possible eggdrop bug: timers over #1000

[snowroom.org]

Hi there,

on the last night i think i recognized an eggdrop timer bug ...

something at 1 pm the eggdrop bot hangs and no sockets or local files could be opened by the eggdrop. there is no bug in the tcl script i have written so it is not necessary to post it here.

now my question to all: does anyone have that problem too ?? does anyone have a solution to solve this bug/failure/or something else?

additional #1: if i try to connect via ctcp on the partyline on this bot i get no response from it. tcl bindings are functioning good but a bit laggy. if i connect via telnet directly to the partyline without any irc client, i can login ... but when i try to rehash the eggdrop it dies.


additional #2: this bug or failure comes up the timer #1000 or something around 1000 .... there are 8-10 timers running all with 2 mins delay opening a socket and get some stuff from websites and posting it to 1-2 channels on irc
----

thats all to this problem ... i hope anyone here knows a solution or made similar experiences with eggdrop´s timers module.

it would be nice to get a fast feedback from the dudes who loves eggdrops ..

so long ... snow

[strikelight]

Sounds more like too many open file/socket descriptors... Make sure your scripts close files/sockets before opening multiple instances..

Also, make sure your scripts check for instances of their timer already in existance, before starting a new one.. Because upon a .rehash, eggdrop will not kill active timers, and thus you may be starting up multiple timers for the same thing...

[snowroom.org]

right man ... your 'fears' about coding fool stuff is good to understand ... but:

i allready added these lines ... like:

Code: Select all

proc setXXXXtimer { } {
 if {![string match "*XXXXdata*" [timers]]} { timer 2 XXXXdata }
}

and

Code: Select all

flush $sock
return 0 < on the end of script

or did i forget anything ?

[strikelight]

flush'ing a socket doesn't close it.... use the 'close' command.

[snowroom.org]

ok .. i 'll try it out ... so long

[snowroom.org]

hmm... very big thx to strikelight ... this was a mistake in the tcl script *wonder* ... have a nice weekend .. cya

[snowroom.org]

hi,

i'm back again ... i addet this close $sock line into the script ... .but the eggdrop is still unstable. After around 24 hs or more the bot laggs very much and makes now more output.

Now, I' dont know again where is the failure ... the sockets gets open correctly and closed. the webservices are still available and so on.

Could You help me again ?

So long ...