Re: hp server hardware monitoring

On Wed, Jul 30, 2014 at 9:50 PM, Bonno Bloksma <b.bloksma@tio.nl> wrote:

Hi,

>> [...]
>> What may be relevant too is that on the g6 server Debian uses the
>> CCISS drivers for the raid hardware, the volume shows up as
>> /dev/cciss/c0d0
>> On the g7 and g8 hardware the raid volume simply shows up as /dev/sda
>
> cciss has been superseded by hpsa, "The hpsa driver is intended to supplant the cciss driver for newer Smart Array controllers.", cf.
> <https://www.kernel.org/doc/Documentation/scsi/hpsa.txt>.

Ok, thanks for the heads up.

>> How can I get this to work under a g7 or g8 server. Do I need a newer
>> version of the package of do I need a different package?
>
> I generally avoid installing third-party tools on the bulk of my servers, and for simple monitoring the packages cciss-vol-status and nagios-plugins-standard from Debian will suffice:

I will try that too. But....

>root@vz02:~# lsmod | grep -e cciss -e hpsa
>hpsa 50787 2

Ok, I have that too
linein:~# lsmod | grep -e hpsa -e cciss
hpsa 40765 2
scsi_mod 162269 5 hpsa,libata,sd_mod,sg,sr_mod

> root@vz02:~# cciss_vol_status /dev/sda
> /dev/sda: (Smart Array P420i) RAID 5 Volume 0 status: OK.

But I get...
linein:~# cciss_vol_status /dev/sda
cciss_vol_status: /dev/sda: Unknown SCSI device.

Which is weird because I have (copy from iLO):
Model: HP Smart Array P420i Controller
Firmware: Version 3.22

Now what?

Bonno Bloksma

Did you give the plugin check_ilo2_health.pl a shot?

The plugin uses ILO to get the status of the hardware. It works fine for servers running with at least ILO2.
Everything you see in ILO is exported and the plugin checks the state. On some older server generations, some hardware parts were missing in ILO (e.g. disks) but in recent servers/ILO versions (G7 and Gen8) the disks are also being checked in ILO.

See https://www.monitoringexchange.org/inventory/Check-Plugins/Hardware/Server/HP-%2528Compaq%2529/check_ilo2_health

I'm monitoring the hardware of 168 HP servers with this plugin.