Quick Start for Server/VM installs
You have two options for installing DataStax Luna Streaming (Pulsar):
Via the provided Helm chart for an existing Kubernetes environment on a laptop or with a cloud provider. See Quick Start for Helm Chart installs.
Via replicated.io packaging for deployment to a single server/VM, or to multiple servers/VMs, as covered in this topic.
Luna Streaming takes care of the replicated.io packaging details and deployment for you. The resulting deployment includes:
The DataStax distribution of Apache Pulsar
Pulsar Admin Console
Prometheus Operator and Grafana Operator
Ready-made Grafana dashboards that visualize the metrics collected via Prometheus
Support for Transport Layer Security (TLS) and token-based authentication
Support for single-node and multi-node (4) installs
Luna Streaming Installation Console
A Linux server or VM
For a single node install, a server with at least 8 CPU and 32 GB of memory is required
For a small HA, 4 servers are required. The servers must be on the same network so they can communicate with each other.
The servers should have at least 50 GB in their root disk volume. If you want to retain messages or improve performance, you should add diskl volume devices to the servers to be used by BookKeeper. Ideally BookKeeper uses 2 volumes devices, one for the journal, and one for the ledgers. The journal device should be 20GB. The ledgers' device should be large enough to hold the expected amount of stored message data, such as 500 GB.
The first server must have a DNS name, which you will need to specify during the steps below.
Install the base system.
curl -sSL https://k8s.kurl.sh/luna-streaming | sudo bash
This will take about 10 minutes to complete.
Near the conclusion of the command output, be sure to note (for later steps) the URL displayed for connecting to the Luna Streaming Installation Console and the password, and other tips.
The following is sample output from an install. Your version will have different values - copy them for use in later steps:
. . . deployment.apps/cert-manager created ⚙ Addon metrics-server 0.3.7 serviceaccount/metrics-server created clusterrole.rbac.authorization.k8s.io/system:aggregated-metrics-reader created clusterrole.rbac.authorization.k8s.io/system:metrics-server created rolebinding.rbac.authorization.k8s.io/metrics-server-auth-reader created clusterrolebinding.rbac.authorization.k8s.io/metrics-server:system:auth-delegator created clusterrolebinding.rbac.authorization.k8s.io/system:metrics-server created service/metrics-server created deployment.apps/metrics-server created apiservice.apiregistration.k8s.io/v1beta1.metrics.k8s.io created configmap/kurl-config created Installation Complete ✔ Kotsadm: http://10.101.32.250:8800 Login with password (will not be shown again): fNBFUn9AM To access the cluster with kubectl, ensure the KUBECONFIG environment variable is unset: bash -l Node join commands expire after 24 hours. To generate new node join commands, run curl -sSL https://kurl.sh/luna-streaming/tasks.sh | sudo bash -s join_token on this node. To add worker nodes to this installation, run the following script on your other nodes: curl -sSL https://kurl.sh/luna-streaming/join.sh | sudo bash -s kubernetes-master-address=10.101.32.250:6443 kubeadm-token=7sdek0.xujj2fo67qh6rs4v kubeadm-token-ca-hash=sha256:28d02e939bfb4fd8d3ea0a9619e9f47e0812773840e0c1e5a72c26b5bfbec08c kubernetes-version=1.19.7 docker-registry-ip=10.96.0.49
If you are planning a multi-node installation (small HA), add the worker nodes. The Luna Streaming configuration is for 4 nodes. The command to run on each node is displayed in the output; the sample values above are just an example. See the text for your environment following, “To add worker nodes to this installation…” in the output. If necessary, you can regenerate the node install command using the following:
curl -sSL https://kurl.sh/luna-streaming/tasks.sh | sudo bash -s join_token
Download and install the Community license.
curl -s https://downloads.datastax.com/license/luna-streaming/Community.yaml -o Community.yaml bash -l kubectl kots install luna-streaming --namespace default --license-file ./Community.yaml
If necessary on your node, preface the
kubectlcommand above with
Output from the install command:
• Deploying Admin Console • Waiting for Admin Console to be ready ✓
Open the Luna Streaming Installation Console’s URL in a browser. The URL was displayed after step 1. In the example command output above, it was http://10.101.32.250:8800. Remember to note the generated password, which you’ll need soon.
In the browser, you may see a message similar to this:
Note the browser security advice, and click the Continue to Setup button. If you encounter a "Your connection is not private" dialog, you can (if you agree) click the Advanced section and click "Proceed to <your IP address>".
The Luna Streaming Installation Console displays an "HTTPS for Luna Streaming admin console" message. It includes an explanation that we’re currently using a self-signed TLS certificate to secure the communication between your browser and the Luna Streaming Installation Console. If you don’t upload your own TLS cert, you’ll see a warning about this scenario in your browser every time you access the Luna Streaming Installation Console.
In development, if you agree, you can click Skip & continue.
Login to the Luna Streaming Installation Console. It presents options before you will launch the Luna Streaming Installation Console. The user name defaults to
admin. Enter the password that was displayed after installing the base system in step 1 above.
Due to a Replicated package default, the Luna Streaming Installation Console login screen’s text refers to the "admin console" even though DataStax calls the follow-on UI the Pulsar Admin Console. Ignore the overloaded term and know that at this point, you are logging into the Luna Streaming Installation Console.
Select the configuration options: The Luna Streaming Installation Console presents its "Configure Luna Streaming" page.
