Apr 2002, Chojin wrote: > > > > What exactly do you mean by fingerprint? You can set klines and > > > bans on masks such as *@*.aol.com already. > > > > > > > I mean fingerprint is a sort a computer id and if I gline a users with > > *@*.aol.com with his fingeprint, even if user changes his IP (with isp > > reconnection) he is still glined because his fingerprint is glined. Other > > users from aol can join. > > You mean maybe like their processor serial/mac address/OS id? > Each OS has his own fingerprint. By example, when sending a SYN|ACK|FIN|RST on a closed port, and user answers with RST, It's a Windows OS (because it doesn't follow the RFC). I also heard MSS (Maximum Segment Size) is different for each user then it could be a sort of user fingerprint. -- Chojin
Re: [Coder-Com] Make erros
Hello, you must upgrade your automake to 1.6.1 -- Chojin - Original Message - From: Inge Hagen To: [EMAIL PROTECTED] Sent: Saturday, April 27, 2002 12:07 PM Subject: [Coder-Com] Make erros I did install autoconf-2.52 but I get this error, someone how to fix it I have try to find a new version of automake but I cant find it. Hope someone know how to fix this.. I have tried to reinstall gnuworld but it did not work [gnuworld@comet gnuworld]$ gmake cd . && /bin/sh /home/gnuworld/gnuworld/missing --run aclocal-1.6 aclocal: configure.in: 30: macro `AM_PROG_LIBTOOL' not found in library WARNING: `aclocal-1.6' is needed, and you do not seem to have it handy on your system. You might have modified some files without having the proper tools for further handling them. Check the `README' file, it often tells you about the needed prerequirements for installing this package. You may also peek at any GNU archive site, in case some other package would contain this missing `aclocal-1.6' program. gmake: *** [aclocal.m4] Error 1 [gnuworld@comet gnuworld]$ Server Admin @ irc-solution.net -Comet-
Re: [Coder-Com] Make erros
New version on automake can be found at ftp.gnu.org - Original Message - From: Inge Hagen To: [EMAIL PROTECTED] Sent: Saturday, April 27, 2002 12:07 PM Subject: [Coder-Com] Make erros I did install autoconf-2.52 but I get this error, someone how to fix it I have try to find a new version of automake but I cant find it. Hope someone know how to fix this.. I have tried to reinstall gnuworld but it did not work [gnuworld@comet gnuworld]$ gmake cd . && /bin/sh /home/gnuworld/gnuworld/missing --run aclocal-1.6 aclocal: configure.in: 30: macro `AM_PROG_LIBTOOL' not found in library WARNING: `aclocal-1.6' is needed, and you do not seem to have it handy on your system. You might have modified some files without having the proper tools for further handling them. Check the `README' file, it often tells you about the needed prerequirements for installing this package. You may also peek at any GNU archive site, in case some other package would contain this missing `aclocal-1.6' program. gmake: *** [aclocal.m4] Error 1 [gnuworld@comet gnuworld]$ Server Admin @ irc-solution.net -Comet-
Re: [Coder-Com] msgs/notices to a channel with/when a client is connected
Yes, an eggdrop ;) -- Chojin - Original Message - From: "Cristian Rauta" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, May 27, 2002 2:07 AM Subject: [Coder-Com] msgs/notices to a channel with/when a client is connected > Hey, > Anyone knows how to make gnuworld to send msgs/notices to a channel with/when a client is connected ? Or .. maybe a service compatible with u2.10.11 who know that > > -- > Cristian Rauta > >
Re: [Coder-Com] Umode +k
> Is it possible to make users +k ?? With a patch or something ... > > Tnx in advance ;-) > Ace > Hello, to allow global operators to use umode +k, in s_user.c you need to modify in set_user_mode() function this: - if (!(setflags & FLAGS_CHSERV)) + if (!(setflags & FLAGS_CHSERV) && !IsOper(Optr)) ClearChannelService(sptr); -- Chojin
[Coder-Com] Re: Mode +k
> This is in ircd/s_user.c: > > if (!(setflags & FLAGS_CHSERV) && !IsOper(Optr)) > ClearChannelService(sptr); > > Compile: > > make[1]: Binnengaan van map `/home/maartenz/IRC/ircu2.10/ircd' > gcc -g -O3 -pipe -I. -I.. -I../include -c s_user.c -o s_user.o > s_user.c: In function `set_user_mode': > s_user.c:1279: `Optr' undeclared (first use in this function) > s_user.c:1279: (Each undeclared identifier is reported only once > s_user.c:1279: for each function it appears in.) > make[1]: *** [s_user.o] Fout 1 > make[1]: Weggaan uit map `/home/maartenz/IRC/ircu2.10/ircd' > sorry, it is sptr, no Optr (it is a variable for my needs) if (!(setflags & FLAGS_CHSERV) && !IsOper(sptr)) ClearChannelService(sptr); -- Chojin
Re: RE : [Coder-Com] Force Nick Change/Mode Change
> -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > A umode change/svsnick MUST be applied to ALL servers... > > Math > [EMAIL PROTECTED] Could someone send to me this SVSNICK patch please ? -- Chojin
Re: [Coder-Com] The message I sent..in plain text.
> That actually would be a good idea...along with something else I've seen > on other networks... > > What if we created some in-channel op commands for "X" like... > !kb (Kickban) > !b (Ban) > !k (Kick) > !op (Set Op) > !dop (Unset Op) > !v (Voice) > !dv (De-Voice) > !ld (LockDown: Kicks everyone that has no access above a certain level > with X) > !tb (Tempban) > > And so on and so forth... > > as it has been said before on this mailing list, X uses +d mode not to see channel messages which generates useless ressources, bandwidth usage. -- Chojin
