Graphing - NTP

This is a discussion on Graphing - NTP ; I've recently become a 'timekeeper'. Keeping a long story short, I'm wondering how I can graph NTPD requests. I've not done much graphing beyond Cacti and NTOP, however I'm wondering about RRD and graphing just NTP. Any info would be ...

+ Reply to Thread
Results 1 to 8 of 8

Thread: Graphing

  1. Graphing

    I've recently become a 'timekeeper'. Keeping a long story short, I'm
    wondering how I can graph NTPD requests. I've not done much graphing
    beyond Cacti and NTOP, however I'm wondering about RRD and graphing
    just NTP. Any info would be great.

    Jon
    epoch.linuxgeneration.org


  2. Re: Graphing

    Jon wrote:
    > I've recently become a 'timekeeper'. Keeping a long story short, I'm
    > wondering how I can graph NTPD requests. I've not done much graphing
    > beyond Cacti and NTOP, however I'm wondering about RRD and graphing
    > just NTP. Any info would be great.
    >
    > Jon
    > epoch.linuxgeneration.org


    This is what I use:

    http://www.david-taylor.myby.co.uk/ntp/NTPandMRTG.txt

    http://www.david-taylor.myby.co.uk/mrtg/daily_ntp.html

    but doubtless it could be improved. You may need different commands for
    your OS.

    David



  3. Re: Graphing

    Jon wrote:

    > I've recently become a 'timekeeper'. Keeping a long story short, I'm
    > wondering how I can graph NTPD requests. I've not done much graphing
    > beyond Cacti and NTOP, however I'm wondering about RRD and graphing
    > just NTP. Any info would be great.
    >
    > Jon
    > epoch.linuxgeneration.org
    >


    I've used GnuPlot. It's freely available and will build and run on just
    about anything and supports most output devices. It never occurred to
    me that "requests" could be graphed and, indeed, I'm not sure what you
    mean by it but if you can express it numerically, it can be graphed somehow.

  4. Re: Graphing

    Please check out http://ntp.isc.org/Support/MonitoringAndControllingNTP and
    add whatever new content you think could be useful.

    H

  5. Re: Graphing

    Great, thanks to everyone. I'll look in to each.


  6. Re: Graphing

    I also should have been more clear. I'm looking to graph everything
    mentioned by everyone who replied, but also bandwidth, and perhaps
    amount of connections. It's nothing serious, I'd just like to see
    how(much) the server is being used. More of an educational experiment.
    Thanks again.


  7. Re: Graphing

    On 2006-11-29, Jon wrote:

    > I also should have been more clear. I'm looking to graph everything
    > mentioned by everyone who replied, but also bandwidth, and perhaps
    > amount of connections. It's nothing serious, I'd just like to see
    > how(much) the server is being used. More of an educational experiment.


    The NTP Pool has monitoring scripts; you might want to contact them.

    You'll find lots of links to NTP monitoring systems through Google.
    Although most of them don't provide documentation about how they did it.

    Here's a few:

    http://www.cs.wisc.edu/cgi-bin/crick...t=%2Fntp-stats
    http://lx.ujf.cas.cz/ntp-lx/lx-pkt.html
    http://www.maths.tcd.ie/local/admin/...p/ntp-pps.html
    http://ntp.isc.org/bin/view/Support/...utimeRefclocks
    http://www.ntp.vanheusden.com/mrtg/ntp.html
    http://fud.no/munin/fud.no/lust.fud.no.html

    You'll also find links to various resources on
    http://ntp.isc.org/bin/view/Support/...ControllingNTP

    --
    Steve Kostecke
    NTP Public Services Project - http://ntp.isc.org/

  8. Re: Graphing

    In article <1164843194.090808.23400@j44g2000cwa.googlegroups.c om>,
    "Jon" writes:
    >I also should have been more clear. I'm looking to graph everything
    >mentioned by everyone who replied, but also bandwidth, and perhaps
    >amount of connections. It's nothing serious, I'd just like to see
    >how(much) the server is being used. More of an educational experiment.
    > Thanks again.


    ntpdc -c monlist
    will show you some info for the servers that are still
    in the servers cache. It isn't quite enough info to
    graph everything you might want to know, but it's a good
    start. More info in the documentation. Beware, the
    default cache size is something like 700 hosts so you
    will loose info if you have more clients that than unless
    you probe often enough.

    --
    These are my opinions, not necessarily my employer's. I hate spam.


+ Reply to Thread