Building a Docker Image for Deploying NGINX Management Suite Without Helm

Building a Docker Image for Deploying NGINX Management Suite Without Helm

Earlier this year we introduced NGINX Management Suite as our new control plane for NGINX software solutions, enabling you to configure, scale, secure and monitor user applications and REST APIs on the NGINX data plane from a single pane of glass.

NGINX Management Suite has a modular design: at its core is the Instance Manager module, which provides tracking, configuration, and visibility for your entire fleet of NGINX Open Source and NGINX Plus instances. As of this writing, API Connectivity Manager is the other available module, used to manage and orchestrate NGINX Plus running as an API gateway.

NGINX Management Suite can run on bare metal, as a Linux virtual machine, or containerized. The recommended way to deploy it on Kubernetes is using the Helm chart we provide, but for specific purposes you might need to build your own Docker image and manage its lifecycle through a custom CI/CD pipeline that doesn’t necessarily rely on Helm.

Prerequisites

We provide a GitHub repository of the resources you need to create a Docker image for NGINX Management Suite, with support for these versions of Instance Manager and API Connectivity Manager:

  • Instance Manager 2.4.0+
  • API Connectivity Manager 1.0.0+

To build the Docker image, you need:

To run the Docker image, you need:

  • A running Kubernetes cluster
  • kubectl with access to the Kubernetes cluster
  • A subscription (or 30-day free trial) for the NGINX Ingress Controller based on NGINX Plus

Building the Docker Image

Follow these instructions to build the Docker image.

Note: We have made every effort to accurately represent the NGINX Management Suite UI at the time of publication, but the UI is subject to change. Use these instructions as a reference and adapt them to the current UI as necessary.

  1. Clone the GitHub repository:

    $ git clone https://github.com/nginxinc/NGINX-Demos
    Cloning into 'NGINX-Demos'... 
    remote: Enumerating objects: 215, done. 
    remote: Counting objects: 100% (215/215), done. 
    remote: Compressing objects: 100% (137/137), done. 
    remote: Total 215 (delta 108), reused 171 (delta 64), pack-reused 0 
    Receiving objects: 100% (215/215), 2.02 MiB | 1.04 MiB/s, done. 
    Resolving deltas: 100% (108/108), done.
  2. Change to the build directory:

    $ cd NGINX-Demos/nginx-nms-docker/
  3. Log in to your account at MyF5.com and navigate to the page for NGINX Management Suite, under either Subscriptions or Trials as appropriate.
  4. Click +Licenses and Certificates, and in the Instance [License] panel that appears, click Download License under Step 3 to copy your license to your local disk.

    Screenshot of 'Instance [License]' panel on TRIALS > NGINX MANAGEMENT SUITE page at f5.com

  5. Click the download the latest version link (in Step 1) to download the NGINX Management Suite software.
  6. On the Downloads page that opens, select these values in the indicated fields for the purposes of this demo (new fields open in response to your choices):

    Screenshot of Downloads window for NGINX API Connectivity Manager software

    • Group – NGINX
    • Product Line – NGINX API Connectivity Manager
    • Product version – The latest (1.2.0 in the screenshot)
    • Linux Distribution – ubuntu
    • Distribution Version – 22.04
    • Architecture – amd64
  7. Scroll down to the Download Files section and click Download in the row for these two .deb packages: nms-api-connectivity-manager_* and nms-instance-manager_*.

    Screenshot of Download Files window for NGINX Management Suite software packages

  8. Copy the downloaded packages into the nim-files/ directory (the result looks like this):

    $ ls -al nim-files/ 
    total 87216 
    drwxrwxr-x 2 f5 f5     4096 Mon DD hh:mm . 
    drwxrwxr-x 8 f5 f5     4096 Mon DD hh:mm .. 
    -rw-r--r-- 1 f5 f5 10724974 Mon DD hh:mm nms-api-connectivity-manager_1.2.0.668430332~jammy_amd64.deb 
    -rw-r--r-- 1 f5 f5 78567074 Mon DD hh:mm nms-instance-manager_2.5.1-663136348~jammy_amd64.deb 
    -rw-rw-r-- 1 f5 f5        0 Mon DD hh:mm .placeholder
  9. Run docker ps to verify that Docker is running and then run the buildNIM.sh script to build the Docker image. The ‑n and ‑a options specify the .deb package filenames for Instance Manager and API Connectivity Manager respectively, and the ‑t option specifies the location and name of the private registry to which the image is pushed.

