Build precise queries to find exactly what you need
Press ESC to close
Your review has been submitted and is pending approval.
Plugin for monitoring Cisco UCS through SNMPv2 and SNMPv3. You don't need install any MIB for this plugin. It's using Net::SNMP PERL module. !!! Warnning !!! When you send SNMP request to virtual IP of UCS Manager you'll get reply from IP active Fabric Interconnect! Many SNMP implementation don't accept this behavior by default!!!
Current Version
2.1.0
Last Release Date
2014-04-04
Owner
Petr Havlicek
License
GPL
Compatible With
check_ucs version 2.1.0. by Petr Havlicek ([email protected]) 2012-2014
Nagios plugin for monitoring Cisco UCS over SNMP.
USAGE: -H HOST_IP -T TYPE -N OBJECT_NAME [-C COMMUNITY] [-u USERNAME] [-a MD5|SHA] [-A PASSPHRASE] [-x AES|DES] [-X PASSPHRASE]
Parameters: -H <HOST_IP> Target hostname -T <TEST> Selected test -N <NAME> Monitoring object name -C <COMMUNITY> SNMPv2 community string -u <USERNAME> SNMPv3 SecurityName -a <MD5|SHA> SNMPv3 authProtocol (Default: MD5) -A <PASSPHRASE> SNMPv3 authPassword -x <AES|DES> SNMPv3 privProtocol (Default: DES) -X <PASSPHRASE> SNMPv3 privKey
Test types: ct - Chassis Temperature ci - Chassis IOCard Status f - Fans Status po - PSUs Operate Status fs - Faults Summary (Dont need -N)
Fabric Interconnects support only these test: f, po Object name examples: switch, switch-A, chassis-1, chassis-10
Check command example: check_command check_ucs!public!ct!chassis-1 check_command check_ucs!public!fs
# History: ############
# Version 2.1.0 - 4th April 2014 # - Add support for SNMP version 3 # - Some cosmetic improvements
# Version 2.0.3 - 16th May 2013 # - Add support for greater OID # - Uknown status with 0 object detected
# Version 2.0.2 23rd January 2013 # - Minor bug fixes
# Version 2.0.1 - 22nd January 2013 # - Added some user input checks # - Object name comapring case insensitive
# Version 2.0 - 4th January 2013 # - Complete re-design of internal functionality # - Now first step is getting sub-tree # - Second step is getting descriptions for all OID in sub-tree # - Third is filter sub tree by specific description # - Last step is getting results state for specific OIDs
# Version 1.3 - 16th November 2012 # - Add support for checking faults
# Version 1.2 - 21st August 2012 # - Based on OID prefix for specific device => No manual OID specification # - Using manual prefix definitation # - Remove PSU Temperature test # - Bugfixes
# Version: 1.1.1 - 7th August 2012 # - Monitoring fan modules insteed individuals fans
# Version: 1.1 - 18th June 2012 # - Add support for prinitg description of monitored objects # - Minor bugfixes
If you get a no OID found this occurs if you have no alerts/warnings or information at all to report, the error kind of threw me for a loop to. I just commented that No OID found as long as I verified I was connecting correctly already. I also made a modification to the faults check to report warnings correctly to Nagios. Here is the modified section I did. if($type eq "fs") { if($results[$i] > 4) { $exit_state = NAGIOS_CRITICAL; $output .= " " . $descriptions[$i] . " "; } if($results[$i] == 4) { if ($exit_state ne NAGIOS_CRITICAL) { $exit_state = NAGIOS_WARNING; } $output .= " " . $descriptions[$i] . " "; } next; }
[nagios@nagios libexec]$ ./check_ucs.bin -H 10.10.10.1 -C public -T ci -N chassis-2 No OID match! Check your -H and -N or -T We have six chassis, only fail in 2 and 5,6. The state of all chassis is ok and we haven't any alarm.
./check_ucs -H 10.1.5.1 -C public -T ct -N chassis-1 No OID match! Check your -H and -N or -T
It's look interesting. Are you sure that IP and comunity are right? Have you enable SNMP in Fabric Interconnect managment? PetrH
The plugin is excellent, but i have detected a bug. When I check a different IP of the UCS, I must receive an error but I receive "0 objects OK".
You must be logged in to submit a review.
To:
From: