Skip to content

About

An example of a helm chart that can set up virtserver in kubernetes

Topics

Resources

Stars

1 star

Watchers

11 watching

Forks

Latest commit

 

History

11 Commits

Folders and files

Repository files navigation

VirtServer Helm Chart

This is a starting point for running VirtServer in kubernetes utilizing helm charts. Users might need to modify it to their needs. This helm chart was last updated 2022-08-18 for VirtServer 3.12.1.

Managing

Installing

If your helm is setup with kubernetes correctly all you need to do to run the helm chart is the following:

helm install virtserver . -f values.yaml

Upgrading

Whenever you need to update your VirtServer to a new version or simply just restart it run:

helm upgrade virtserver . -f values.yaml

This usually makes your VirtServer unavailable for a minute and any virtual services not configured to autostart will need to be manually started again.

Uninstalling

When it is time to bring down VirtServer you run:

helm uninstall virtserver

Configuration

Within this repository there is an example of what values.yaml could look like.

licenseServer

Where ProtectionLS license server is running. Make sure that any network configuration needed is done so that the VirtServer node can reach the license server. If the license server can't be reached or doesn't have a valid VirtServer license then VirtServer won't be able to start.

Example: host.minikube.internal:1443

Note: This example works for when the ProtectionLS license server is running at the same host as a minikube cluster. Other environments will need a different domain or IP and port. Also note that ProtectionLS in this particular configuration is not running on its default port but rather 1443.

exposePorts

An array with exactly two elements representing the start and stop of a range of ports that should be exposed of VirtServer hosts. These ports maps to your virtual services running in VirtServer. The port that you would access them through will probably not be the same as the port you decide to open. If you have 10 virtual services then you probably want to expose 10 ports. The virtual services need to run on a port within this range.

Example: [8000, 8010]

To figure out what port it gets mapped to(and is accessible on outside the Kubernetes cluster) run:

$ kubectl get svc ready-api-virtserver
NAME                   TYPE       CLUSTER-IP       EXTERNAL-IP   PORT(S)                                                                                                                                                                AGE
ready-api-virtserver   NodePort   10.100.126.233   <none>        9090:32298/TCP,8000:31051/TCP,8001:30639/TCP,8002:32653/TCP,8003:30283/TCP,8004:30470/TCP,8005:31451/TCP,8006:31511/TCP,8007:31808/TCP,8008:31729/TCP,8009:32241/TCP,8010:30207/TCP   5d18h

configStorageSize

How big storage you want to give VirtServer for storing deployments, HAR-logs, audit logs, regular logs and more. Can start of as a small value of for example 1Gi, but ensure you then monitor storage space, so you can upgrade it or clean out old logs whenever needed. Valid values are anything that would be valid as Memory resource units.

Example: 1Gi

image

Which VirtServer image you want to run. This allows you to upgrade versions quick and simple by just changing the version. Need to be the full image name including repository.

Example: smartbear/ready-api-virtserver:3.12.1

About

An example of a helm chart that can set up virtserver in kubernetes

Topics

Resources

Stars

1 star

Watchers

11 watching

Forks

Releases

Packages

Used by

Contributors