Monitoring MegaRAID using StorCLI on Linux in an agentless environment

jm-es

Hi,

I am monitoring servers running Ubuntu systems using SNMPv3 with AVAGO 3108 MegaRAID cards in them,

The systems are not running the Agent, their hardware management is also not accessible. They are not Out of the Box supported hardware for monitoring purposes (Cisco UCS C series in stand-alone mode and SuperMicro units).

I believe that I cannot discover and monitor the RAID volumes without running the Agent, running the LSI LSA server tool or access to poll the hardware management.

Does that sound correct?

Broadcom seems to indicate that the controller is not exposed to the Linux operating system in a way that an SNMP MIB would reach.

So - that being said - I am thinking that I am going to need to use SAM and create a MegaRAID Application with component monitors.

StorCLI now supports output in JSON format, which makes it less difficult.

For example I know that if I want a quick response as to the health of a RAID volume, I can use this:

storcli64 /call/vall show J | jq --raw-output '.Controllers[0]["Response Data"]["Virtual Drives"][].State'

mesverrum

This all sounds about correct, the main trick is just to take that StorCLI output and reformat it into what SAM script monitors expect, which is a single line output that looks roughly like
Statistic: <somenumber>

and has an appropriate exit code based on whatever logic you want to implement

More detail about script monitors in this article

thwack.solarwinds.com/.../sam-script-component-monitors---everything-you-need-to-know