Cisco Wireless Access Point - CPU 100% (but it is not true)

Help
1

Dear LibreNMS,

I added a wireless AP device to our LibreNMS instance but the processors graph shows that the CPU is stuck at 100% and the displayed uptime shows permanently 1 second, whereas both are not true. The AP is fonctionning correctly. When I reboot the AP, the information becomes correct, but after few hours, it becomes wrong again.

AP model: Cisco Small Business WAP4410N-E.

Thank you for your help.
Best regards,

2

Bonjour @Auguste
Please check if the device is replying correct value in SNMP at first (just after reboot) and then again when the value is wrong, and compare both.
You can debug the replies from the device directly from librenms server CLI, running ./poller.php -h -v -d and checking all the SNMP replies for uptime and CPU.

3

Hello,
thank you for the answer.
I checked the SNMP replies and i saw this:

Error in packet
Reason: (noSuchName) There is no such variable name in this MIB.
Failed object: hrSystemProcesses.0

Error in packet
Reason: (noSuchName) There is no such variable name in this MIB.
Failed object: hrSystemUptime.0

Best regards,

4

I added another Cisco AP (WAP200 model) but LibreNMS identifies it as “Linux2.4.27-star” and not as “Cisco Wireless Access Point” in the “Operating System” column. And for this AP, the information is correct, both in terms of CPU and uptime.

5

Those devices are different (aside of the name WAP) and do not behave at all the same. If we don’t receive the SNMP values we expect, the display will be wrong.
So we need contributors to correct the support. Feel free to send us patches on github.