Hello,
I'm using Zabbix 2.0 on few Linux & HP-UX hosts. Recently I noticed that zabbix agent shows different CPU utilization values than other tools ( glance/top/vmstat/measureware) on some HP-UX servers.
For example, right now on one of the HP-UX hosts system.cpu.util[] returns values from 64.907216 to 69.867061. In the same time other tools I mentioned before were showing CPU usage between 24 and 29.
This is 11.31 PA-RISC 8800 server.
The difference is significant. I wonder if anyone else noticed such problem? How to track the problem?
br,
Kamil
I'm using Zabbix 2.0 on few Linux & HP-UX hosts. Recently I noticed that zabbix agent shows different CPU utilization values than other tools ( glance/top/vmstat/measureware) on some HP-UX servers.
For example, right now on one of the HP-UX hosts system.cpu.util[] returns values from 64.907216 to 69.867061. In the same time other tools I mentioned before were showing CPU usage between 24 and 29.
This is 11.31 PA-RISC 8800 server.
The difference is significant. I wonder if anyone else noticed such problem? How to track the problem?
br,
Kamil