For more information regarding the security incident at F5, the actions we are taking to address it, and our ongoing efforts to protect our customers, click here.

Forum Discussion

cbenton210_2064's avatar
cbenton210_2064
Icon for Nimbostratus rankNimbostratus
Aug 11, 2016

NTP Health Monitor on V12

Does anyone know a way to setup a Valid NTP health monitor on V12. I have a virtual server setup with 3 Local NTP servers in a pool. Would like to know that there are actually serving time as appose...
application delivery
iRules
LTM
mimlo_61970's avatar
mimlo_61970
Icon for Cumulonimbus rankCumulonimbus
Aug 11, 2016

You'd have to create an external monitor of some kind. Something like

    !/bin/bash
    ip=${1}
    ntpdate -q ${ip}
    if [ $? -eq 0 ]
    then
            echo "UP"
    fi

Whatever you echo should be the RECV= variable for the monitor. See https://devcentral.f5.com/articles/ltm-external-monitors-the-basics for info on setting up an external monitor. Script above is very basic, not very good error handling, making sure it isn't already running and so on. Could use some work. The basic is ntpdate will return non-0 if it wasn't able to query the ntp server.

StephanManthey's avatar
StephanManthey
Icon for Nacreous rankNacreous
Apr 24, 2019

I like your ntpdate based approach (+1). It looks upgrade safe to me compared to the Perl based monitor which requires a library import. (It used to get lost with a software update.) By default the ntpdate runs 4 polls. A single poll should be enough. The number of polls can be controlled via the -p paramter followed by a numerical value. A stratum of 1-15 should be okay. Thats why my current solution looks as follows:

!/bin/sh
ntpdate -q -p 1 ${1} 2> /dev/null | grep -Ewo 'stratum\s+([1-9]|1[0-5])' > /dev/null
return_codes="${PIPESTATUS[*]}"

if [ "${return_codes}" == "0 0" ]
then
    echo "up"
fi