This repository contains a Powershell script sensor for Paessler PRTG which allows you to monitor the health of Aruba switches with ArubaOS-CX operating system (6000 series, 8000 series).
The sensor allows you to monitor the following parameters of individual switches and VSF stacks as well as members of VSX stacks
- Temperatures
- highest currently measured CPU temperature (of all chassis in the VSF stack)
- highest currently measured intake temperature (of all chassis in the stack)
- Tranceivers
- Transceiver laser errors
- Transceiver power errors
- Transceiver temperature errors
- Transceiver voltage errors
- Power supplies
- Current usage of PoE budget per Switch in percent
- State of available power supplies (in percent)
- State of available fan trays (in percent)
- Firmware version baseline check (disabled by default, can be configures in settings file
AOS-CX\settings.json) - Unsaved configuration changes (comparison of running config and saved config)
- All sensor channels can be activated/deactivated via the configuration file
AOS-CX\settings.json
The sensor has already been successfully tested on the following devices:
- Aruba 6000
- Aruba 6100
- Aruba 6200F/6200M
- Aruba 6300M
- Aruba 8325
- Switch does not have a CPU temperature sensor, but only a temperature sensor in a line card that has “CPU” specified as the location. This is used
Please let me know if you have successfully/unsuccessfully tested another Switch model.
The script uses the REST API 10.08 of the ArubaOS-CX switches. It is therefore necessary that the switches to be monitored have at least version 10.08.
To install the sensor, copy the files (AOS-CX folder and AOS-CX-Sensor.ps1) to the PRTG Network Monitor\Custom Sensors\EXEXML folder on your PRTG core server and remote probes that are supposed to use the script sensor.
Since PRTG uses the x86 powershell, the script execution policy must be adjusted in this x86 powershell so that the script sensor can be executed.
# Windows PowerShell (x86) on PRTG servers
Set-ExecutionPolicy Unrestricted
The Windows login data for the device set in PRTG is passed to the script as environment values.
In order to use the REST API of a ArubaOS-CX switch, the web server must be enabled in the VRF, which is used to access the switch from PRTG. Additionally, a password must be set for the admin user to enable the API feature.
# access via default vrf
https-server vrf default
# access via OOB management port
https-server vrf mgmt
# access-mode can be set to readonly
https-server rest access-mode read-only
# (optional, but recommended) add a separate monitoring user
user monitoring group administrators password plaintext <securepassword>The API credentials must be stored in the Windows credentials of the device (or inherited to the device via the group).
In the PRTG sensor settings, the "Environment" setting must be set to "Set placeholders as environment values" so that the switch hostname/IP address and windows login data is passed to the script as environment variables.
For a VSF stack only one sensor needs to be added, VSX stacks need to be monitored individually (one sensor per device).
All sensor channels and settings can be configured in the AOS-CX\settings.json settings file (generated when the sensor is first executed, if not already present).
{
"Connection": {
"TrustAllCertificates": "true" // ignore certificate errors
},
"System": {
"EnableFanTrayMonitoring": "true", // enable fan tray channels
"EnablePowerSupplyMonitoring": "true", // enable power supply channels
"EnableTemperatureMonitoring": "true", // enable temperature channels
"EnablePowerOverEthernetMonitoring": "true", // enable PoE channels
"CpuTemperatureThreshold": "65", // cpu temperature threshold for error
"AirInletTemperatureThreshold": "45" // inlet temperature threshold for error
},
"Transceiver": {
"EnableTransceiverMonitoring": "true", // enable tranceiver channels
"IgnoreWaitingForLinkPorts": "true", // do not generate errors for ports that are in waiting for link state
"IgnoreAdminDownPorts": "true" // do not generate errors for ports that are admin shutdown
},
"Version": {
"EnableVersionMonitoring": "false", // enable firmware version monitoring (disabled by default)
"MinimumVersion": { // Minimum version for which no error is generated (here 10.11.0)
"Major": "10",
"Minor": "11",
"Patch": "0"
}
},
"Configuration": {
"EnableConfigurationSavedMonitoring": "true" // generate error, if the running config and the saved config are different
}
}To test the sensor, the script can be executed on the PRTG server in debug mode.
In order for the debug messages to be displayed, $DebugPreference = “Continue” must be set.
The script can then be called with the parameters -Username, -Password and -IPAddress.
The output should look like this:
# Debug mode
PS E:\PRTG Network Monitor\Custom Sensors\EXEXML> $DebugPreference = "Continue"
# Execute script with parameters
PS E:\PRTG Network Monitor\Custom Sensors\EXEXML> .\AOS-CX-Sensor.ps1 -Username "api" -Password "<XXXXXX>" -IPAddress "192.168.180.50"
DEBUG: [login] POST https://192.168.180.50/rest/v10.08/login?username=api&password=<XXXXXX>
DEBUG: [login] Response is: 200
DEBUG: [login] Login successful
DEBUG: [login] Set-Cookie is: id=XXXXXXXXXXXXXX==; Path=/; HttpOnly; Secure; SameSite=Lax
DEBUG: [platform_data_name] GET https://192.168.180.50/rest/v10.08/system?attributes=platform_name
DEBUG: [platform_data_name] Response is: 200
DEBUG: [platform_data_name] Platform name is: 6200
DEBUG: [subsystem_data] GET https://192.168.180.50/rest/v10.08/system/subsystems?attributes=fans,temp_sensors,power_supplies&depth=6
DEBUG: [subsystem_data] Response is: 200
DEBUG: [interface_data] GET https://192.168.180.50/rest/v10.08/system/interfaces?attributes=l1_state,pm_info&depth=2
DEBUG: [interface_data] Response is: 200
DEBUG: [config_hash] GET https://192.168.180.50/rest/v10.08/fullconfigs/hash/running-config
DEBUG: [config_hash] Response is: 200
DEBUG: [config_hash] GET https://192.168.180.50/rest/v10.08/fullconfigs/hash/startup-config
DEBUG: [config_hash] Response is: 200
DEBUG: [Logout] POST https://192.168.180.50/rest/v10.08/logout
DEBUG: [Logout] Response is: 200
DEBUG: [Logout] Logout successful
<prtg>
<result>
<channel>Chassis 1 PSU Health</channel>
<value>100</value>
<unit>Percent</unit>
<limitminerror>100</limitminerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>CPU Temp</channel>
<value>46.1</value>
<unit>Temperature</unit>
<float>1</float>
<limitmaxerror>65</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Inlet-Air Temp</channel>
<value>21.2</value>
<unit>Temperature</unit>
<float>1</float>
<limitmaxerror>45</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Transceiver power errors</channel>
<value>0</value>
<unit>Count</unit>
<limitmaxerror>0</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Transceiver temperature errors</channel>
<value>0</value>
<unit>Count</unit>
<limitmaxerror>0</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Transceiver laser bias errors</channel>
<value>0</value>
<unit>Count</unit>
<limitmaxerror>0</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Transceiver voltage errors</channel>
<value>0</value>
<unit>Count</unit>
<limitmaxerror>0</limitmaxerror>
<limitmode>1</limitmode>
</result>
<result>
<channel>Config saved</channel>
<value>1</value>
<unit>Custom</unit>
<valuelookup>prtg.standardlookups.yesno.stateyesok</valuelookup>
</result>
</prtg>