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

rraver_79489's avatar
rraver_79489
Icon for Nimbostratus rankNimbostratus
May 08, 2009

External health monitor works when ran from CLI but doesn't when applied to profile

I have the below health monitor that works when I run it in the command line, an UP statement gets put to the screen but doesn't when I apply it to the profile. I have tried it with both the static IP and the variable IP.

 

 

IP=`echo ${1} | sed 's/::ffff://'`

 

IP="10.2.3.10"

 

PORT=${2}

 

URI="/acct/ping.jsp"

 

 

send request & look for expected respons

 

STATUS=$(curl -kINfs https://${IP}:${PORT}$URI | grep -i "200 OK")

 

STATUS=$(curl -kINfs https://10.2.3.12:443/acct/ping.jsp | grep -i "200 OK")

 

 

mark node up if response is HTTP STATUS 200

 

if [ "$STATUS"=="HTTP/1.1 200 OK" ]

 

then

 

echo "UP"

 

fi

 

 

 

any ideas or suggestions? It's to check the page HTTP status code and verify that it's code 200.
management
monitoring

4 Replies

  • The_Bhattman's avatar
    The_Bhattman
    Icon for Nimbostratus rankNimbostratus
    May 10, 2009
    Why didn't you choose to use the built-in http monitor. If you are looking for a status code then it can be easily obtained with that.

     

     

     

    CB

     

  • Yossi_Attia_100's avatar
    Yossi_Attia_100
    Icon for Nimbostratus rankNimbostratus
    May 20, 2009
    Hey,

     

     

    did you manage to use the UP response in order to correctly monitor it?

     

     

    Thanks

     

    Yosi
  • hoolio's avatar
    hoolio
    Icon for Cirrostratus rankCirrostratus
    May 27, 2009
    Any standard output from an external script will be interpreted as a positive response for the member state, so the case (or the content) of the "up" message shouldn't matter.

     

     

    It might help to log debug messages to a log file. Here are some quick notes for adding debug to a shell script with both standard error and standard out being redirected to the log file.

     

     

    Based on info from: http://tldp.org/LDP/abs/html/io-redirection.html

     

     

    DEBUG='1'

     

    LOGFILE="/var/log/${0}.log"

     

     

    If debug is enabled, log the date to a log file. Redirect standard error and out to the log file.

     

    if [ ${DEBUG} -eq 1 ]; then echo "`date +"%Y-%m-%d %H:%M:%S"`: Debug enabled" >> ${LOGFILE} 2>&1; fi

     

     

    Aaron