Running the tool in-cluster ?

[sundowndev]

Hi, I like the way clusterlint helps me fixing issues in my cluster and I'd like to make it a continuous process. Since I don't manage my cluster locally but with Terraform, I was wondering if it could be possible to run clusterlint in-cluster and retrieve scan results via some metrics (JSON, Prometheus, ...) or web UI for example. This could be useful for collaboration as well. I'd be happy to work on that, is it relevant ? Is it an anti-pattern to run the tool in-cluster ?

[varshavaradarajan]

@sundowndev - the client does not use the in-cluster rest config as DOKS accesses the cluster externally. However, it's not an anti-pattern to run the tool in-cluster. We'd be happy to accept a PR to support in-cluster config. You'd need to make the change here

Once in-cluster config is supported, you can have a job on your k8s cluster to run clusterlint and parse the results.

Alternatively, if you are using DOKS, you can leverage the DOKS clusterlint APIs to do this:

• To run clusterlint checks: https://docs.digitalocean.com/reference/api/api-reference/#operation/run_clusterlint
• To fetch clusterlint results: https://docs.digitalocean.com/reference/api/api-reference/#operation/get_clusterlint_results Have a job to make these API requests (note the fetching the results will take time depending on how big your cluster is) and parse the results.