Writing Custom Plugin with String output

[chrisbooth]

I realized it wasn't the best idea to run a check within another check so i removed the check_by_ssh and just used ssh itself

Code: Select all

if ssh admin@"$HOSTNAME"  /usr/local/aw/bin/nsdchat -s 'awsock:/admin:Operation8@localhost:9001 -c Job failed 90' | grep -v empty > /dev/tty; then
#if ./check_by_ssh -H "$HOSTNAME" -l admin -C '/usr/local/aw/bin/nsdchat -s awsock:/admin:Operation8@localhost:9001 -c Job failed 90' | grep -v empty > /dev/tty; then
   echo "Syncs Failed"
   exit 2
else
   echo "OK"
   exit 0
fi

I've just realized no matter what i put in the host when i run a service check it responds with OK

Screen Shot 2016-12-07 at 22.27.56.png

other checks do fail so i'm obviously an issue specific to my plugin

Screen Shot 2016-12-07 at 22.31.48.png

[gormank]

For some reason my last post never got posted :(

Is the if reacting to the return value of the command, or to the text in the command?
You can make a test script that returns the value you tell it and figure things out before trying in real life. Call w/ 1/2/3/255

test.sh n

echo "Return value: $1"
exit $1

Something like the above...

Is " -c Job failed 90" part of the current check or an arg to the check_ plugin.
I'd think interms of the return value of the underlying command executed since it should theoretically be meaningful. Maybe not.

[chrisbooth]

ok i think i've found the issue

Code: Select all

if ssh admin@"$HOSTNAME"  /usr/local/aw/bin/nsdchat -s 'awsock:/admin:Operation8@localhost:9001 -c Job failed 90' | grep -v empty > /dev/tty; then

changed it from /dev/tty; to /dev/null;

/dev/tty; was my way of getting the output of /usr/local/aw/bin/nsdchat -s 'awsock:/admin:Operation8@localhost:9001 -c Job failed 90' and not just passing it onto the if statement

changed to:

Code: Select all

if ssh admin@"$HOSTNAME  /usr/local/aw/bin/nsdchat -s 'awsock:/admin:Operation8@localhost:9001 -c Job failed 90' | grep -v empty > /dev/null; then

the only reason i had /dev/tty; is becuase that way i get the codes of the failed jobs.

i'm not sure how i can use /dev/null; but also get that output

[gormank]

> /dev/null: redirect all output to nowhere. Not good.
2>&1: redirect STDERR to STDOUT
2>&1 > /dev/null: redirect STDERR to nowhere

Maybe the return status isn't meaningful, but it seems you're trying to react to return text rather than status. Why not try using the status in $?

[dwhitfield]

@gormank thanks again for your help!

@chrisbooth, I know this was already mentioned, but you might find quicker responses for the output/scripting questions at https://unix.stackexchange.com/.

[chrisbooth]

thanks for your help. i only posted here as i wasn't sure if it was a script issue or a nagios issue.

thanks again for your generous help.

[rkennedy]

Did you have any further questions about this?

[chrisbooth]

thanks so much. all this information helped

[tmcdonald]

I'll be closing this thread now, but feel free to open another if you need anything in the future!