Sirius wrote:
> Hi Michael,
>
> Ah, okay. I thought there would be a simple way to share this data
> once a user uses /OPER, and then burst that in a netburst too.
>
> If you think this is a really bad idea I can leave it, it would just
> be great to add it to our new command.
>
> Thanks!
What d
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Also note, by the way, that bans on the user's real host mask are also
> effective even when the user is +x.
although for some reason X doesn't enforce these bans
- --
No evil can happen to a good man. -- Plato
-BEGIN PGP SIGNATURE-
Versio
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> While +r cuts down the number of spam bots extremely well, it also seems
> to starve the channel of the "normal" people/newbies. It would be nice
> if the server sent instructions something like "login to X to join"
> or "register at http://cs
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> The point of my rant is, I don't care if I'm a few seconds lagged to
> the server I'm connected to, I just I want to connect to where I want
> to connect to. Geographically close doesn't necessarily mean close in
> terms of networks.
>
> Please j
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>
> I think it would be the decent thing to do to make it easier for the
> impaired to chat on Undernet, and as isomer said, it's being worked on atm
we will try, however rolling out new versions is a royal PIA :)
- --
Van
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> The user mentioned that he knew other blind people who this is affecting.
> Is there any way to "fix" the problem?
> I don't give a rats a** about MS chat, but if it's stopping innocent,
> handicapped, people from continuing to chat on Undernet, it
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> > If people set themselves /away multiple times only the first is
> > broadcast to all servers. If I had my way, we wouldn't broadcast /away
> > at all :)
>
> I recommend storing the away message locally only, making it only
> available on a /whois
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> I notice that when a user sets their away message it propagates to the
> remote servers. If a user does not clear their away message, but sets it
> to a new one, it updates on the local server *only*. Remote servers still
> hold the old away mes
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> I have a litle problem with ircu2.10,
> There cant be more than 1 operator
> 'inlogged' with the /oper command
> at the same time (if I am an ircop, it says to
> everyone else who tryes to login as a oper;
> 'No O-Lines for your host' but when I am
>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> In the past days I noticed a new kind of trojan... they use the realname "0"
> (zero, without the quotes).
>
> I stupidly had the idea to who then using /who 0 xr ... guess... I got
> flooded off, that matches all users, probably because 0 = NULL.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Coder-Com.
>
> Just a quick question. I'm running GNUWorld and IRCu (11.04). I changed the
> ACCOUNTLEN in include/ircd_defs.h as I want to allow the usernames with
> mod.cservice to be longer and i put this to '25'. Is it possible to create
> us
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> I would suggest that everyone who is member of this mailinglist to
> download the newest definitions for your Virusscanner becaus' It's
> annoying to recieve viruses every day. For more information about the
> virus look here > http://vil.mcafee.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> [13:02] [13:01] -Season`d:@#zt- @ @ @ @ hummm better paste me
> what is happening
> [13:02] [18:04] -BlakJ|Wrk:@#zt- @ @ @ @ [Ops/#zt] so its your
> scdrip tthen :P
>
>
> Forgive my dodgy typing, but it would appear that opnotices are prefixed
>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Mirc allows you to do a channel search via "key word". With Mirc you can
> type in a word, "help" for example and the client will display all channels
> with that word in it's name and/or topic (you can also set max and min
> users). Would it be p
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> For what it's worth, the problem discussed below (by reed) is a problem for
> families" where more than one person uses IRC. If mom is an op on
> #adult_chat. and wants to turn the computer over to her son John for a
> while, then being able to de
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>Hey! How about we make everybody happy and create a F:line to handle -x
>activation or deactivation?
>
Haven't we just been over this? No.
* We don't want people flooding -x/+x
* It leads to race conditions in the network
* If your IP is hi
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Is there any plans to implement a -x feature, or at least make it an
> *optional* feature? I've noticed that on some of the servers that +x is
> automatically set, which is quite frustrating.
>
- -x is racy, and kinda pointless, the servers that s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
We've release 2.10.11.03. We'd like undernet admin's to upgrade as soon
as practically possible as this fixes several desync's that we've seen
on the network.
Other changes:
* users logged into services, but not +x (hidden host) should still
* match
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>
>
> Interesting..did our quit/part protocol change somehow recently? Probably
> worth recreating this problem with mschat on undernet and another network
> to see what they're choking on..
>
Due to some coding changes, parts without a part messag
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>
> > Shouldn't it be sending a KILL upstream?
> >
> > This means that the first server thats clock is slow will squit every
> > other server on the network. Should be fun!
