Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
executable file 206 lines (152 sloc) 6.93 KB

This plugin checks the health of HP servers with iLo 4 interface via SNMP.

The following components are checked:

  • status of powersupply
  • status of physical and logical drives (incl. drive temperatures)
  • status of fans
  • status of the temperature sensors
  • the power state of the server (On or Off)
  • status of the memory
  • the global health status

The plugin currently does not provide performance data.

For older systems we use the plugin. In our opinion the ilo2 plugin does have some disadvantages:

  • a username and password for the iLo interface must be provided in the configuration
  • the ilo2 plugin is using the xml interface of the ilo interface. The xml file is very large and that causes a lot of network traffic, and a high load on our icinga server. Additonally we had some crashes and timeouts of our old ilo2 interfaces, due to a too fast monitoring interval.

With iLo 4 HP introduced a real SNMP interface for iLo. The SNMP interface is able to get all relevant status information from the server. Including the hard drive status (was missing older iLOs)

Tested with:

  • HP DL380 Gen8
  • HP DL380 Gen9
  • HP DL360p Gen8
  • HP DL320e Gen8 v2

The plugin requires pynag and python-netsnmp.

Options for

  -h, --help            show this help message and exit
  -H HOSTNAME, --hostname=HOSTNAME
                        Hostname or ip address
                        SNMP community of the SNMP service on target host.
                        SNMP version. (1 or 2)
  --retries=RETRIES     Number of SNMP retries.
  -U SECNAME, --securityname=SECNAME
                        SNMPv3: security name (e.g. bert)
  -L SECLEVEL, --securitylevel=SECLEVEL
                        SNMPv3: security level (noAuthNoPriv, authNoPriv,
  -a AUTHPROTO, --authprotocol=AUTHPROTO
                        SNMPv3: authentication protocol (MD5|SHA)
  -A AUTHPASS, --authpass=AUTHPASS
                        SNMPv3: authentication protocol pass phrase
  -x PRIVPROTO, --privproto=PRIVPROTO
                        SNMPv3: privacy protocol (DES|AES)
  -X PRIVPASS, --privpass=PRIVPASS
                        SNMPv3: privacy protocol pass phrase
  --drives=AMOUNT_DRVS  Amount of physical drives
  --ps=AMOUNT_PWR_SPLY  Amount of connected power supplies
  --fan=AMOUNT_FANS     Amount of fans
  --scan                (obsolete)
  --noStorage           Do not check global storage condition
  --noSystem            Do not check global system state
  --noPowerSupply       Do not check global power supply condition
  --noPowerState        Do not check power state
  --noTemp              Do not check Overall thermal environment condition
  --noTempSens          Do not check temperature sensor condition
  --noDriveTemp         Do not check temperature sensor of the hard drives
  --noFan               Do not check global fan condition
  --noMemory            Do not check memory condition
  --noController        Do not check controller condition
  --noLogicalDrives     Do not check the logical drives
  --noPowerRedundancy   Do not check powersupply redundancy

  Generic Options:
    --timeout=50        Exit plugin with unknown status after x seconds
    --threshold=range   Thresholds in standard nagios threshold format
    --th=range          Same as --threshold
    --extra-opts=@file  Read options from an ini file. See
    -d, --debug         Print debug info

  Display Options:
    -v, --verbose       Print more verbose info
    --no-perfdata       Dont show any performance data
    --no-longoutput     Hide longoutput from the plugin output (i.e. only
                        display first line of the output)
    --no-summary        Hide summary from plugin output
    --get-metrics       Print all available metrics and exit (can be combined
                        with --verbose)
    --legacy            Deprecated, do not use



-C, --community         "public"
-V, --snmpversion       2


Normal check:

./ -H --fan 2 --ps 1 --drives 4

In this case we excpect that the server does have two fans, one power supply and four hard drives. The script requires that you enter the correct amount of components. We decided to rely not only on the status of the component, we also want to count the components. We had one server where a fan was completely dead. So it even did not report a degraded status. It was just completely missing.

OK - HP-Server0815  - Serial number:AB1234567C
Global storage status: ok

Global system status: ok

Global power supply status: ok

Server power status: poweredOn

Overall thermal environment status: ok

Temperature sensors status: ok

Fan(s) status: ok

Memory status: ok

Controller 0 status: ok
Controller 1 status: ok

Physical drive 0: ok
Physical drive 1: ok

Temperature of physical drive 0: 35 Celsius (threshold: 60 Celsius)
Temperature of physical drive 1: 33 Celsius (threshold: 60 Celsius)

Logical drive 0 status: ok

Power supply 0: ok

Temperature 0: 22 Celsius (threshold: 42 Celsius)
Temperature 1: 40 Celsius (threshold: 70 Celsius)
Temperature 2: 27 Celsius (threshold: 87 Celsius)
Temperature 3: 35 Celsius (threshold: 60 Celsius)
Temperature 4: 48 Celsius (threshold: 105 Celsius)
Temperature 5: 36 Celsius (threshold: 115 Celsius)
Temperature 6: 28 Celsius (threshold: 65 Celsius)
Temperature 7: 52 Celsius (threshold: 77 Celsius)
Temperature 8: 53 Celsius (threshold: 100 Celsius)
Temperature 9: 42 Celsius (threshold: 68 Celsius)
Temperature 10: 86 Celsius (threshold: 100 Celsius)
Temperature 11: 41 Celsius (threshold: 69 Celsius)
Temperature 12: 44 Celsius (threshold: 73 Celsius)
Temperature 13: 33 Celsius (threshold: 68 Celsius)
Temperature 14: 42 Celsius (threshold: 67 Celsius)

Fan 0: ok
Fan 1: ok
Fan 2: ok

Disable several checks

./ -H --fan 0 --ps 0 --drives 0 --noSystem --noPowerSupply

In this case we don't check the fans, the power supply, the drives, the global system status and the global power supply status.

We added these parameters because some of out HP DL380 G9 report a degraded power supply status, but everything is fine. In this case only one power supply is installed.


./ -H --scan 

With the scan option you can check how many components are found.

Unknown - HP-Server0815  - Serial number:AB1234567C. This is not a health status!
Physical drive 0: ok
Physical drive 1: ok

Power supply 0: ok

Fan 0: ok
Fan 1: ok
Fan 2: ok


  • Implement SNMPv3
  • Code cleanup
You can’t perform that action at this time.