[ZBXNEXT-8909] Thresholds for some metrics on Cisco Catalyst 4500 (and others) are too low Created: 2023 Dec 18  Updated: 2023 Dec 22

Status: Confirmed
Project: ZABBIX FEATURE REQUESTS
Component/s: Templates (T)
Affects Version/s: 7.0.0alpha9
Fix Version/s: None

Type: Change Request Priority: Trivial
Reporter: Joan Assignee: Aleksandrs Petrovs-Gavrilovs
Resolution: Unresolved Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Tested on 6.4.7 but template has no changes on this aspect on current master

Attachments: Text File cisco_walk.txt    

 Description   

On Cisco IOS by SNMP template, there are three temperature triggers

        - '{$TEMP_CRIT}' value: '60'
        - '{$TEMP_CRIT:"CPU"}' value: '75'
        - '{$TEMP_CRIT_LOW}' value: '5'

But the sensors on a properly functioning system report higher thresholds for some of the values (values on column temperature between parentheses are for minor,major or critical values)

{{#sh environment temperature  

Module Sensor                     Temperature          Status       
------+--------------------------+--------------------+------------ 
1      air inlet                  31C (45C,60C,70C)              ok 
1      air outlet                 37C (61C,76C,86C)              ok 
2      air inlet                  29C (45C,60C,70C)              ok 
2      air outlet                 35C (61C,76C,86C)              ok 
3      Air inlet                  32C (56C,68C,71C)              ok 
3      Air inlet remote           25C (46C,59C,62C)              ok 
3      Air outlet                 32C (66C,76C,79C)              ok 
3      Air outlet remote          27C (60C,71C,74C)              ok 
5      Air inlet                  25C (50C,66C,72C)              ok 
5      Air outlet                 46C (73C,80C,92C)              ok 
5      XPP                        73C (101C,120C,123C)           ok 
5      IFE                        40C (65C,90C,120C)             ok 
5      CONAN                      47C (67C,82C,100C)             ok 
5      CPU                        45C (69C,102C,105C)            ok 
6      Air inlet                  24C (50C,66C,72C)              ok 
6      Air outlet                 45C (73C,80C,92C)              ok 
6      XPP                        71C (101C,120C,123C)           ok 
6      IFE                        37C (65C,90C,120C)             ok 
6      CONAN                      46C (67C,82C,100C)             ok 
6      CPU                        37C (69C,102C,105C)            ok 
7      Air inlet                  27C (44C,59C,62C)              ok 
7      Air inlet remote           28C (47C,61C,64C)              ok 
7      Air outlet                 30C (47C,62C,65C)              ok 
7      Air outlet remote          32C (52C,66C,69C)              ok 
8      air inlet                  25C (41C,56C,66C)              ok 
8      air outlet                 32C (50C,64C,74C)              ok 
9      air inlet                  25C (41C,56C,66C)              ok 
9      air outlet                 34C (50C,64C,74C)              ok 
10     air inlet                  25C (43C,58C,61C)              ok 
10     air outlet (front)         25C (51C,66C,69C)              ok 
10     air outlet (back)          26C (43C,58C,61C)              ok

}}

So for  example, the default value of 60ºC for a critical would be below the normal value for these metrics is throwing errors 5 - Air outlet, 5 XPP

The most extreme case is the XPP value where the minor warning starts at 101ºC, and currently it's throwing errors on 60ºC.

Having a higher preset value for this metric would be the appropiate such a $TEMP_CRIT:"XPP"

 

 Comments   
Comment by Aleksandrs Petrovs-Gavrilovs [ 2023 Dec 19 ]

Hello and thank you for this report.

Could you please confirm that to thresholds on the Cisco were modified using the Threshold manager?

Best Regards,
Aleksandrs

Comment by Joan [ 2023 Dec 20 ]

Hi @Aleksandrs, I hadn't heard previously of the threshold manager, the values are preset by cisco as far as I know and the reading is from a module type WS-X45-SUP8-E.

When doing a snmwalk with the proper MIBs installed, one can read the preset threshold at

CISCO-ENVMON-MIB::ciscoEnvMonTemperatureThreshold.43 (currently 120ºC)

So the rethinking on my post the most appropiate way to fix the triggers on cisco hardware would be to read the ciscoEnvMonTemperatureThreshold and act according to that

I found a post entry of a user called Xyue that changed the cisco template to do that precisely, maybe that would be a good directon to improve the cisco template. The post is https://www.zabbix.com/forum/zabbix-help/45943-discovery-error

 

See the full snmpwalk for my switch

cisco_walk.txt

 

Regards

Comment by Aleksandrs Petrovs-Gavrilovs [ 2023 Dec 22 ]

Thank you for the suggestion, it does seems as a more appropriate approach

Just to inform you, this section of the tracker is for bug reports only. The case you have submitted can not be qualified as one. Instead it can be classified as a Feature change, and should be opened as a ZBXNEXT.

With that, I'm moving this ticket to the correct section in order for it to be checked.

Generated at Tue May 13 09:15:09 EEST 2025 using Jira 9.12.4#9120004-sha1:625303b708afdb767e17cb2838290c41888e9ff0.