[mrtg] Re: problem polling cpu statistics from solaris 2.6 box running ucd-snmpd

[Jason Wright]

One thing which I forgot to mention in my original post is that I am
successfully polling traffic stats from server for an hme interface on
'server' (Target[server]: 2:public at server).

On Thu, Mar 09, 2000 at 12:51:18AM +0100, Alex van den Bogaerdt spewed forth:

> Seems to be okay.  Now try the other OIDs.
> > 
> > SNMPGET Problem for 1.3.6.1.4.1.2021.11.9 1.3.6.1.4.1.2021.11.10 sysUptime sysName on public at server
> 
> So: you need to check:
> 1.3.6.1.4.1.2021.11.9
> 1.3.6.1.4.1.2021.11.10

Without leading periods, using both of these with snmpwalk returns

client:~ ] snmpwalk server public 1.3.6.1.4.1.2021.11.9
system.sysContact.0 = System Team <system at example.com>

which isn't what I want and makes me think that snmpwalk wants the
leading periods on numerical OIDs.  With the leading periods,

client:~ ] snmpwalk server public .1.3.6.1.4.1.2021.11.9
enterprises.ucdavis.systemStats.ssCpuUser.0 = 70
client:~ ] snmpwalk server public .1.3.6.1.4.1.2021.11.10
enterprises.ucdavis.systemStats.ssCpuSystem.0 = 0

> sysUptime

client:~ ] snmpwalk server public sysUptime             
system.sysUpTime.0 = Timeticks: (51953299) 6 days, 0:18:52.99

That uptime is wrong though:

server:~ ] uptime
 12:09am  up 144 day(s), 23:19,  7 users,  load average: 0.88, 1.09, 1.29

so some counter might be overflowing somewhere.  The uptime on server's
hme0 traffic graph does show the 6 day value, but it does display.

> sysName

client:~ ] snmpwalk server public sysName
system.sysName.0 = server

PeeWee

-- 
email: peewee at scc.mi.org - useless: http://www.scc.mi.org/peewee/ - efnet:  Pwe
I said you were a state of mind, I believe. I said that if you ran very swiftly
and were acceptably violent, you would be admired. - "The Era of Great Numbers"

--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:mrtg-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/mrtg