Best Practises for UCS Monitoring

[narenblr]

Hi,

What are the current methods that have been followed to monitor UCS.

I know there are third party agents/modules such as SCOM Management Pack etc.

Howeer, are there any best practises / suggestions for using SNMP as a primary protocol for monitoring UCS.

Thanks,

Naren

[kg6itcraig]

I have used the standard Cisco template in Cacti, and it will show the connections to FI and IOM. That is fairly useful.

Have not had the patience to lookup useful OID’s from the MIB’s. If someone found the proper OID’s for vHBA’s and vNIC’s and other useful into I would be happy to update and share a “Good” Cacti template for UCS. I consider Cacti a “dead project” as far as further development but if you have a good template for something no reason not to use it. Free.

Zenoss has some monitoring plugins for UCS, but in my opinion they are crap. Just shows what you already know.

Is there really a SCOM plugin? If it could show performance statistics it might be useful.

For alerts like “critical” etc, you can enable call home. You can create monitoring profiles that can be pretty detailed.

Craig

[Paul O'Leary]

Cisco UCS Pack for SCOM R2:

http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/scom/quick/start/guide/ucsMPQS.html

The Cisco UCS SCOM (System Center Operations Manager) R2 Management Pack is a plug-in for SCOM 2007 R2, installed on a SCOM server. It is used to monitor the health of the UCS system in the data center. With this plug-in, you can monitor chassis, blades, and service profiles across multiple UCS systems. Additionally, the Cisco UCS SCOM management pack enables correlation of faults and events between the Cisco UCS infrastructure, and both, bare-metal and virtualized operating systems already managed by SCOM.