Skip to content
This repository has been archived by the owner on Oct 23, 2024. It is now read-only.

Latest commit

 

History

History
173 lines (136 loc) · 5.75 KB

installing.md

File metadata and controls

173 lines (136 loc) · 5.75 KB
Raw

Installing the KUDO Cassandra Operator

Table of Contents

Requirements

  • The KUDO CLI installed on your machine
  • KUDO running in your cluster

Make sure that the KUDO version is at least 0.8.0, both for the CLI and the KUDO running in your cluster.

To see the KUDO CLI version:

kubectl kudo version

To see the KUDO version running in your cluster:

kubectl get pods/kudo-controller-manager-0 \
        -n kudo-system \
        -o jsonpath='{.spec.containers[0].image}' \
  | cut -d: -f2

Installing the operator

It is possible to install multiple instances of the KUDO Cassandra Operator. Each instance is a managed Cassandra cluster. KUDO allows differentiating instances with the --instance parameter.

KUDO operator instances are also namespaced via Kubernetes namespaces. It is possible to have similarly named instances in different namespaces via the --namespace parameter, and differently named instances in the same namespace.

The command below installs a KUDO Cassandra operator instance named "analytics-cassandra" in the "production" namespace. Omitting the --namespace parameter will cause the instance to be installed in the "default" namespace, and omitting the --instance parameter will cause the instance name to be "cassandra". If providing a namespace, make sure it exists.

kubectl kudo install cassandra \
        --instance analytics-cassandra \
        --namespace production

By default a 3-node Cassandra cluster is installed, with each Cassandra node requiring 1 CPU and 4GiB memory.

The total resources needed are 3 CPUs and 12GiB memory

The command above will start the operator instance installation. To check the installation progress, the KUDO CLI provides us with the plan status command. Notice that the --instance and --namespace parameters must be passed on every KUDO CLI command, so that it interacts with the correct operator instance.

Running plan status right after the install command will likely show the plan as still "in progress". This means that the operator is still deploying all necessary pods, services, etc.

kubectl kudo plan status deploy \
        --instance analytics-cassandra \
        --namespace production
Plan(s) for "analytics-cassandra" in namespace "production":
.
└── analytics-cassandra (Operator-Version: "cassandra-0.1.0" Active-Plan: "deploy")
    └── Plan deploy (serial strategy) [IN_PROGRESS]
        └── Phase nodes [IN_PROGRESS]
            └── Step node (IN_PROGRESS)

After a minute or so the deployment should report as "complete":

kubectl kudo plan status deploy \
        --instance analytics-cassandra \
        --namespace production
Plan(s) for "analytics-cassandra" in namespace "production":
.
└── analytics-cassandra (Operator-Version: "cassandra-0.1.0" Active-Plan: "deploy")
    └── Plan deploy (serial strategy) [COMPLETE]
        └── Phase nodes [COMPLETE]
            └── Step node (COMPLETE)

The operator instance pods will also report as "running":

kubectl get pods -n production
NAME                         READY   STATUS    RESTARTS   AGE
analytics-cassandra-node-0   2/2     Running   0          124m
analytics-cassandra-node-1   2/2     Running   0          123m
analytics-cassandra-node-2   2/2     Running   0          122m

The Cassandra cluster should also report all nodes as "UN":

kubectl exec pod/analytics-cassandra-node-0 \
        -n production \
        -c cassandra \
        -- \
        bash -c "nodetool status"
Datacenter: datacenter1
=======================
Status=Up/Down
|/ State=Normal/Leaving/Joining/Moving
--  Address          Load       Tokens       Owns (effective)  Host ID                               Rack
UN  192.168.180.232  219.93 KiB  256          68.7%             664c3243-a7b4-48cf-840d-3173aadf9595  rack1
UN  192.168.246.123  193.24 KiB  256          66.2%             38a639d0-6ead-4dcf-b301-f1272e7f870c  rack1
UN  192.168.144.100  191.78 KiB  256          65.1%             18c470c3-f210-4ced-8512-c720bd2828d8  rack1

The operator deploys a service that provides a DNS record for containers to interact with the Cassandra cluster.

kubectl exec -it pod/analytics-cassandra-node-0 \
        -n production \
        -c cassandra \
        -- \
        bash -c 'cqlsh analytics-cassandra-svc.production.svc.cluster.local'
Connected to analytics-cassandra at analytics-cassandra-svc.production.svc.cluster.local:9042.
[cqlsh 5.0.1 | Cassandra 3.11.4 | CQL spec 3.4.4 | Native protocol v4]
Use HELP for help.
cqlsh>

Check out the parameters reference for a complete list of all configurable settings.

Check out the "configuration" section in the "managing" page for help with changing an existing operator instance's parameters and the operating page for help with managing Cassandra operators and their underlying Cassandra clusters.

Required Permissions

KUDO Cassandra requires certain permissions in the cluster to operate. By default, it creates one service account, role and role binding in the same namespace as the installed instance. This service account has the permissions to execute commands in pods.

If the operator is configured to use a NODE_TOPOLOGY for a multi datacenter setup, additional permissions are required and explained in the corresponding section.