The log info and error messages that are a part of the operator-sdk scaffolding are not formatted in a structured log format that follows the msg, key, value pair format. It would be better if they are formatted using the standard recommendations given by logr in :https://godoc.org/github.com/go-logr/logr and https://github.com/kubernetes-sigs/controller-runtime/blob/master/TMP-LOGGING.md which su

When selecting a good Redis Cluster operator I found KubeDB elsewhere, but in this table it is not clear that it provides operators for:

• elasticsearch,
• memcached,
• mongodb,
• mysql,
• postgresql,
• redis

Multiple paths to go:

• add tags and host this page as something interactive with some tag-filtering on Github Pages?
• add these db's to comment column
• add kubedb also at "#3 R

Feature Request

Is your feature request related to a problem? Please describe:

At present, we support to define the percentage of injection errors and provides a number from 0-100 when to use IOChaos, but sometimes we want the smaller probability, such as 0.01%

**Describe the fe

What happened?

The default Prometheus ClusteRole in kube-prometheus is missing some permissions compared to the Prometheus ClusterRole from documentation from the do

There are currently options to enable load balancers for master and replica nodes, but that would be cool to support balancing for pooler service too.

Is your feature request related to a problem? Please describe.
Investigate github.com/google/go-containerregistry to replace the heroku/docker-registry-client which seems to be unmaintained.

Describe the solution you'd like
Check the go-containerregistry library and see if it fits our needs.

Describe alternatives you've considered
We already use a forked version of the docker-

Background

As discussed in #223 comments (zalando-incubator/kopf#223 (comment) and below), there is one type of handlers is missing: when an operator restarts, it should process all the objects that existed before, but were neither updated, nor deleted.

Originally, this was a technical solution for @kopf.on.resume() handlers in #96. But then it was

It looks like the namespace on line 13 of vault-tokenreview-binding.yaml should be "default" instead of "vault-services". I have followed the documentation steps to deploy vault-operator and configure k8s auth, but failed to authenticate. Changing namespace has resolved the issue.

What would you like to be added:
Ignore IDE files by default in order to not face issues as:

$ kubectl kudo install ./first-operator
Error: failed to resolve package CRDs for operator: ./first-operator: while reading package from the file system: unexpected file when reading package from filesystem: first-operator/.idea/first-operator.iml

The Readme would be more comprehensive if it contained instructions to delete all the resources that are created in the given examples.

twitchtv/retool is out of maintenance and alecthomas/gometalinter has been deprecated.

Now that we have upgraded to go v1.13 in which go module is enabled by default, it is better to drop retool and manage these tools in go module in scripts instead.

The deprecated tool gometalinter should be dropped too.

Could someone add docs into README.md on how to configure sinks?
Preferably in a form that k8s noobs could use

Expected behaviour

When running make integration-test, I expect all tests to pass. But that is not the case.

Actual behaviour

make integration-test produces:

$ make integration-tests
./scripts/integration-tests.sh
=> Preparing minikube for running integration tests
😄  minikube v1.8.1 on Darwin 10.15.4
✨  Automatically selected the hyperkit driver. Other choices: do

Which is misleading and should be probably logged when waiting on kafka to be up during the first deployment.

The operator should deploy a replica set instead of single pods as it's a bad practice to run individual pods in Kubernetes, even singleton pods.

Ref: https://kubernetes.io/docs/concepts/workloads/pods/pod-overview/#working-with-pods

Description :

From the current documentation & whatever experimentation I have done on operator metering, I don't see a feature to upload or push the operator metering metrics to a NoSQL database like Cassandra or Kafka topic instead of Presto DB

This can be useful to query operator metering metrics from web applications as Presto is relatively slow

Please share your thoughts on this or a

Currently, we think that the ASO server implementation will be entirely stateless. There may be caches, of course, but the state that ASO considers will be stored in the Kubernetes etcd backend and the AWS services themselves.

So, the upgrade process for ASO itself should be relatively simple: just increment the image version/tag for the Deployment YAML that describes the ASO application in t

Add example manifests for different persistence options (EBS volume / SSD mount / RAM disk)

With the documentation for upgrading a CRD's API version coming soon operator-framework/operator-sdk#1541 operator-framework/operator-sdk#1680 the memcached-operator should have an example of multiple api versions with a CRD manifest that uses the versions field and a conversion webhook that handles conversions between the two apiversions.

Cur