snmp Instance Identifiers for CPU and MEM?

[ricelchan]

When we snmp walk the following MIB OID:

CPU    - .1.3.6.1.4.1.9.9.109.1.1.1.1.5
MEM    - .1.3.6.1.4.1.9.9.48.1.1.1.5. and .1.3.6.1.4.1.9.9.48.1.1.1.6.

It return a instance value at the end of the output, sometimes it will return instance 1 to 7:

CPU:

snmpwalk R4_SW007 .1.3.6.1.4.1.9.9.109.1.1.1.1.5 (return instance 1)

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.1: Gauge32: 4

snmpwalk R4_SW021 .1.3.6.1.4.1.9.9.109.1.1.1.1.5 (return instance 1 - 7)

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.1 : Gauge32: 4

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.2 : Gauge32: 5

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.3 : Gauge32: 0

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.4 : Gauge32: 0

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.5 : Gauge32: 0

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.6 : Gauge32: 0

cisco.ciscoMgmt.ciscoProcessMIB.ciscoProcessMIBObjects.cpmCPU.cpmCPUTotalTable.cpmCPUTotalEntry.cpmCPUTotal5min.7 : Gauge32: 0

Memory:

snmpwalk R4_SW007 .1.3.6.1.4.1.9.9.48.1.1.1.5. (return instance 1 - 2)

cisco.ciscoMgmt.ciscoMemoryPoolMIB.ciscoMemoryPoolObjects.ciscoMemoryPoolTable.ciscoMemoryPoolEntry.ciscoMemoryPoolUsed.1 : Gauge32: 28413088

cisco.ciscoMgmt.ciscoMemoryPoolMIB.ciscoMemoryPoolObjects.ciscoMemoryPoolTable.ciscoMemoryPoolEntry.ciscoMemoryPoolUsed.2 : Gauge32: 9758696

I can't found any doc in cisco.com explain the meaning of this instance identifier. expert please help. 

Note: R4_SW007 is C3750 and R4_SW021 is C7609.

[phiharri]

Greetings,

In cpmCPUTotalTable, each entry has cpmCPUTotalPhysicalIndex (1.3.6.1.4.1.9.9.109.1.1.1.1.2) which will map the index in cpmCPUTotalTable to an index from entPhysicalTable where you can retrieve for example, entPhysicalName (1.3.6.1.2.1.47.1.1.1.1.7) describing which CPU you are looking at, example:

cpmCPUTotal5min values

$ snmpwalk -v2c -c public 10.48.91.71 1.3.6.1.4.1.9.9.109.1.1.1.1.5
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.5.1 = Gauge32: 1
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.5.2 = Gauge32: 23
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.5.3 = Gauge32: 1
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.5.4 = Gauge32: 3

The corresponding physical indexes

$ snmpwalk -v2c -c public 10.48.91.71 1.3.6.1.4.1.9.9.109.1.1.1.1.2
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.2.1 = INTEGER: 4017
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.2.2 = INTEGER: 4001
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.2.3 = INTEGER: 5007
SNMPv2-SMI::enterprises.9.9.109.1.1.1.1.2.4 = INTEGER: 2011

The description for these physical indexes

$ snmpget -v2c -c public 10.48.91.71 1.3.6.1.2.1.47.1.1.1.1.7.4017
SNMPv2-SMI::mib-2.47.1.1.1.1.7.4017 = STRING: "CPU of Routing Processor 6"
$ snmpget -v2c -c public 10.48.91.71 1.3.6.1.2.1.47.1.1.1.1.7.4001
SNMPv2-SMI::mib-2.47.1.1.1.1.7.4001 = STRING: "CPU of Switching Processor 6"
$ snmpget -v2c -c public 10.48.91.71 1.3.6.1.2.1.47.1.1.1.1.7.5007
SNMPv2-SMI::mib-2.47.1.1.1.1.7.5007 = STRING: "CPU of Sub-Module 9 CFC Card"
$ snmpget -v2c -c public 10.48.91.71 1.3.6.1.2.1.47.1.1.1.1.7.2011
SNMPv2-SMI::mib-2.47.1.1.1.1.7.2011 = STRING: "CPU of Sub-Module 3 DFC Card"

So above, the Switching Processor in slot 6 has a 5 min utilization value of 23%.

Memory utilization in ciscoMemoryPoolTable returns values for each memory pool, on the switches you mentioned the DRAM is divided into Processor (used to run IOS processes) and I/O (for packet buffering) pools when the device boots. You can check ciscoMemoryPoolName (1.3.6.1.4.1.9.9.48.1.1.1.2) to determine which pool the index refers to, example:

$ snmpwalk -v2c -c public 10.48.91.71 1.3.6.1.4.1.9.9.48.1.1.1.5

SNMPv2-SMI::enterprises.9.9.48.1.1.1.5.1 = Gauge32: 157069352

SNMPv2-SMI::enterprises.9.9.48.1.1.1.5.2 = Gauge32: 11891792

$ snmpwalk -v2c -c public 10.48.91.71 1.3.6.1.4.1.9.9.48.1.1.1.2

SNMPv2-SMI::enterprises.9.9.48.1.1.1.2.1 = STRING: "Processor"

SNMPv2-SMI::enterprises.9.9.48.1.1.1.2.2 = STRING: "I/O"

So the Processor pool has allocated around 150MB while I/O uses around 11MB. You will see similar output in 'show memory summary' on the device CLI.

Hope this helps!

/Phil

[ricelchangmail]

Thank your very much, your replied is extereme helpful.