kbrew-dev / kbrew-registry Goto Github PK

A recipe using the VictoriaMetrics operator and ability to configure multiple configurations would be nice. For ex. the HA installation documentation is here

Rook installation exposes quite a few items and needs a lot of configuration for an install:

https://www.digitalocean.com/community/tutorials/how-to-set-up-a-ceph-cluster-within-kubernetes-using-rook

The Harbor's Helm chart seems a bit complex than usual charts. It has multiple components.
https://goharbor.io/docs/1.10/install-config/harbor-ha-helm/

A recipe for Akka Cluster

If we look at OpenEBS as an example: https://docs.openebs.io/docs/next/installation.html it is fairly complicated and could be a good candidate for a recipe.

Istio installation is great candidate for a recipe including upgrade and uninstall scenarios. The docs here

Hazelcast is a in memory data grid and the installation has operator and enterprise operator flavours. Would be good to have a recipe for Hazelcast.

Splunk Recipe

Tried on DO K8S - single node cluster. It looks like it does not really wait for 30 minutes even if I specify 30 min as timeout, finally results in error:

It took only 6.5 odd minutes but it failed:

kbrew install kafka-operator -t 30m  32.13s user 8.73s system 10% cpu 6:34.59 total

And error:

zookeepercluster.zookeeper.pravega.io/zookeeper created
Waiting for zookeeper cluster to be ready
timed out while waiting for cluster to be ready
Failed to report event. exit status 1
Error: exit status 1

Also exit status is not right:

$ echo $?
0

https://operatorhub.io/operator/mongodb-atlas-kubernetes

A recipe for Spinnaker Operator

Minio is a great example of a fairly important stack that needs to be installed & configured properly. A recipe for MinIO would be good to attempt - based on the operator that is available, install docs etc!

LongHorn installation needs a bunch of checks and then installs itself on cluster - could be a good candidate for us to write a recipe: https://longhorn.io/docs/1.0.0/deploy/install/

Installation steps seem to be dependent on cloud providers. This could be a good candidate for kbrew-recipe - https://docs.k8ssandra.io/install/

A recipe may not be always well described in a single file. Maybe in the future, we might move some of the elements from being inline to a file in the same directory. So it might be worth thinking if a recipe is a file or a directory (Collection of files)

General instructions with dos and don'ts.

A recipe for Spark

A recipe for Seldon

Hi,

i would like to add OpenFaas https://artifacthub.io/packages/helm/openfaas/openfaas to kbrew.

A recipe for Kubeflow

The RabbitMQ operator installation is quite simple but the customisations on top are quite a few: https://www.rabbitmq.com/kubernetes/operator/using-operator.html

We need to think of how we can either set most customisations to some defaults and how to expose them to CLI for users to override an easy way