I know this is slightly off-topic, but since it's still related to the
list, I thought I'd give it a try. I am wondering what systems are out
there (open source, preferably) for data collection and processing of
hardware health data (temperature, CPU clock, fan speeds, etc). Ideally
brand agnostic and location agnostic as well.
I know of Cacti, but it would require SNMP enabled devices AFAIK, so
room/generator/misc monitors wouldn't necessarily be included.
Just for getting your feet wet and doing so on a (tiny) budget..... If you want to monitor non-SNMP devices such as things like room temp probes, water leak detection, generator/ats/ups alarm outputs, etc . You could look into something like the APC AP9340 units
These support APC's own temp/humidity probes, various user input, modbus rs-485 port, etc.
Even cheaper, but a little more DYI, you can look into building a small Linux box, load MRTG (which you should be running anyway), and crafting small probe scripts that would feed the "traffic" grapher. For switch closures like on water-sensors, you will need an I/O board, but they are readily available and pretty easy to script.
For temperature/voltage alarms, those same scripts can send alarm e-mail when particular values fall outside of the range. Ditto switch sensing.
Also, there are SNMP-based solutions you may not have thought of. Have Cisco routers? The environmental sensors are available via SNMP.
We're using PRTG from Paessler (http://www.paessler.com). We're monitoring > 50k sensors (storage, network, hardware, applications, a/c, generators, door locks, liquid detection system in datacentres, etc) ... Best decision ever!
This is a product designed for use on Windows only, no mention of ports to other operating systems. For some people, this is fine. For others, who don't want to mess with Windows at all, it's a concern.
Looking at some of the product sheets, it looks boss at what it does. In particular, the "David Letterman" view is an interesting quick snapshot look at what is going on.
This is a product designed for use on Windows only,
No. The monitoring itself requires windows as OS but only for the Mgmt service, DB service, etc. You do not need a client (like for Nagios/etc) to monitor other systems. You simply monitor devices via http (e.g. APIs, etc) or SNMP, etc. You can also integrate other coding languages (Perl, PHP, C++, etc) if you need something unsupported.
From my POV PRTG is nearly a 100% solution and you do not need much more tools to get an good view of your running inventory. Beside PRTG we're only running some tools for flow analysis, NetApp storage analysis, etc. In sum we're running <5 tools to monitor EVERYTHING (hardware, software, datacentre infra, etc).
I know this is slightly off-topic, but since it's still related to the
list, I thought I'd give it a try. I am wondering what systems are out
there (open source, preferably) for data collection and processing of
hardware health data (temperature, CPU clock, fan speeds, etc).
Ideally brand agnostic and location agnostic as well.
I know of Cacti, but it would require SNMP enabled devices AFAIK, so
room/generator/misc monitors wouldn't necessarily be included.
You're going to find that the most commonly recommended solution, I think,
will be proxy SNMP, and let your SNMP monitor log it; there are *lots* of
reasons not to want to run two infrastructures for that.
