diff --git a/Makefile b/Makefile
index df625369b5..9b44c21278 100644
--- a/Makefile
+++ b/Makefile
@@ -13,7 +13,12 @@ CHARTS_INTEGRATION    := wire-server databases-ephemeral redis-cluster fake-aws
 # (e.g. move charts/brig to charts/wire-server/brig)
 # this list could be generated from the folder names under ./charts/ like so:
 # CHARTS_RELEASE := $(shell find charts/ -maxdepth 1 -type d | xargs -n 1 basename | grep -v charts)
-CHARTS_RELEASE        := wire-server redis-ephemeral redis-cluster databases-ephemeral fake-aws fake-aws-s3 fake-aws-sqs aws-ingress  fluent-bit kibana backoffice calling-test demo-smtp elasticsearch-curator elasticsearch-external elasticsearch-ephemeral minio-external cassandra-external nginx-ingress-controller nginx-ingress-services reaper sftd restund coturn inbucket
+CHARTS_RELEASE := wire-server redis-ephemeral redis-cluster databases-ephemeral	\
+fake-aws fake-aws-s3 fake-aws-sqs aws-ingress fluent-bit kibana backoffice		\
+calling-test demo-smtp elasticsearch-curator elasticsearch-external				\
+elasticsearch-ephemeral minio-external cassandra-external						\
+nginx-ingress-controller nginx-ingress-services reaper sftd restund coturn		\
+inbucket k8ssandra-test-cluster
 KIND_CLUSTER_NAME     := wire-server
 
 package ?= all
diff --git a/changelog.d/5-internal/k8ssandra-cluster-helm-chart b/changelog.d/5-internal/k8ssandra-cluster-helm-chart
new file mode 100644
index 0000000000..ce269c8ef6
--- /dev/null
+++ b/changelog.d/5-internal/k8ssandra-cluster-helm-chart
@@ -0,0 +1 @@
+Add Helm chart to configure clusters managed by k8ssandra-operator for test environments.
diff --git a/charts/k8ssandra-test-cluster/.helmignore b/charts/k8ssandra-test-cluster/.helmignore
new file mode 100644
index 0000000000..0e8a0eb36f
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/.helmignore
@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/
diff --git a/charts/k8ssandra-test-cluster/Chart.yaml b/charts/k8ssandra-test-cluster/Chart.yaml
new file mode 100644
index 0000000000..b67746d307
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/Chart.yaml
@@ -0,0 +1,9 @@
+apiVersion: v2
+name: k8ssandra-test-cluster
+description: K8ssandra (Cassandra cluster) K8ssandraCluster object for wire test servers. (This does not install K8ssandra itself!)
+
+type: application
+
+version: 0.1.0
+
+appVersion: "0.39.2"
diff --git a/charts/k8ssandra-test-cluster/README.md b/charts/k8ssandra-test-cluster/README.md
new file mode 100644
index 0000000000..865183c156
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/README.md
@@ -0,0 +1,89 @@
+# k8ssandra-test-cluster Helm chart
+
+`k8ssandra-test-cluster` provides a `K8ssandraCluster` object to create a
+*Cassandra* database with
+[`k8ssandra-operator`](https://artifacthub.io/packages/helm/k8ssandra/k8ssandra-operator).
+**It does not install `k8ssandra-operator` itself!** This configuration is meant
+to be used in test environments: **It lacks crucial parts like backups
+(`medusa`)!**
+
+## Usage in Helmfile
+
+### Prerequisites
+
+You need a *storage class* that can automatically request storage volumes. For
+Hetzner's cloud see: [Container Storage Interface driver for Hetzner
+Cloud](https://github.com/hetznercloud/csi-driver)
+
+### Usage
+
+These entries are used in the `helfile` file:
+
+``` yaml
+...
+
+repositories:
+  - name: wire
+    url: 'https://s3-eu-west-1.amazonaws.com/public.wire.com/charts'
+  - name: k8ssandra-stable
+    url: https://helm.k8ssandra.io/stable
+
+...
+
+releases:
+  - name: k8ssandra-operator
+    chart: 'k8ssandra-stable/k8ssandra-operator'
+    namespace: databases
+    version: 0.39.2
+    values:
+      # Use a cass-operator image that is compatible to the K8s cluster version
+      - cass-operator:
+            image:
+              tag: v1.10.5
+
+  # Installs CDRs automatically
+  - name: k8ssandra-test-cluster
+    chart: "wire/k8ssandra-test-cluster"
+    namespace: "databases"
+    version: {{ .Values.wireChartVersion | quote }}
+    needs:
+      - 'databases/k8ssandra-operator'
+    wait: true
+    waitForJobs: true
+
+  - name: 'wire-server'
+    namespace: 'wire'
+    chart: 'wire/wire-server'
+    version: {{ .Values.wireChartVersion | quote }}
+    values:
+      - './helm_vars/wire-server/values.yaml.gotmpl'
+    secrets:
+      - './helm_vars/wire-server/secrets.yaml'
+    needs:
+      - 'databases/k8ssandra-test-cluster'
+
+...
