monitoring SAS drives on a PE2950?

[Terry Gliedt]

Ryan Langseth wrote:
> 
> On Thu, 9 Oct 2008, Adam Williams wrote:
> 
>> Is there a way to monitor the drives on a PE2950 for failure?  I know
>> there is OMSA, but I have to check the web interface daily.  is there a
>> way to have a script check the drives daily and email me if one fails?
>>
> 
> I have been trying to figure this out on our linux systems too, I really 
> don't want to install OMSA to do it.  there are nagios plugins to monitor 
> megaraid controllers. On our FreeBSD systems, drive failures are logged by 
> the kernel, on CentOS, there are no log messages when a disk fails.

I monitor 70+ machines (Dell, IBM and Sun) every day with a simple 
script that dumps the hardware event logs of each machine. I keep each 
log cleared and filter out non-events (like 'log cleared). The remaining 
lines (if any) are sent me each morning as mail and look like this:

############ SEL Information for 'p35' ['192.168.12.24'] ############
Complete SEL log saved in '/data/sel/p35.log'
SEL can be cleared with:   ipmitool -I lan -U MYROOT -a -H 192.168.12.24 
sel clear
    2 | 09/25/2008 | 14:47:21 | Power Supply #0x64 | Predictive failure 
| Asserted
    3 | 09/25/2008 | 14:47:31 | Power Supply #0x64 | Predictive failure 
| Deasserted
    4 | 09/25/2008 | 15:18:06 | Power Supply #0x64 | Predictive failure 
| Asserted

The key is to use IPMI. See

   http://www.hps.com/~tpg/notebook/ipmi.php

for the core of what I needed to know. Hopefully that is enough for you too.