Hi Experts,
I want to check health of hp-ux box.
Basically I want to check if there are possibilities of network/memory/cpu bottleneck?
Are there are any commands available other than glance in hp-ux for the same?
Measureware (OVPA) is another one.
Hi Scrutinizer,
Sorry but I didn't understand what is the meaning measure ware(OVPA)? 
It is the name of the product, Measureware or OV Performance Agent.
Ok..thanks got it..
However my PL is suggesting me to use commands only to check the system health. So please suggest if there are commands available to check the same.
With commands, that would mean to use the standard unix tools: top, iostat, vmstat, netstat, nfsstat, etc... or write your own e.g. using pstat...
An sar . Very useful for maintaining a history of performance metrics when properly configured.
1 Like
Forgot sar, really getting old...
I totally disagree. Very useful and free.
Here's edited highlights of an incident earlier. User CPU% jumped from 7% to 29% and triggered an alert. Quick look with top and found a looping orphan process. Killed it.
sar -u # Edited output
00:00:01 %usr %sys %wio %idle
09:40:00 7 11 8 74
10:00:00 7 11 3 79
10:20:00 7 12 4 78
10:40:00 20 12 14 54
11:00:00 29 15 7 49
11:20:00 29 15 4 52
11:40:00 6 11 3 80
12:00:00 7 12 6 76
Can also analyse the various reports from sar for sizing reports.
Methyl: I ment Me... 
I like execute first of all
vmstat 1 50
and I have a quickly overview about if the system is swapping or has CPU and memory problems. If you detect something wrong, you can execute other commands to investigate more deeply.
@vbe
Doh! I get so used to reading strange grammar on this site I didn't read what you wrote.
@RuBiCK
I run vmstat from cron giving it a 2-minute burst every 20 mins and append the result to a day file (having ignored the first line output because is garbage on HP-UX). The memory statistics from vmstat are nice and easy to work with for Excel graphs or writing crude alerts.