Thomas Sluyter wrote: > Yay, debugging! \o/ > > Could be handy if we move this off-list, unless no-one objects to us > handling this through the list. > > >>> bha! end of the script now looks like this >>> >>> ### FINALLY... RETRIEVING THE VALUES ### >>> >>> EXITCODE=`$SNMPWALK -v 1 -c $COMMUNITY $HOST $STATUS` >>> [ $DEBUG -gt 0 ] && echo "Retrieve exit code is $EXITCODE" >> >>> $DEBUGFILE >>> >>> OUTPUT=`$SNMPWALK -Ovq -v 1 -c $COMMUNITY $HOST $STRING` >>> [ $DEBUG -gt 0 ] && echo "Retrieve status message is: $OUTPUT" >> >>> $DEBUGFILE >>> >>> if [ $EXITCODE -eq 0 ]; then >>> echo "$OUTPUT" | grep -v CRITICAL >>> [ $? -eq 1 ] && EXITCODE=2 >>> fi >>> >>> echo $OUTPUT >>> exit $EXITCODE >>> >>> but when it runs i get this >>> >>> ../libexec/retrieve_custom_snmp: line 84: [: -eq: unary operator >>> expected >>> "CRITICAL emcpowera has 2 dead, 2 alive" >>> >> Urgh. My bad. What does EXITCODE contain when you run it in debug >> mode? > > It should contain the exact same value it contains when not running > in debug mode: the exit code for the Nagios script being run through > SNMP exec. >
Yes, ofc, or we'd have a Heisenbug :) What I meant was that it doesn't seem to be a numerical value. If I look at it I can see that it may very well be a textual representation, so perhaps we just have a clash in nomenclature. > Could you tell me what exactly is contained on line 84 in your case? > Because in the version of the script up on my website, line 84 is > part of the Nagios "help" text. > > I think I -should- look into the whole "-Ovq" bit a little bit > though. I think the $EXITCODE may benefit from this as well... > Ah, that'd be it. I'm guessing the current output is expected to look something like this: oid.some.more.of.the.oid = INTEGER: $REAL_EXITCODE which is indeed numerical, but with a lot of junk in front of it. -Ovq would indeed help for this. > If I find some time this afternoon, I'll start digging... > > >> (on a side-note, whoever names a variable containing textual output >> "EXITCODE" should be shot) > > Yes, that would be me thank you... What kind of gun would you prefer? > I would like to smoke a pipe before being executed, though. Mind? > Not at all. If you come to sweden I'll buy you a whiskey to go along with that smoke. A water squirt gun to put it out when you're done sounds about right ;-) > But before you pull that trigger, would you please go back just a wee > a bit and read my docu on what $EXITCODE contains? Why yes, it -does- > in fact contain the exit code of the Nagios script that was run on > the client side! OMG! Let me quote the bit in question: > > # Output: > # The exit code is the exit code retrieved from OID.100.0. > # The output string is the string retrieved from OID.101.1. > > /me puts on a blind fold. > > Any suggestions for a variable name change are more than welcome. And > I guess I -could- elaborate the docu a bit more. Expect a newer > version, some time this afternoon. > EXITCODE works fine for me. As for the docu-change, it shouldn't be necessary assuming either -Ovq or some sed magic removes all the cruft that's *not*, in fact, the exit-code. >>> ho hum - enjoy the beer >>> >> Good plan. Working on it ;-) > > Right about now, I'm wishing I had a nice, cold bottle of "Lambiek > Kriek" in my desk drawer. > <offtopic> Never tried that. Where's it from? </offtopic> -- Andreas Ericsson [EMAIL PROTECTED] OP5 AB www.op5.se Tel: +46 8-230225 Fax: +46 8-230231 ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Nagios-users mailing list Nagios-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
