Anope IRC Services

Anope Support => 1.8.x/1.7.x Support (Read Only) => Topic started by: madman on January 01, 2009, 03:40:28 PM

Title: modes condition
Post by: madman on January 01, 2009, 03:40:28 PM
Hello and happy new year to all. I dont know if there's another topic for this cause i couldnt find something similar... When you try to give a mode to a user from chanserv or a botserv's bot it doesnt check if the user has already that mode. For example if you use /cs op #channel nick chanserv will give +o even if the user has already that mode. I think there should be a condition to check if the user has already a mode and if not then give it.
Title: Re: modes condition
Post by: Jobe on January 01, 2009, 04:36:34 PM
This is intended behaviour on Anope's side in an attempt to save processing time.

In my honest opinion, the IRCd is at fault for sending the mode through as if it was not already set to begin with.
Title: Re: modes condition
Post by: katsklaw on January 13, 2009, 05:24:54 AM
This is intended behaviour on Anope's side in an attempt to save processing time.

In my honest opinion, the IRCd is at fault for sending the mode through as if it was not already set to begin with.

it is also intended behavour on ircds as well as it removes the above mentioned check to conserve CPU cycles.
Title: Re: modes condition
Post by: ShawnSmith on January 13, 2009, 07:11:58 PM
it is also intended behavour on ircds as well as it removes the above mentioned check to conserve CPU cycles.
I believe the ircds do the check for most modes, but since multiple users can have +v/+h/+o/+a/+q it doesn't bother doing the check for them.
Title: Re: modes condition
Post by: katsklaw on January 13, 2009, 07:43:40 PM
I believe the ircds do the check for most modes, but since multiple users can have +v/+h/+o/+a/+q it doesn't bother doing the check for them.

some do and some dont. Bahamut for example doesn't check any of them. The ircd insures the user is authorized the mode of course but doesn't check if the mode is already set.
Title: Re: modes condition
Post by: Master on February 17, 2009, 04:27:36 PM
Although, I kind of wish there was some sort of protection against flooding with this type. With it the way it is, users can simply flood those commands and flood the channel with it.
Title: Re: modes condition
Post by: casper1256 on February 18, 2009, 10:37:55 AM
If it happens and flooding seems to be in the rules just /os ignore and in the worst case scenario you could /cs drop it as a lesson.

Although these are 'extreme' solutions, I've never had to use them before under this circumstances and I think it's needless.

Just makes me wonder Malcolm, how many user you have had flooding in that way?
Title: Re: modes condition
Post by: katsklaw on February 19, 2009, 01:07:28 AM
modes are controlled by the IRCd, nothing services can do except react and by then it's too late. Secondly, suspend is a better solution than drop. Third, lets not forget about sendQ/recvQ.