[c-nsp] Monitoring Nexus 7000 platform

[Lincoln Dale]

hi Ash,

On 14/08/2009, at 7:01 AM, Ash Net wrote:

> Yep, we know that already. I'm finding that there isn't a lot of
> management systems (OV/Concord atleast) that can natively monitor the
> 7k's since they haven't certified the platform yet.

actually, there is quite a lot of management partners that support  
N7K / NX-OS.

a non exhaustive list is:
   HP Opsware NAS (Configuration and Compliance
   EMC Smarts (Fault Management)
   IBM Tivoli Netcool (Fault Management)
   CA Spectrum  (Fault Management)
   CA eHealth (Network Perf. Monitoring)
   SolarWinds Orion [Network Performance Monitor , Network  
Configuration Manager, NetFlow Traffic Analyzer (Network Perf.  
Monitoring)
   Alterpoint [Network Authority] (Config. and Compliance
   BMC BladeLogic for Networks (BCAN) (Compliance)
   CiscoWorks Lan Management Solution (LMS) (General Purpose)
   CiscoWorks Network Compliance Manager (NCM) (Compliance)
   Cisco Network Analysis Module (NAM)  (Traffic and Flow Analysis)

there is no doubt a more complete list, the above is what i explicitly  
know about.
in addition to the above, there are numerous MIBs, SNMP traps and  
Netflow v5/v9 that all sorts of 3rd party management and monitoring  
systems plug in to


>
>
> Wondering how people are monitoring elements such as CPU Health, intf
> utilization, topology change event traps of the 7K Chassis etc. There
> doesn't appear to be a comprehensive MIB that has all the elements
> defined.

see http://ftp-sj.cisco.com/pub/mibs/supportlists/nexus7000/Nexus7000MIBSupportList.html 
  for a list of MIBs.

we do also have a list of 'key performance indicators' that best  
practice would say that you poll for.

e.g.:
MIB: CISCO-PROCESS-MIB
OID: cpmCPUTotal5minRev
Loc: .1.3.6.1.4.1.9.9.109.1.1.1.1.8
Range: 0..100 (%)
Desc: The overall CPU busy percentage in the last 5 minute period.
Normal operating range: Value should remain below 80% normal conditions.
Poll interval: once every 5 minutes

MIB: CISCO-SYSTEM-EXT-MIB
OID: cseSysMemoryUtilization
Loc: .1.3.6.1.4.1.9.9.305.1.1.2
Range: 0..100 (%)
Desc: The average utilization of memory on the active supervisor.
Thresholds for RMON should probes should be set based on baselining  
the memory utilization within a production environment
Poll interval: once every 5-15 minutes

can provide you with the complete list if you wish (quite long).  ping  
me off list if you want that.


interface utilization is provided via standard IF-MIB.
suggest you use high speed counters, e.g. ifHCInOctets, ifHCOutOctets,  
ifHCInUcastPkts, ifHCOutUcastPkts, etc.


cheers,

lincoln.