>
> Actually, an addendum to my last email--I couldn't decide whether to
>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> In my situation, as one of the ops in #usa (which tends to have 200-350 users
>during any part of the day), the host-hiding option further stresses what I consider
>to be a far too small banlist size even further; just the simple addition of this
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Hi,
>
> This is just something I was wondering... If I would be abusing in a
> channel with my real host matching the folowing mask
> *!capricorne@*.kabel.telenet.be, I probably would get banned. I could
> return with another host & still get banned
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> On Thu, Oct 17, 2002 at 07:33:45PM +0100, Chris Crowther wrote:
> > On Thursday 17 Oct 2002 2:40 pm, Richard Smith wrote:
> >
> > > > regitrations from the same e-mail sub-domain would likely be queried.
> > >
> > > Are queried ;)
> >
> > Thoug
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> > I was whois'ing a friend that is connected to mesa.* and I observed that a
> > remote whois (/whois nick nick) was returning me a different username.
>
> I attempted to replicate this and was unable to do so. Do you have more
> examples of this
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> > This will block all eembots that were active on the channel because of =
> > 'Internal database doesn't match with server' and it is not compliant to =
> > the RFC1459. Can you please take a look at this?
I believe this is fixed in one of the rec
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
2.10.11.02 Released.
* Fixed the stats core bug. Oops.
* Added 001 properganda
* Some code cleanups
* 005 shows for non opers in /version now too. Oops.
* RPL_ENDOFNAMES now is sent when it should be. It was left off in a
few minor cases in .0
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>
> I seen that if there is no reason from client:
>
> :DaDreamer!~[EMAIL PROTECTED] PART #tg.jiu :bye bye honey
> :DaDreamer!~[EMAIL PROTECTED] JOIN :#tg.jiu
> :DaDreamer!~[EMAIL PROTECTED] PART :#tg.jiu
an IRC line is split up into up to 15 param
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> On Sun, Aug 18, 2002 at 01:23:51PM +0200, Dani?l Boeije wrote:
> > Hello,
> >
> > is there a mode B(Bot) to mark bots in IRCu?
>
> There should be one, its a good idea.
>
> Bots that are spotted and do not use it should
> be kill targets (to moti
> > It is very, very important that the max users field be of the correct
> > form.
> > --
> > Kevin L. Mitchell <[EMAIL PROTECTED]>
>
> indeed. i hope it'll be fixed? (127 looks very ok to me)
>
> i think of another idea: in ircu, check the numeric of the S message and
> force it to be 2^n-1 (o
> Is this affecting only bans that attempt to ban nicks containing the
> character \, or are nicks containing \ being totally able to bypass all
> bans?
>
> In the latter case, then yes there would be an urgent problem to deal with.
> However, if it's only a matter of not being able to specifical
at build grand
structures out of nothing but ideas and thought. I'm honoured to have
been able to serve the Undernet community, and it (literally) brings
tears to my eyes to be forced by things beyond my control to abandon it
at this late hour.
- Isomer (Perry Lorier)
--
The End, run the credits, cue the final music.
> IIRC, there were plans to convert ircu to use the ADNS library in the future
> (perhaps u2.10.12), instead of the old resolver code, no?
I'm very interested in getting rid of the old res.c, it's very old and
frigile. Replacing it with ADNS appears to be the best solution.
Sharing code between
> ok
> Heres the complication:
> Due to the stupidness of the FAT(32) and possibly NTFS file systems,
> Configure.in will overwrite configure.in and will lead to some complications
> when running `make config'.
2.10.11 does away with "make config" and instead puts almost everything
into ircd.conf
> My proposal is to add a new channel setting (level
> 500). For example FULLBAN (OFF or ON) which allow (or
> not) users to set bans on *.* . I guess is a good
> ideea... 10x & please excuse my english.
>
How does the IRC daemon know if a ban is a full ban?
*!*@*
*!*@*.*
*!*@*.???
*!~*@
> On Sat, May 11, 2002 at 11:04:49PM -0400, [EMAIL PROTECTED] wrote:
> > hello:
> > maybe is a server error or whatever is it , i've saw many users reporting the
> > same ... after a few minutes i got the same problem.
> > disconnected (from Amsterdam.NL.EU.undernet.org) Reason: Closing Link:
>
> On Sat, 11 May 2002, Bas wrote:
>
> > proposal MD5+salt password, feel free to comment on this
>
> I'm toying around with doing this right now, I have a patch for
> plain MD5 atm, just needs changing to salt the MD5.
>
in 2.10.11 we pass the full salt to crypt(3), so if the system uses
>
> I'd like to ask where is the whocmd.s in ircu2.10.07 ?