    $ ./scripts/buildNIM.sh -n nim-files/nms-instance-manager_2.5.1-663136348~jammy_amd64.deb -a nim-files/nms-api-connectivity-manager_1.2.0.668430332~jammy_amd64.deb -t registry.ff.lan:31005/nginx-nms:2.5.1 
    ==> Building NGINX Management Suite docker image 
    Sending build context to Docker daemon  92.19MB 
    Step 1/18 : FROM ubuntu:22.04 
    ---> a8780b506fa4 
    Step 2/18 : ARG NIM_DEBFILE 
    ---> Running in 0f2354280c34 
    Removing intermediate container 0f2354280c34
    […]
    ---> 0588a050c852 
    Step 18/18 : CMD /deployment/startNIM.sh 
    ---> Running in d0cc5466a43d 
    Removing intermediate container d0cc5466a43d 
    ---> 25117ec0410a 
    Successfully built 25117ec0410a 
    Successfully tagged registry.ff.lan:31005/nginx-nms:2.5.1 
    The push refers to repository [registry.ff.lan:31005/nginx-nms] 
    9c4918474e3a: Pushed
    42543d044dbb: Pushed
    1621b2ec0a5e: Pushed
    c6a464fc6a79: Pushed
    75fa1d3c61bb: Pushed
    3501fcf5dbd8: Pushed
    d4a221057e67: Pushed
    9ad05eafed57: Pushed
    f4a670ac65b6: Pushed
    2.5.1: digest: sha256:9a70cfdb63b71dc31ef39e4f20a1420d8202c85784cb037b45dc0e884dad74c9 size: 2425

Running NGINX Management Suite on Kubernetes

Follow these instructions to prepare the Deployment manifest and start NGINX Management Suite on Kubernetes.

  1. Base64‑encode the NGINX Management Suite license you downloaded in Step 4 of the previous section, and copy the output to the clipboard:

    $ base64 -w0 nginx-mgmt-suite.lic
    TulNRS1WZXJz...
  2. <a name="edit-nginx-nim"

    Using your favorite editor, open manifests/1.nginx-nim.yaml and make the following changes:

  3. On line 19, substitute your Docker image name for the default (your.registry.tld/nginx-nim2:tag):

    16    spec: 
    17      containers: 
    18      - name: nginx-nim2 
    19        image: registry.ff.lan:31005/nginx-nms:2.5.1
  4. In the Instance Manager Environment section, configure authentication credentials:

    • On line 28, optionally substitute a different admin account name for the default, admin.
    • On line 30, substitute the admin password for the default, nimadmin.
    • On line 32, substitute the base64‑encoded license you generated in Step 1 above for <BASE64_ENCODED_LICENSE_FILE>.
    25 env: 
    26   ### NGINX Instance Manager environment 
    27   - name: NIM_USERNAME 
    28     value: admin 
    29   - name: NIM_PASSWORD 
    30     value: nimadmin 
    31   - name: NIM_LICENSE 
    32     value: ""
  5. Check and modify files under manifests/certs to customize the TLS certificate and key used for TLS offload by setting the FQDN you want to use. By default, the scripts/nimDockerStart.sh startup script publishes the containerized NGINX Management Suite through NGINX Ingress Controller.
  6. Optionally, edit manifests/3.vs.yaml and customize the hostnames used to reach NGINX Management Suite.
  7. Start NGINX Management Suite in your Kubernetes cluster. As indicated in the trace, it runs as the nginx-nim2 pod. The script also initializes pods for ClickHouse as the backend database and Grafana for analytics visualization. For more information, see the README at the GitHub repo.

    $ ./scripts/nimDockerStart.sh start 
    namespace/nginx-nim2 created 
    ~/NGINX-NIM2-Docker/manifests/certs ~/NGINX-NIM2-Docker 
    Generating a RSA private key 
    .....................................+++++ 
    .....................................+++++ 
    writing new private key to 'nim2.f5.ff.lan.key' 
    ----- 
    secret/nim2.f5.ff.lan created 
    configmap/clickhouse-conf created 
    configmap/clickhouse-users created 
    persistentvolumeclaim/pvc-clickhouse created 
    deployment.apps/clickhouse created 
    service/clickhouse created 
    deployment.apps/nginx-nim2 created 
    service/nginx-nim2 created 
    service/nginx-nim2-grpc created 
    persistentvolumeclaim/pvc-grafana-data created 
    persistentvolumeclaim/pvc-grafana-log created 
    deployment.apps/grafana created 
    service/grafana created 
    virtualserver.k8s.nginx.org/nim2 created 
    virtualserver.k8s.nginx.org/grafana created 
    ~/NGINX-NIM2-Docker
  8. Verify that three pods are now running:

    $ kubectl get pods -n nginx-nim2 
    NAME                        READY     STATUS    RESTARTS   AGE 
    clickhouse-759b65db8c-74pn5   1/1     Running   0          63s 
    grafana-95fbbf5c-jczgk        1/1     Running   0          63s 
    nginx-nim2-5f54664754-lrhmn   1/1     Running   0          63s

Accessing NGINX Management Suite

To access NGINX Management Suite, navigate in a browser to https://nim2.f5.ff.lan (or the alternate hostname you set in Step 4 of the previous section). Log in using the credentials you set in Step 2 of the previous section.

Stopping NGINX Management Suite

To stop and remove the Docker instance of NGINX Management Suite, run this command:

$ ./scripts/nimDockerStart.sh stop 
namespace "nginx-nim2" deleted

The post Building a Docker Image for Deploying NGINX Management Suite Without Helm appeared first on NGINX.

Source: Building a Docker Image for Deploying NGINX Management Suite Without Helm

About KENNETH 19694 Articles
지락문화예술공작단

Be the first to comment

Leave a Reply

Your email address will not be published.


*


이 사이트는 스팸을 줄이는 아키스밋을 사용합니다. 댓글이 어떻게 처리되는지 알아보십시오.