[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
  Web www.spinics.net

high-end routing

I've got a RH7.3 box that I have succesfully setup to route 4 subnets 
out to our upstream provider. Routing-wise, everything works fine. My 
problem is that every 5 minutes (almost to the second) the box starts 
dropping packets like mad, and generally just goes unresponsive. The 
server itself is not locked up, but pinging either of it's interfaces, 
or pinging through it fails utterly. Oddly enough, about 1 minute later 
everything returns to normal. It is important to note that this is a 
_heavy_ traffic network, passing approximately 20-30Mb/s through those 
interfaces. There are no IPTables rules in the way (i've removed all and 
still see the behavior). The box itself should not be a problem, I 
think: dual P3-700 512MB ram, dual 3Com 100TX FD. I have also confirmed 
that it is not a problem on our upstream. So I guess my question is: are 
there kernel tweaks I need to do to better tune this? Is an ARP cache 
getting updated too frequently? The regularity, and the return to 
normalcy really make me think it's some setting I just need to tune, but 
so far I haven't come across anything terribly helpful.

Any thought are appreciated.

Thanks, Matthew
Matthew Boeckman			(816) 777-2160
Manager - Systems Integration		Saepio Technologies
LinuxManagers mailing list - http://www.linuxmanagers.org
submissions: LinuxManagers@linuxmanagers.org
subscribe/unsubscribe: http://www.linuxmanagers.org/mailman/listinfo/linuxmanagers

[Home]     [Kernel List]     [Linux SCSI]     [Video 4 Linux]     [Linux Admin]     [Yosemite News]     [Motherboards]

Powered by Linux