[network::cisco::standard::snmp::plugin] Monitoring switches on the stack (cpu, memory, uptime)

[Aleksey-Maksimov]

Hello.

In the Cisco stack, all switches work as a single unit and have one IP address for management and monitoring. We connected a monitoring stack of 4 Cisco 2960X switches. And now it is not clear how to interpret information in some modes in the plugin.

For example, some plug-in modes (environment, memory-flash, interfaces) correctly display information on all switches in the stack.

# /centreon_plugins.pl --plugin=network::cisco::standard::snmp::plugin 
--mode=environment --no-component --verbose
--change-perfdata='( RED | YELLOW | GREEN ),_' 
--change-perfdata='\.,_'
--filter-perfdata='^(?!count_)' --filter-uom='%' 
--hostname='SW40' --snmp-community='public' --snmp-version='2c'

OK: All 16 components are ok [4/4 fans, 4/4 modules, 4/4 psus, 4/4 temperatures]. | 
'temp_SW#1, Sensor#1,_'=34;;~:68;; 
'temp_SW#2, Sensor#1,_'=34;;~:68;; 
'temp_SW#3, Sensor#1,_'=34;;~:68;; 
'temp_SW#4, Sensor#1,_'=36;;~:68;;
Environment type: c37xx
Checking fans
fan 'Switch#1, Fan#1' status is normal [instance: 1007].
fan 'Switch#2, Fan#1' status is normal [instance: 2007].
fan 'Switch#3, Fan#1' status is normal [instance: 3007].
fan 'Switch#4, Fan#1' status is normal [instance: 4007].
Checking power supplies
Power supply 'Sw1, PS1 Normal, 
RPS Normal' status is normal [instance: 1006] [source: ac]
Power supply 'Sw2, PS1 Normal, 
RPS Normal' status is normal [instance: 2006] [source: ac]
Power supply 'Sw3, PS1 Normal, 
RPS Normal' status is normal [instance: 3006] [source: ac]
Power supply 'Sw4, PS1 Normal, 
RPS Normal' status is normal [instance: 4006] [source: ac]
Checking temperatures
Temperature 'SW#1, Sensor#1, 
GREEN ' status is normal [instance: 1008] [value: 34 C]
Temperature 'SW#2, Sensor#1, 
GREEN ' status is normal [instance: 2008] [value: 34 C]
Temperature 'SW#3, Sensor#1, 
GREEN ' status is normal [instance: 3008] [value: 34 C]
Temperature 'SW#4, Sensor#1, 
GREEN ' status is normal [instance: 4008] [value: 36 C]
Checking voltages
Checking modules
Module 'Stacking Module' status is ok [instance: 1063]
Module 'Stacking Module' status is ok [instance: 2067]
Module 'Stacking Module' status is ok [instance: 3063]
Module 'Stacking Module' status is ok [instance: 4063]
Checking physicals
Checking sensors

But there are plugin modes (cpu, memory, uptime) where information is only shown from the MASTER switch (?), but there is no information about the status of other switches in the stack.

# ./centreon_plugins.pl --plugin=network::cisco::standard::snmp::plugin 
--mode=cpu 
--snmp-version='2c' --snmp-community='public' --hostname='SW40' 
--warning-average-5m=90 --critical-average-5m=98

OK: 1 CPU(s) average usage is 34.00 % (5s) 34.00 % (1m) 35.00 % (5m) 
- CPU '1' usage 34.00 % (5s) 34.00 % (1m) 35.00 % (5m) | 
'total_cpu_5s_avg'=34.00%;;;0;100 
'total_cpu_1m_avg'=34.00%;;;0;100 
'total_cpu_5m_avg'=35.00%;0:90;0:98;0;100 
'cpu_5s'=34.00%;;;0;100 
'cpu_1m'=34.00%;;;0;100 
'cpu_5m'=35.00%;;;0;100

[garnier-quentin]

Could you provide a full snmpwalk ? I will try to find OIDs about cpu stack (if it exists)

[garnier-quentin]

Could you provide a snmpwalk for following OIDS ? .1.3.6.1.2.1.47.1.1.1 .1.3.6.1.4.1.9.9.109.1.1.1

[garnier-quentin]

For old catalyst, seems there is a specific community syntax: https://www.cisco.com/c/en/us/support/docs/ip/simple-network-management-protocol-snmp/40367-camsnmp40367.html

[Aleksey-Maksimov]

For old catalyst, seems there is a specific community syntax: https://www.cisco.com/c/en/us/support/docs/ip/simple-network-management-protocol-snmp/40367-camsnmp40367.html

I do not quite understand what this means.

[garnier-quentin]

Maybe it's not relevant for stack. I could look with your snmpwalk

[Aleksey-Maksimov]

Could you provide a snmpwalk for following OIDS ? .1.3.6.1.2.1.47.1.1.1 .1.3.6.1.4.1.9.9.109.1.1.1

Just sent you an email

[Aleksey-Maksimov]

Hello Quentin, Can I help you with something else to solve this problem?

[garnier-quentin]

It should be maybe: cpmCPUTotalPhysicalIndex = .1.3.6.1.4.1.9.9.109.1.1.1.1.2

But in your snmpwalk, you have only one entry: .1.3.6.1.4.1.9.9.109.1.1.1.1.2.1 = INTEGER: 1001

It's a stacked switch ?

[Aleksey-Maksimov]

Yes, stacked (stack of 4 Cisco 2960X switches) If necessary, I can try to unload snmpwalk again

[garnier-quentin]

You can try to get .1.3.6.1.4.1.9.9.109.1.1.1.1.2. Could you ask to Cisco support ?

[Aleksey-Maksimov]

Hello @garnier-quentin

Unfortunately, we do not have a service contract to support these switches, so cisco technical support will not help us here. My colleague found a link to more information:

https://www.cisco.com/c/en/us/support/docs/ip/simple-network-management-protocol-snmp/15215-collect-cpu-util-snmp.html#multiple

Maybe this will help us somehow?

[garnier-quentin]

I already check OIDs from your link.

[fmattesct]

Hi, Thanks for your interest in Centreon. Requests for new features and enhancements must be suggested here. Troubleshooting and questions must now be asked here (cf our new issue template.

Thank you for your understanding.