Quickstart ¶
For the quickstart we have examples for Kind or k3s.
- Kind - is suitable for CPU only installations.
- k3s - is suitable for CPU or GPU installations.
Kubernetes >= 1.25
The current version of the chart supports kubernetes version 1.25 and above.
Please select your target Kubernetes variant:¶
Kind (kubernetes in docker kind.sigs.k8s.io)¶
This installation in a kind cluster is for trying out the operators and the database in a non-production environment.
CPU Only
This method currently only supports installing a CPU version of the database.
Please contact Kinetica Support to request a trial key.
Create Kind Cluster 1.29¶
wget https://raw.githubusercontent.com/kineticadb/charts/72.0.11/kinetica-operators/kind.yaml
kind create cluster --name kinetica --config kind.yaml
Set Kubernetes Context
Please set your Kubernetes Context to kind-kinetica before performing the following steps.
Kind - Install kinetica-operators including a sample db to try out¶
Review the values file charts/kinetica-operators/values.onPrem.kind.yaml. This is trying to install the operators and a simple db with workbench installation for a non production try out.
As you can see it is trying to create an ingress pointing towards local.kinetica. If you have a domain pointing to your machine, replace it with the correct domain name.
Kind - Install the Kinetica-Operators Chart¶
helm repo add kinetica-operators https://kineticadb.github.io/charts/latest
FQDN or Local Access
By default we create an ingress pointing towards local.kinetica. If you have a domain pointing to your machine, replace/set the FQDN in the values.yaml with the correct domain name or by adding --set.
If you are on a local machine which is not having a domain name, you add the following entry to your /etc/hosts file or equivalent.
wget https://raw.githubusercontent.com/kineticadb/charts/72.0.11/kinetica-operators/values.onPrem.kind.yaml
helm -n kinetica-system upgrade -i kinetica-operators kinetica-operators/kinetica-operators --create-namespace --values values.onPrem.kind.yaml --set db.gpudbCluster.license="your_license_key" --set dbAdminUser.password="your_password"
or if you have been asked by the Kinetica Support team to try a development version
helm search repo kinetica-operators --devel --versions
helm -n kinetica-system upgrade -i kinetica-operators kinetica-operators/kinetica-operators/ --create-namespace --values values.onPrem.kind.yaml --set db.gpudbCluster.license="your_license_key" --set dbAdminUser.password="your_password" --devel --version 72.0.11
Accessing the Workbench
You should be able to access the workbench at http://local.kinetica
k3s (k3s.io)¶
Install k3s 1.29¶
curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="--disable=traefik --node-name kinetica-master --token 12345" K3S_KUBECONFIG_OUTPUT=~/.kube/config_k3s K3S_KUBECONFIG_MODE=644 INSTALL_K3S_VERSION=v1.29.2+k3s1 sh -
Once installed we need to set the current Kubernetes context to point to the newly created k3s cluster.
Select if you want local or remote access to the Kubernetes Cluster: -
For only local access to the cluster we can simply set the KUBECONFIG environment variable
For remote access i.e. outside the host/VM k3s is installed on: -
Copy /etc/rancher/k3s/k3s.yaml on your machine located outside the cluster as ~/.kube/config. Then edit the file and replace the value of the server field with the IP or name of your K3s server.
sudo chmod 600 /etc/rancher/k3s/k3s.yaml
mkdir -p ~/.kube
sudo cp /etc/rancher/k3s/k3s.yaml ~/.kube/config
sudo chown "${USER:=$(/usr/bin/logname)}:$USER" ~/.kube/config
# Edit the ~/.kube/config server field with the IP or name of your K3s server here
export KUBECONFIG=~/.kube/config
K3s - Install kinetica-operators including a sample db to try out¶
Review the values file charts/kinetica-operators/values.onPrem.k3s.yaml. This is trying to install the operators and a simple db with workbench installation for a non production try out.
FQDN or Local Access
By default we create an ingress pointing towards local.kinetica. If you have a domain pointing to your machine, replace/set the FQDN in the values.yaml with the correct domain name or by adding --set.
If you are on a local machine which is not having a domain name, you add the following entry to your /etc/hosts file or equivalent.
K3S - Install the Kinetica-Operators Chart (CPU)¶
helm repo add kinetica-operators https://kineticadb.github.io/charts/latest
wget https://raw.githubusercontent.com/kineticadb/charts/72.0.11/kinetica-operators/values.onPrem.k3s.yaml
helm -n kinetica-system install kinetica-operators kinetica-operators/kinetica-operators --create-namespace --values values.onPrem.k3s.yaml --set db.gpudbCluster.license="your_license_key" --set dbAdminUser.password="your_password"
or if you have been asked by the Kinetica Support team to try a development version
helm search repo kinetica-operators --devel --versions
helm -n kinetica-system install kinetica-operators kinetica-operators/kinetica-operators --create-namespace --values values.onPrem.k3s.yaml --set db.gpudbCluster.license="your_license_key" --set dbAdminUser.password="your_password" --devel --version 7.2.0-2.rc-2
K3S - Install the Kinetica-Operators Chart (GPU)¶
If you wish to try out the GPU capabilities, you can use the following values file, provided you are in a nvidia gpu capable machine.
wget https://raw.githubusercontent.com/kineticadb/charts/72.0.11/kinetica-operators/values.onPrem.k3s.gpu.yaml
helm -n kinetica-system install kinetica-operators charts/kinetica-operators/ --create-namespace --values values.onPrem.k3s.gpu.yaml --set db.gpudbCluster.license="your_license_key" --set dbAdminUser.password="your_password"
Accessing the Workbench
You should be able to access the workbench at http://local.kinetica
Uninstall k3s¶
Default User
Username as per the values file mentioned above is kadmin and password is Kinetica1234!