SNMP correct OID

[enediel]

Hello:

I need a reference where to find the correct OID for HPUX, specifically I face problems to monitor the cpu load.

Thanks in advance for any help
Enediel

[Geoff Wild]

See my post in here for PARISC:

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=707417

Note - I'm currently looking for Itamium OID's....

RGds...Geoff

[enediel]

Thanks for the answer,
as the matter of fact that message was the reason to become a member of this forum, but still I have some doubts. I'm using nagios to monitor other hosts, and I need to understand the meaning of those OIDs, for one and multiple cpus, different concepts were included here, system cpu, user cpu, etc.
?where to find the documentation for SNMP on HPUX?

One more time Thanks for your help
Regards
Enediel

[Geoff Wild]

I found some stuff on the net:

http://lists.ee.ethz.ch/mrtg/msg09049.html

http://www.admins-tipps.net/management/mrtg.htm?template=/template/print.tpl

Rgds...Geoff

[enediel]

Thanks again, but look at this.
Executing a walk I obtained the following values

# /usr/bin/snmpwalk -m ALL -v 2c -c public 130.100.80.1 1.3.6.1.4.1.11.2.3.1.1
....
SNMPv2-SMI::enterprises.11.2.3.1.1.13.0 = Counter32: 3498429
SNMPv2-SMI::enterprises.11.2.3.1.1.14.0 = Counter32: 2697088
SNMPv2-SMI::enterprises.11.2.3.1.1.15.0 = Counter32: 66922512
SNMPv2-SMI::enterprises.11.2.3.1.1.16.0 = Counter32: 3690397

Checking with the report for the cpu load directly on the HPUX, I don't know how to understant the values for ..3.1.1.13.0 and ..3.1.1.14.0

Regards
Enediel

[Geoff Wild]

Download and install getif:

http://www.wtcs.org/snmp4tpc/getif.htm

Grab http://mit.edu/afs/athena/astaff/project/snmp/mibs/hp-unix.mib and put in the C:\Program Files\Getif 2.3.1\Mibs directory.

In the MRTG example above, it takes the the values from 13,14,15,16 does something, and computes the load.

I don't understand it fully - but I know it works...

BTW - same values do work on Itanium - as I have that working now.

Rgds...Geoff

[enediel]

I'm sorry if I insist

In this page
http://www.admins-tipps.net/management/mrtg.htm?template=/template/print.tpl#Monitoring%20HPUX%20mit%20MRTG:
I can read
1.3.6.1.4.1.11.2.3.1.1.13.0 - Percentage of user CPU time

In the picture you sent to me
computerSystemUserCPU (Average time in seconds spent by all processors un User Mode)
for you this value was 418080

for me at this moment is 3775708

this value grows up always, so for that I think that the second definition has sense.
My point now is to answer the question ?which is the current user cpu load for this hp unix (%)? having this value

Should I save this value and calculate the second passed between the questions, calculate the differences between these values, etc,
or there is another way to obtain the answer quickly?

In your example on
http://forums1.itrc.hp.com/service/forums/publicProfile.do?forumId=1&userId=CA648840
Target[sha1.internaldomain.netcpup]:.1.3.6.1.4.1.11.2.3.1.1.15.0&.1.3.6.1.4.1.11.2.3.1.1.13.0:public@sha1.pcacorp.net +
.1.3.6.1.4.1.11.2.3.1.1.16.0&.1.3.6.1.4.1.11.2.3.1.1.14.0:public@sha1.internaldomain.net
MaxBytes[sha1.internaldomain.netcpup]: 100
Options[sha1.internaldomain.netcpup]: integer, growright, nopercent
?how you declare on mrtg to calculate any percentage when it received those big numbers?

Thanks in advance for any answer
Regards
Enediel

[Geoff Wild]

Yah - that part I don't know - got that from the net somewhere - I'm not too sure what the + and the & are doing to those 4 values...

Rgds...Geoff

[enediel]

with your doubt I can help you, on mrtg you can have 2 curves to compare the values between them, to have this case the use the & symbol.

in the mrtg declaration, it looks like you merge
1.3.6.1.4.1.11.2.3.1.1.15.0 of sha1.pcacorp.net
with
.1.3.6.1.4.1.11.2.3.1.1.16.0 of sha1.internaldomain.net
(for uknown reason they are not declared with the same host)

and the resulting curve you compare with the merge of
.1.3.6.1.4.1.11.2.3.
1.1.13.0
and
.1.3.6.1.4.1.11.2.3.1.1.14.0 respectively

Regards
Enediel