NetApp IP MC - Switch monitoring or Cisco switch in general

[BrendonA667]

Is your feature request related to a problem? Please describe.
Feature request if this is possible

Describe the solution you'd like
If possible any metric from the Cisco Cluster switches specially from the Metro Cluster. We had and issue with one of the replication port where down and ONTAP did not reported this. Case is open for investigation but independent from this it would be a huge improvement when harvest could collect some data from the supported/recommended Cisco switches.

Additional context
From my point of view harvest collects at the moment all possible metrics that are currently available in ONTAP. If this could be extended to Cisco switches this would be awesome.

[cgrinds]

While we evaluate the feasibility of adding a Cisco switch collector to Harvest, I gathered a list of existing open-source Cisco exporters. Have you tried any from the list below @BrendonA667?

You should be able to use one of these to:

• collect and export Cisco switch metrics
• update your existing instance of Prometheus to scrape the new endpoint
• use those metrics in Grafana

Looks like most of the Cisco switches do not provide a REST API and instead, the collectors listed below, gather the metrics via ssh. Some of the Nexus switches provide a REST API. Most of the projects below have not been updated in a few years which may mean they are abandoned or feature complete.

We have not tried them and can not vouch for their quality.

• https://github.com/wobcom/Cisco-Exporter
• https://github.com/lwlcom/cisco_exporter
• https://github.com/greenpau/network_exporter
• https://github.com/AS-Path/cisco_exporter
• https://karneliuk.com/2022/08/tools-11-running-prometheus-exporters-on-dis-aggregated-data-center-switches-cisco-nx-os-nvidia-cumulus-linux-arista-eos/

[BrendonA667]

Hi @cgrinds Thank you for your feedback. We will discuss this if we use one of these exporters dedicated for Cisco or like a generic one with SNMP. My idea behind this request is, that since ONTAP already monitor the switches, there should (in my small world) any informations that Harvest could collect

[cgrinds]

hi @BrendonA667 understood. I like your idea. Can you see if this returns anything for you? Replacing $ip, $user, $pass with correct values.

curl -sk -u$user:$pass 'https://$ip/api/storage/switches?fields=*'

[BrendonA667]

Hi @cgrinds with the API api/network/ethernet/switches?fields=* there is like the CLI output but unfortunately nothing which adds a big value regarding performance or the port status. I'm not sure how ONTAP collects information which is showed in the system health alert section when for example a port goes down.

[cgrinds]

@BrendonA667 looks like api/storage/switches?fields=*' should tell you if a port is down along with some speed and environmental metrics. Can you try that endpoint instead of network/ethernet/switches?

[roybatty2019]

Having monitoring of MetroCluster ISL links in Harvest + nabox would be really useful. For example, we'd like to see the utilisation of the links to know if we need to upgrade the bandwidth between sites. Of course the up/down status is also really important.