I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a shame.
I am wondering whether accessing it from within the OS is the best option or whether writing something that will talk to it over the ssh console makes the most sense.
What’s everyone else doing these days? I’m looking to target things like unplugged PSUs, failed disks etc.
I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a shame.
I am wondering whether accessing it from within the OS is the best option or whether writing something that will talk to it over the ssh console makes the most sense.
What’s everyone else doing these days? I’m looking to target things like unplugged PSUs, failed disks etc.
I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a shame.
I am wondering whether accessing it from within the OS is the best option or whether writing something that will talk to it over the ssh console makes the most sense.
What’s everyone else doing these days? I’m looking to target things like unplugged PSUs, failed disks etc.
By default, the iLO's SNMP is set to passthrough, which requires you to run HP's SNMP agents on the host to collect data. You won't get any data via SNMP on iLO unless they are running.
However, a new feature in iLO 4 is agentless management. This means iLO can pull data from the sensors, hp raid arrays, etc directly, and you can poll iLO via SNMP without involving the OS at all. You'll have to disable passthrough and enable agentless management via the web interface or hponcfg. For the latter, I believe this will do the job:
I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a
shame.
By default, the iLO’s SNMP is set to passthrough, which requires you to
run HP’s SNMP agents on the host to collect data. You won’t get any data
via SNMP on iLO unless they are running.
However, a new feature in iLO 4 is agentless management. This means iLO
can pull data from the sensors, hp raid arrays, etc directly, and you can
poll iLO via SNMP without involving the OS at all. You’ll have to disable
passthrough and enable agentless management via the web interface or
hponcfg. For the latter, I believe this will do the job:
Unfortunately, this only works on iLO 4, so you’ll still have to run
HP’s SNMP agents on your ilO 3 hosts if you want to collect host data via
iLO.
I went with /usr/local/bin/check_ipmi_wrapper for historic reasons. $OMD_ROOT/local/bin/ would work as well I guess.
I created a new Agent Type, and a rule under Datasource Programs -> Individual program call instead of agent access. I have a Rule for IPMI 1.x
and one for IPMI 2.0. For 2.0 the command is:
/usr/local/bin/check_ipmi_wrapper user password --driver-type=LAN_2_0
I’ve modified the script to take a 4th parameter. Depending on your version of ipmi-sensors you might have to omit --ignore-not-available-sensors
–legacy-output in the original scripts. I’ve you don’t get any output in cmk, try running ipmi-sensors directly and see what you get.
I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a shame.
I am wondering whether accessing it from within the OS is the best option or whether writing something that will talk to it over the ssh console makes the most sense.
What’s everyone else doing these days? I’m looking to target things like unplugged PSUs, failed disks etc.
I haven’t done the array part yet to be honest. It looks like it’ll need a separate check. However, just installing hp-health and hp-snmp-agents, gets me temperature, cpu, memory, fan, raid controller, and physical disk monitoring via SNMP.
For Windows, you just need to run through the installer on the HP Support Pack (or whatever they’re calling it these days).
I’m just wondering what the current best way of monitoring HP ILO3/4 is?
It can send SNMP alerts but from what I can see it’s not walkable which is a shame.
I am wondering whether accessing it from within the OS is the best option or whether writing something that will talk to it over the ssh console makes the most sense.
What’s everyone else doing these days? I’m looking to target things like unplugged PSUs, failed disks etc.
I haven’t done the array part yet to be honest. It looks like it’ll need a separate check. However, just installing hp-health and hp-snmp-agents, gets me temperature, cpu, memory, fan, raid controller, and physical disk monitoring via SNMP.
For Windows, you just need to run through the installer on the HP Support Pack (or whatever they’re calling it these days).