Look at all those data retries ! – Do you want to make something like this? Sure you do! It is based on a pre-made dashboard, and if you read Dave Benham’s blog Viewing Network Telemetry from the Catalyst 9800 with Grafana you get it up and running in less than an hour. If you need to learn how to get the 9800CL up and running, he got a blog for that also. Building a Catalyst 9800-CL Lab with an Intel NUC. I recommend reading his telemetry blog before reading this if you do not know much about telemetry and Netconf-Yang.
Reason for me getting into this was to my «surprise» that the EWC (Embedded Wireless Controller on Cisco Access Points) did not support SNMP, so for those who got Prime and do not want to buy a DNAC there had to be some alternatives – and sure it is!
Spoiler alert ! – I have not got it working on the EWC for some weird reasons, only on a bunch of 9800CL and 9800-L-F. – Nicholas Swiatecki has tested to see if it works and it does, he was helped me a lot – follow him on Twitter!
When I get this working with the EWC, I can have drop-downs for the different EWCs and troubleshooting dashboards and alarms when something happens. Disjoin for example, all with Grafana, and the API to Slack. Very nice to have for customers that would want to use the EWC and for me that use an EWC at home.
Note: This is the first part of 2 or 3 blog posts, and I will not go into heavy details that Dave Benham has already covered in this blog.
First up – how to find the correct Xpath.
The Advanced Netconf Explorer
If you have banged your head trying to get the Yang Explorer working, you are not alone. I even tried to used the docker, but I did not want to turn back time and force Flash into browser again.
Do not worry, we have a solution! The Advanced Netconf Explorer, and I will tell you how you can get this up and running and find all the metrics you want to graph up !
This tool is great to find the Xpath, and also see what is currently running on your 9800CL. I really hope they keep on maintaining this!
And here is how you get it up and running.
First you need to download the Zip file that includes all the Docker Files.
kjetil@LabbyMcLab:~/dockerfiles$ wget https://github.com/cisco-ie/anx/archive/master.zip
Next you unzip this file.
If you do not have Docker-Compose, I recommend that you get this installed. See this guide. How To Install and Use Docker Compose on Ubuntu 20.04
Then you just type «(sudo) docker-compose up -d» while in the folder and let it run until it is done.
Then type «
docker ps -a» to see the docker, and go into localhost:9269
kjetil@LabbyMcLab:~/dockerfiles/anx-master$ docker ps -a
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
13d6a8c4c117 anx-master_anx "/usr/share/jetty9/b…" 38 seconds ago Up 37 seconds 0.0.0.0:9269->8080/tcp anx-master_anx_1
Boom ! And there it is. – Just type in the username and password for the controller and it will load everything the device support. Just like a typical SNMP OiDViewer.
Below you see the GUI, and to start here is the Xpath for the new external antenna monitoring function, and the Xpath you need to add to the telemetry config to get this into Grafana.
Or what about the new OpenConfig modules (I have not got this data to Telegraf yet, even with my Xpath being valid)
You can also see what is currently running on the device and what information they contain. When choosing «Show Data – Running» it will load all the current valid telemetry subscriptions that is running on the device.
This next picture (below) is the telemetry subscription you need to make the fancy RSSI and SNR graphs.
You can also just search for words, for example RSSI and see if there is something you would like to graph up in Grafana. Here you see one about RRM.
It is so much to find here, and you need to have the correct xpath to get them working in Grafana. If you use the whole container, you often end up in not being able to map it to the particular client.
You will understand when I get to my next blog post up. This was just about the Advanced Netconf Explorer that is highly needed to find more things you would like to add to Grafana.
Sneak peak to my next post.
More about the actual graph configuration, for example that colorful thing at the bottom on the first picture.
The roaming history for every single client from the drop-down menu.
AP-Name, mac:adr, channel, roaming-reason, run latency. Add that with the RSSI, SNR, Tx, Rx, Data Retries and more and you have a very good troubleshooting dashboard for all those cases you get from customers saying «I had an issue last week at 5:65 AM» – Yeah, we know – we have the data.
And best of all. It is FREE! You do not even need so many resources 🙂