If you're waiting for snips support, you're in for a _long_ wait.
My first guess is that the number etherload reports is a percentage of
the estimated maximum # of packets possible over ethernet. The only
documentation I was able to find on etherload is:
etherload - it has a little more complicated initialization and
monitoring since its very specific to the operating system.
(from src/docs/design.html)
...oh wait, here we go. In snips/etc/samples/etherload-confg:
# Bandwidth is represneted in percentage of total.
# Packets per sec is a variable depending on the pkt size:
and a table follows.
So the > 100 errors really are errors, since you shouldn't have 117%
bandwidth usage on your ethernet device. That should be reported in
run/*.error, because it's a misconfiguration you wouldn't want
cluttering up your snipslog file.
Tuc wrote:
>
> etherload: Thu Jan 3 16:50:47 2002 Bad bandwidth 117 >100
> etherload: Thu Jan 3 16:56:03 2002 Bad bandwidth 115 >100
> etherload: Thu Jan 3 17:06:42 2002 Bad bandwidth 114 >100
>
> Does it NEED to print this.... I would think thats what snipslogd is for..
|