Eric Greer
Eric Greer
It would be nice to be able to query a single `khcheck`, rather than having to (as a user) query a list of all checks and then iterate on them....
In many cases, users do not want the daemonset check to run on any node with a state of `Not Ready`, cordoned, or when taints are present of any kind....
The new `kh-run-uuid` header may not be in the documentation for manual check creation. We should be sure this is listed for those who are creating checks that aren't written...
The current helm chart uses a values system like this: ``` imageRegistry: {} image: registry: kuberhealthy repository: kuberhealthy tag: v2.4.1 ``` Given that the format of images is: `//:`, and...
During an upgrade of the Kubernetes package versions in PR #636 a bunch of `context.TODO()` function call parameters were added. We should go through and plug these into any existing...
We should make a pass to improve the use of logging on resources Kuberhealthy works with. If we are modifying a `khcheck`, we can lave an event on it. Same...
We should have an email that users can reach out to in order to report potentially sensitive security concerns.
The check reaper leaves pods that have errored for five days [statically in the code](https://github.com/Comcast/kuberhealthy/blob/5c8da01d1ef853c63b7a815d896373fbfed21a87/cmd/kuberhealthy/reaper.go#L166-L168). It would be nice to make this configurable by the user so they aren't forced...
Kuberhealthy is great at tracking with synthetic monitors fail and why, but it would be nice if there were some way to hook into triggers for automatic remediation of some...
We should do some extra layers of security around checks that ingested into Kuberhealthy. Because of the nature of Kuberhealthy to run lots of remote images that are being updated...