+```
+
+Please note the `needs` relations of the releases: `wire-server` *needs*
+`k8ssandra-test-cluster` which *needs* `k8ssandra-operator`.
+
+`wait` and `waitForJobs` are mandatory for `k8ssandra-test-cluster` in this
+setup: These settings ensure that the database really exists before resuming
+with the deployment.
+
+## Implementation details
+
+### k8ssandra-cluster.yaml
+
+Contains the `K8ssandraCluster` object. Its schema is described in the [CRD
+reference](https://docs-v2.k8ssandra.io/reference/crd/k8ssandra-operator-crds-latest/#k8ssandracluster)
+
+The specified *Cassandra* cluster runs on a single Node with reasonable
+resources for test environments.
+
+### check-cluster-job.yaml
+
+Defines a job that tries to connect to the final *Cassandra* database. Other
+deployments can wait on this. This is useful because `wire-server` needs a
+working database right from the beginning of it's deployment.
diff --git a/charts/k8ssandra-test-cluster/templates/check-cluster-job.yaml b/charts/k8ssandra-test-cluster/templates/check-cluster-job.yaml
new file mode 100644
index 0000000000..dbff6e0332
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/templates/check-cluster-job.yaml
@@ -0,0 +1,19 @@
+# This job fails until the Cassandra created database is reachable. The Helmfile
+# deployment can wait for it. This is used to start wire-server deployments only
+# with a reachable database.
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: check-cluster-job
+  namespace: databases
+spec:
+  template:
+    spec:
+      containers:
+      - name: cassandra
+        image: cassandra:3.11
+        command: ["cqlsh", "k8ssandra-cluster-datacenter-1-service"]
+      restartPolicy: OnFailure
+  # Default is 6 retries. 8 is a bit arbitrary, but should be sufficient for
+  # low resource environments (e.g. Wire-in-a-box.)
+  backoffLimit: 8
diff --git a/charts/k8ssandra-test-cluster/templates/k8ssandra-cluster.yaml b/charts/k8ssandra-test-cluster/templates/k8ssandra-cluster.yaml
new file mode 100644
index 0000000000..4139f672e5
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/templates/k8ssandra-cluster.yaml
@@ -0,0 +1,43 @@
+apiVersion: k8ssandra.io/v1alpha1
+kind: K8ssandraCluster
+metadata:
+  name: k8ssandra-cluster
+  namespace: databases
+spec:
+  auth: false
+  cassandra:
+    serverVersion: "3.11.11"
+    telemetry:
+      prometheus:
+        enabled: true
+    resources:
+      requests:
+        cpu: 1
+        memory: "4.0Gi"
+      limits:
+        memory: "4.0Gi"
+    config:
+      jvmOptions:
+        # Intentionally, half of the available memory
+        heap_max_size: "2G"
+        heap_initial_size: "2G"
+        gc_g1_rset_updating_pause_time_percent: 5
+        gc: "G1GC"
+        gc_g1_max_gc_pause_ms: 300
+        gc_g1_initiating_heap_occupancy_percent: 55
+        gc_g1_parallel_threads: 16
+    datacenters:
+        - metadata:
+            name: datacenter-1
+          size: 1
+          storageConfig:
+            cassandraDataVolumeClaimSpec:
+              storageClassName: {{ .Values.storageClassName }}
+              accessModes:
+                - ReadWriteOnce
+              resources:
+                requests:
+                  storage: {{ .Values.storageSize }}
+  reaper:
+    autoScheduling:
+      enabled: true
diff --git a/charts/k8ssandra-test-cluster/values.yaml b/charts/k8ssandra-test-cluster/values.yaml
new file mode 100644
index 0000000000..3aabc8db1a
--- /dev/null
+++ b/charts/k8ssandra-test-cluster/values.yaml
@@ -0,0 +1,13 @@
+# The values in k8ssandra-cluster.yaml are well choosen. Please only export and
+# override them if you are confident the change is needed.
+
+# storageClassName: the name storageClass to use. This defines where the data is
+# stored. Storage is automatically requested if the storage class is correctly
+# setup.
+storageClassName: hcloud-volumes-encrypted
+
+# storageSize: Size of the storage (persistent volume claim) to request. At
+# Hetzner's cloud the smallest volume is 10GB. So, even if you need much less
+# storage, it's fine to request 10GB. The memory units are described here:
+# https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/#meaning-of-memory
+storageSize: 10G