Start by selecting the configuration type: single node or small HA. If you select small HA, make sure you have previously added the required number of nodes.
In the Hostname field, enter a DNS resolvable name for the cluster. For a single node install, this should resolve to the IP address of the single node. For a multi-node install, the DNS name should resolve to a list of all the IP addresses in the cluster.
If you know your node’s IP address but not its DNS name, use an
nslookupcommand. For example, if the node’s IP is 10.101.32.250:
nslookup 10.101.32.250 Server: 10.100.6.66 Address: 10.100.6.66#53 250.32.101.10.in-addr.arpa name = ip-10-101-32-250.srv101.dsinternal.org.
The DNS name in this example is
ip-10-101-34-250.srv101.dsinternal.organd you would enter this in the required Hostname field.
Optionally enable TLS enabled for Pulsar clients. If you entered TLS information in step 6, that certificate will be used. The certificate should be signed by a trusted certificate authority (for example, Let’s Encrypt). If you want to use a self-signed certificate, select “Generate Self-Signed Certificate”.
Optionally enable token-based authentication for Pulsar admin and clients. The installation will automatically generate keys and tokens. The superuser token can be retrieved from the Pulsar Admin Console.
The BookKeeper storage settings allow you to optionally use attached volumes for the BookKeeper journal and ledgers. By default, they will use the OS volume of the servers. However, if you have attached devices, the installation will automatically detect them and map them to BookKeeper. The journal device should be 20 GB. The ledger device can be any size, but defaults to 50 GB. Make sure the attached volume device is at least as large as the specified size. If it is not large enough, it will not be successfully mapped.
Scroll down the Installation Console’s Config page. In the Admin Console Values section, if you’ll authenticate with username/password, copy the credentials for a subsequent Pulsar Admin Console login. Example:
Once you have entered the config options, click Continue.
Let the pre-flight checks complete. Example:
Once the application is deployed and the Application status is green on the Luna Streaming Installation Console’s Application tab, open the Pulsar Admin Console in a new browser window or tab.
For example, if you specified ip-10-101-32-250.srv101.dsinternal.org as the hostname in the Luna Streaming Installation Console’s Config tab, and if you did not configure TLS, launch the Pulsar Admin Console in a browser:
Or in this example, open https://ip-10-101-32-250.srv101.dsinternal.org if you configured TLS from the Luna Streaming Installation Console.
Log in to the DataStax Luna Streaming Pulsar Admin Console. The username is
admin. You can find the password in the Luna Streaming Installation Console’s Config tab, under Admin Console Values.
Work with your new Pulsar cluster. You can view/create topics, namespaces, tenants, functions, sinks, and source.
You can connect a test client (Test Clients) directly from the Pulsar Admin Console and you can view the built-in Grafana dashboards (Cluster/Monitoring). The Grafana user name is
adminand the password is the same as for the Admin Console, which can be found under Dashboard/Config tab of the installation interface in the Admin Console Value section.
DataStax Luna Streaming is a Replicated KOTS application. You can update the application by using the Version history tab of the Admin Console, or via the KOTS CLI. For KOTS details, see Updating a KOTS application.
On the Version history tab, click Check for update. If a newer version of Luna Streaming is available, consider updating the server. Once the update is ready, if you want to proceed, click Deploy. Wait a few minutes after the deployment, and check its Ready status on the Applications tab. Example:
You can also check the status of the pods from the command line. Example:
kubectl get pods
NAME READY STATUS RESTARTS AGE kotsadm-67c8d478f7-xs7mx 1/1 Running 0 32m kotsadm-migrations 0/1 Completed 0 44m kotsadm-operator-75c844ffdc-k95q6 1/1 Running 0 44m kotsadm-postgres-0 1/1 Running 0 44m kurl-proxy-kotsadm-79df797c77-jhvjb 1/1 Running 0 43m prometheus-operator-984cbc6b8-g8ggz 1/1 Running 0 8m45s prometheus-pulsar-single-kube-prometh-prometheus-0 2/2 Running 1 6m54s pulsar-autorecovery-5596f6464d-wtd22 1/1 Running 2 8m45s pulsar-bastion-5567576b79-jrvjh 1/1 Running 0 8m45s pulsar-bookkeeper-0 1/1 Running 0 8m45s pulsar-broker-77848bccb8-bj7bj 1/1 Running 2 8m45s pulsar-dashboard-6cd77957b9-mqvd6 2/2 Running 0 8m45s pulsar-function-0 2/2 Running 0 8m45s pulsar-proxy-59cc75485-88c5b 3/3 Running 0 8m45s pulsar-pulsarmonitor-55c8c5454d-bq88d 1/1 Running 0 8m45s pulsar-single-grafana-78bfcd7848-n6jqh 2/2 Running 0 8m45s pulsar-single-kube-state-metrics-9c7644667-g9n27 1/1 Running 0 8m45s pulsar-single-prometheus-node-exporter-bk52h 1/1 Running 0 8m45s pulsar-zookeeper-0 1/1 Running 0 8m45s pulsar-zookeeper-metadata-r7r42 0/1 Completed 0 8m45s
Logs can be viewed from the installation node using the Kubernetes command line interface. To view the logs for each component, use
To list all the pods, enter:
kubectl get pods
To view the logs of a specific pod, for example:
kubectl logs <podname>
If the pods has multiple containers, use the following command to view the logs for one of the containers:
kubectl logs <podname> -c <containername>