Psybnc ping timeout on undernet

[umic]

I have a problem with my psybnc. It is installed on Red Hat 9 and until a few weeks ago it worked just fine. I have more than one ip on the server.. 8 exactly so I can connect more than 16 users with no problems. I have 10 users now and after an hour or at most a day all the users drop in ping timeout at the same time. If I "kill" the bouncer and start it again it works just fine but the same reaction.. one hour and they all drop. I have no idea what happened to them. I tried a new psybnc, I am using 2.4 now and it all happens in the same way.
Does anyone have any idea why? Please give me a hint

[sjaz]

Probably polarhome's line not coping.

[DenisF]

The whole "I have" bit in his post kinda indicates that he ain't talking about polarhome chaz m8

As far as your problem goes, umic, id'e say that either your line can't cope with all of the simultaneous pings, or the server itself can't, and packet loss occurs.
If i were you, id'e upgrade it to fedora2 or atleast slap a 2.6 kernel in there, that might fix it.

but do that only if you're one of the brave few

[sjaz]

kernel? I think he meant hes using 2.4 psybnc.

Also I guess your right as he has "more than 1 ip".

[DenisF]

I know that he meant psybnc 2.4 [which doesn't exist btw, 2.3.2-4 is the latest]
but he mentioned that his server runs rh9, which uses the 2.4 kernels.

and that's what i meant, that he should upgrade toa newer distro and/or a new kernel.