PSINet/Cogent Latency

Alexander Koch koch at tiscali.net
Tue Jul 23 06:34:40 UTC 2002


On Tue, 23 July 2002 02:25:36 -0400, Phil Rosenthal wrote:
> I have a small RRD project box that polls 200 interfaces and has it
> takes 1 minute, 5 seconds to run with 60%  cpu usage (so obviously it
> can be streamlined if I wanted to work on it). I guess the limit in this
> implementation is 1000 interfaces per box in this setup -- but I see
> most of the CPU usage is in the forking of snmpget over and over.  Im
> sure I could write a small program in C that could do this at least 10X
> more efficiently.  That's 10,000 interfaces with RRD on one intel -- if
> you are determined to do it.
> 
> I think if you are billing 10k interfaces, you can afford a 2nd intel
> box to check the 2nd 10,000, no?

Phil,

imagine some four routers dying or not answering queries,
you will see the poll script give you timeout after timeout
after timeout and with some 50 to 100 routers and the
respective interfaces you see mrtg choke badly, losing data.

You see, the poll script is doing one after the other,
mainly, so you wait too long and then the next run starts
and then something.

mrtg/rrd is not the tool of choice for accounting / billing
but nice enough for showing you 'backup' graphs for visitors
probably.

Alexander




More information about the NANOG mailing list