Q: HP-UX performance monitoring using sar and gnuplot - HP UX

This is a discussion on Q: HP-UX performance monitoring using sar and gnuplot - HP UX ; Hi! Does anybody have some scripts to convert the output of the HP-UX (11.31) sar command to gnuplot graphs? We have one machine that shows a severe performance problem (e.g. top takes over 10 seconds to display the first page), ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: Q: HP-UX performance monitoring using sar and gnuplot

  1. Q: HP-UX performance monitoring using sar and gnuplot

    Hi!

    Does anybody have some scripts to convert the output of the HP-UX (11.31) sar
    command to gnuplot graphs? We have one machine that shows a severe performance
    problem (e.g. top takes over 10 seconds to display the first page), and I'm
    absolutely clueless what the problem is. It's not CPU; it might be I/O; it may
    also be a kernel bug.

    As I'm tight on spare time, some tools to adjust would help me writing
    everything from scratch (I'd prefer using the sar/gnuplot approach to be able
    to compare the results with other systems (like Linux)).

    Regards,
    Ulrich

  2. Re: Q: HP-UX performance monitoring using sar and gnuplot

    Ulrich Windl wrote:
    > We have one machine that shows a severe performance
    > problem (e.g. top takes over 10 seconds to display the first page)


    Is this PA or IPF? If the latter, you can use free caliper on top.
    And of course there is glance for both.

  3. Re: Q: HP-UX performance monitoring using sar and gnuplot

    Dennis Handly writes:

    > Ulrich Windl wrote:
    >> We have one machine that shows a severe performance
    >> problem (e.g. top takes over 10 seconds to display the first page)

    >
    > Is this PA or IPF? If the latter, you can use free caliper on top.
    > And of course there is glance for both.


    Yes, I saw caliper as part of the 11.31 Base_OE (IA64), but I'm afraid it takes
    significant time to understand that tool set. Thanks for the hint.

    Ulrich

  4. Re: Q: HP-UX performance monitoring using sar and gnuplot

    Dennis Handly wrote:
    > Ulrich Windl wrote:
    > > We have one machine that shows a severe performance
    > > problem (e.g. top takes over 10 seconds to display the first page)


    > Is this PA or IPF? If the latter, you can use free caliper on top.
    > And of course there is glance for both.


    If PA, then one might use Prospect:

    http://www.hp.com/go/prospect

    rick jones
    --
    denial, anger, bargaining, depression, acceptance, rebirth...
    where do you want to be today?
    these opinions are mine, all mine; HP might not want them anyway...
    feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

  5. Re: Q: HP-UX performance monitoring using sar and gnuplot

    Ulrich Windl wrote:
    > Yes, I saw caliper as part of the 11.31 Base_OE (IA64),


    One can also I suspect follow pointers from:

    http://www.hp.com/go/caliper

    > but I'm afraid it takes significant time to understand that tool
    > set. Thanks for the hint.


    Modulo finding an "Up and running with Caliper in 30 seconds" writeup,
    you could try a tusc trace of a top command, with the options to do
    both syscall entry and exit tracing, with timestamps. That might
    yield some interesting tidbits.

    rick jones
    --
    oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
    these opinions are mine, all mine; HP might not want them anyway...
    feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

  6. Re: Q: HP-UX performance monitoring using sar and gnuplot

    Rick Jones wrote:
    > Ulrich Windl wrote:
    > > Yes, I saw caliper as part of the 11.31 Base_OE (IA64),

    > One can also I suspect follow pointers from:
    > http://www.hp.com/go/caliper


    I've been reminded there is a FAQ off that link above. One that might
    be of help for a first-time Caliper user is titled "I'm new to HP
    Caliper. What's the first measurement I should take to start?"

    http://h21007.www2.hp.com/portal/sit...5d6e10RCRD#new

    rick jones
    --
    No need to believe in either side, or any side. There is no cause.
    There's only yourself. The belief is in your own precision. - Jobert
    these opinions are mine, all mine; HP might not want them anyway...
    feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

+ Reply to Thread