how long does it take to reboot the list server?
As long as it takes to reboot a gserver. Rebooting it is not the reason its taking so long. Finding out why its acting up and fixing the problem is.
no sorry just reboot
You can’t just reboot you have to find the issue first or else it will happen again. You wouldn’t want that would you?
Thinking aloud… would it be possible to attach gdb to it, get a ‘running’ crashdump of it (no, I don’t know gdb enough to know if it is really possible), maybe just changing it’s port where it listen with gdb, and launch a new instance with the usual port, and keep the old instance listening on a debugging port.
I hope this gonna work again soon 
:\ You guys should know that it’s not just a simple fix, if that were the case it would be up by now. It’s gotta be a problem if Nalin can’t just flip the switch and all is happy in Graal-Land.
It isn’t crashing. So that wouldn’t do any good.
XD Arcain its funny lol
Looking at gdb help I see this that seems usefull:
http://sourceware.org/gdb/current/onlinedocs/gdb/Checkpoint_002fRestart.html#Checkpoint_002fRestart (Linux only)
I don’t know if they save to disk… or just in memory however… I think it is just in memory, sorry.
I will be home this weekend. I’ll try to figure out what the problem is when I get home and have access to my full programming environment. The odd thing is that the serverlist is now using the same exact socket code as the gserver. Everything looks like it should work. But, for some reason, the serverlist is accepting incoming connections, then terminating them. I haven’t been able to replicate this behavior on any of my machines which further complicates this whole issue.
I almost wonder if it is time to migrate all my socket code over to boost.asio.
is any of the code written in Perl? if it is maybe I can help … hope some debug statements at the right places can help. or see if some other program or cron task is running that is kicking out the connections right when they are established.
then again, it might just be from a missing semicolon that somehow still lets it compile
i know you just joined, but so far you are the only one on this forum that talks sense. everyone else is nuts.
You say that like it’s a problem. :colbert:
So what is the news now ?
Still down. 
That i know but do Nalin make something ?
Better yet, couldn’t Cadavre just compile and run an older version?
So how many people have the knowledge/machine access to fix these sorts of problems when they come up? seems like the list server has a low bus factor