> We'd like to notice Opers when someone makes whois on them
> but we couldnt find it in whocmds.c in ircu2.10.11.
ircd/m_whois.c, not that I think that it's a wise idea :)
--
Overdrawn? But I still have checks left!
Because of our move to line buf's, we've done some research into the
line lengths of lines sent on IRC (serverside and clientside), Below
are the stats so far, I hope to do more "accurate" ones later.
It's a very very long tailed distribution, with 90% of lines being less
than 150 charactors lon
> There are people on coder-com that can answer these questions, perhaps
> better than some server admins. It would be great for these people to speak
> up, and help the admins by sharing their expertise.
Undernet has for a very long time had problems with floodnets. These
use up connections a
> Is the default setting 1 connection per user@host? If so, that may be the
> problem. If that's the case, how about setting the default to 2 and give
> the admins the choice to change the default?
The default is actually "unlimited", except for "unresolved" addresses
which are limited to 1 con
> Okay... number two... noticed that some servers are not responding
> with the idle times when doing a /whois nick nick request. Again I'm
> on amsterdam.nl.eu.undernet.org so I don't know if that has anything
> to do with it. Is it just me again?
>
Known bug, we're playing merry games
> ack, I sent this e-mail to early, thinking that it *did* remove the ban.
> However, it did not, a couple of minutes later:
>
> What's going on?!
The bans were placed with an access higher than yours. You then tried
to remove it and couldn't, but X let you unset the ban from the channel
(since
> /usr/local/pgsql/data/base/1...IpcSemaphoreCreate: semget(key=1, num=17,
> 03600) failed: Function not implemented
>
Postgres is trying to use the "semget" syscall. This is a "System V
Shared Memory" system call from memory. This isn't compiled into your
kernel. Time for a recompile.
--
I
> hello
> I have ircu2.10.11 beta and CS can not links. He is connected but he not joined
>#cservice.
>
> Can you help to links CS6.0 or 5.1
>
> thanks you ( iam french man and it's very difficult to me to find help ) +++
CS doesn't work with recent version of our daemon. One of the major
p
I thought it's probably time to write an email explaining what we're up
to in the world at the moment. So heres what we've done to 2.10.11
in 2.10.11:
* /names requires a "0" to list all names since people tend to flood
themselves off.
* Server side /JUPE
* Server side /GLINE
* Massive en
> Another thing is: why do you actually use
> @.users.undernet.org and not @users.undernet.org
> ? For opers it might become @opers.undernet.org.
>
> Another stupid idea ? No, just think about it, I did, now it's your turn to
> think about it as well...
Undernet usernames are valid domain names,
> The section contains:
>
> -DEXTENDED_NUMERICS This option configures the server to send
> extended numerics as well as parse them. This option should only
> be used on networks that run ircu2.10.07 and above only.
> -DFERGUSON_FLUSHER If you have a server with a lot of resources
> available thi
>
> > -Original Message-
> > From: Tom Rons>
> >
> >
> > > Is there some kind of Spam/DOS-prevention that could cause this
> > > behaviour?
> > >
> >
> > Yes, ircu contains client connection throtte code, which disallows
> fast
> > multiple connections from the same ip address. This is
> How about chanmode +B to be able to ban users based on there userid
> (the lower would be for ban, the higher for userid, simply check on AC token
> if it matches, in combination with chanmode +r you should have a nice system
> to keep certain users out without the option to evade such a ban.
> While implementing apass1-2, my eye fell on m_create.
> It seems that what happens there should be changed.
>
> Firstly, the last exit_client KILLs sptr, a client. It doesn't SQUIT
> the server of sptr. But changing this into squitting the server of
> sptr would be at least as wrong - conside
> >> The P10 protocol as i know it,
> >> SENDERNUM B #channel TS +modes users :%bans
> >> doesn't have timestamps for the bans
> >>
> >
> > Like all channel modes, it uses the TS of the channel, there aren't
> > TS's on individual modes, they are just "merged".
> >
> i know the TS isn't the TS of
> Hey,
>
> Just an insignificant suggestion--I thought of a few more ISUPPORT TAGS that
> could perhaps be considered:
>
>
> USERMODES=
This is available in 004.
> KEYLEN=
Theres a good argument for this. Currently for some rather obscure
reason, it's currently "23" charactors. No idea at
> > i guess that those bans were actually *originally* set prior to any
> > other BUT ... when you *see* a server SET bans .. it doesn really
> > set them at the present moment like if a user did, it is just
> > resynchronizing with the list that server has on its side .. I
> > guess it just pass
53 matches
Mail list logo
