1

We have a lot of IBM x86 servers that we need to monitor for hardware issues(disk failure/temperature/fans). Around 50% of them have IMM ports hooked up and we are able to monitor for SNMP traps using Nagios. In case of any hardware issues IMM will send a trap to the Nagios server IP and we get notified.

The issue here is that we have a lot of servers in various locations that we can't connect the IMM ports. Is there an agent that we can install on these servers to monitor for hardware failure? Something that does not need IMM port connectivity and will use the regular network Interface? The OS running on these servers are Windows and RHEL.

We could install Nagios agent(nsclient++ or NRPE) but as per the information I have received they can't read hardware information.

What are my options?

Debianuser
  • 421
  • 4
  • 12
  • 29

2 Answers2

0

You can use one of the various check_ipmi variants run via NRPE or check_by_ssh (at least in Linux) if you load the kernel IPMI module.

Start by loading (i.e., modprobe) ipmi_devintf and ipmi_si.

Some of the checks use ipmitools, others might use openipmi or freeipmi. You'll have to hunt around for one that meets your needs. (They may also be able to check hardware via the IMM using IPMI over LAN, too.)

I don't know what you can do for Windows, though.

Keith
  • 4,637
  • 15
  • 25
0

Just like Keith said, you can do this by loading the kernel IPMI. You can do this or via Nagios, but I recommend you try out Pandora FMS and their take on IPMI. Pandora FMS is a lesser known monitoring solution I've been testing and tinkering around with, and I'm starting to recommend it as an alternative to others.

  • Thanks, but we already have Nagios so need something that works with it. Any idea what I can use to read hardware information on Windows? – Debianuser May 04 '16 at 10:53