Configuration and secrets

In this chapter, we will show how to use and store sensitive and non-sensitive application configurations properly.

In previous chapters, we added the configuration directly to containers during the build or used the container’s environment variables to pass parameters during the deployment.

Now, you will learn how to store application parameters in ConfigMaps and Secrets for security and flexibility. We will show how you can use Helm chart values and werf secrets and discuss parameterization and configuration reuse approaches. In addition, you will learn how to store sensitive data along with the code in the application’s Git repository.

The application described in this chapter is not intended for use in production environments as-is. Note that successful completion of this entire guide is required to create a production-ready application.

Preparing the environment

If you haven’t prepared your environment during previous steps, please, do it using the instructions provided in the “Preparing the environment” chapter.

If your environment has stopped working or instructions in this chapter don’t work, please, refer to these hints:

docker run hello-world

Unable to find image 'hello-world:latest' locally
latest: Pulling from library/hello-world
b8dfde127a29: Pull complete
Digest: sha256:9f6ad537c5132bcce57f7a0a20e317228d382c3cd61edae14650eec68b2b345c
Status: Downloaded newer image for hello-world:latest

Hello from Docker!
This message shows that your installation appears to be working correctly.

docker run hello-world

Unable to find image 'hello-world:latest' locally
latest: Pulling from library/hello-world
b8dfde127a29: Pull complete
Digest: sha256:9f6ad537c5132bcce57f7a0a20e317228d382c3cd61edae14650eec68b2b345c
Status: Downloaded newer image for hello-world:latest

Hello from Docker!
This message shows that your installation appears to be working correctly.

sudo systemctl restart docker

sudo systemctl status docker

● docker.service - Docker Application Container Engine
     Loaded: loaded (/lib/systemd/system/docker.service; enabled; vendor preset: enabled)
     Active: active (running) since Thu 2021-06-24 13:05:17 MSK; 13s ago
TriggeredBy: ● docker.socket
       Docs: https://docs.docker.com
   Main PID: 2013888 (dockerd)
      Tasks: 36
     Memory: 100.3M
     CGroup: /system.slice/docker.service
             └─2013888 /usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock

dockerd[2013888]: time="2021-06-24T13:05:16.936197880+03:00" level=warning msg="Your kernel does not support CPU realtime scheduler"
dockerd[2013888]: time="2021-06-24T13:05:16.936219851+03:00" level=warning msg="Your kernel does not support cgroup blkio weight"
dockerd[2013888]: time="2021-06-24T13:05:16.936224976+03:00" level=warning msg="Your kernel does not support cgroup blkio weight_device"
dockerd[2013888]: time="2021-06-24T13:05:16.936311001+03:00" level=info msg="Loading containers: start."
dockerd[2013888]: time="2021-06-24T13:05:17.119938367+03:00" level=info msg="Loading containers: done."
dockerd[2013888]: time="2021-06-24T13:05:17.134054120+03:00" level=info msg="Daemon has completed initialization"
systemd[1]: Started Docker Application Container Engine.
dockerd[2013888]: time="2021-06-24T13:05:17.148493957+03:00" level=info msg="API listen on /run/docker.sock"

docker run hello-world

Unable to find image 'hello-world:latest' locally
latest: Pulling from library/hello-world
b8dfde127a29: Pull complete
Digest: sha256:9f6ad537c5132bcce57f7a0a20e317228d382c3cd61edae14650eec68b2b345c
Status: Downloaded newer image for hello-world:latest

Hello from Docker!
This message shows that your installation appears to be working correctly.

minikube start

kubectl config set-context minikube --namespace=werf-guide-app

😄  minikube v1.20.0 on Ubuntu 20.04
✨  Using the docker driver based on existing profile
👍  Starting control plane node minikube in cluster minikube
🚜  Pulling base image ...
🎉  minikube 1.21.0 is available! Download it: https://github.com/kubernetes/minikube/releases/tag/v1.21.0
💡  To disable this notice, run: 'minikube config set WantUpdateNotification false'

🔄  Restarting existing docker container for "minikube" ...
🐳  Preparing Kubernetes v1.20.2 on Docker 20.10.6 ...
🔎  Verifying Kubernetes components...
    ▪ Using image gcr.io/google_containers/kube-registry-proxy:0.4
    ▪ Using image k8s.gcr.io/ingress-nginx/controller:v0.44.0
    ▪ Using image registry:2.7.1
    ▪ Using image docker.io/jettech/kube-webhook-certgen:v1.5.1
    ▪ Using image docker.io/jettech/kube-webhook-certgen:v1.5.1
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v5
🔎  Verifying registry addon...
🔎  Verifying ingress addon...
🌟  Enabled addons: storage-provisioner, registry, default-storageclass, ingress
🏄  Done! kubectl is now configured to use "minikube" cluster and "werf-guide-app" namespace by default

Restarting existing docker container for "minikube"

minikube tunnel --cleanup=true

minikube start --namespace werf-guide-app

kubectl config set-context minikube --namespace=werf-guide-app

😄  minikube v1.20.0 on Ubuntu 20.04
✨  Using the docker driver based on existing profile
👍  Starting control plane node minikube in cluster minikube
🚜  Pulling base image ...
🎉  minikube 1.21.0 is available! Download it: https://github.com/kubernetes/minikube/releases/tag/v1.21.0
💡  To disable this notice, run: 'minikube config set WantUpdateNotification false'

🔄  Restarting existing docker container for "minikube" ...
🐳  Preparing Kubernetes v1.20.2 on Docker 20.10.6 ...
🔎  Verifying Kubernetes components...
    ▪ Using image gcr.io/google_containers/kube-registry-proxy:0.4
    ▪ Using image k8s.gcr.io/ingress-nginx/controller:v0.44.0
    ▪ Using image registry:2.7.1
    ▪ Using image docker.io/jettech/kube-webhook-certgen:v1.5.1
    ▪ Using image docker.io/jettech/kube-webhook-certgen:v1.5.1
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v5
🔎  Verifying registry addon...
🔎  Verifying ingress addon...
🌟  Enabled addons: storage-provisioner, registry, default-storageclass, ingress
🏄  Done! kubectl is now configured to use "minikube" cluster and "werf-guide-app" namespace by default

Restarting existing docker container for "minikube"

kubectl create namespace werf-guide-app
kubectl create secret docker-registry registrysecret \
  --docker-server='https://index.docker.io/v1/' \
  --docker-username='<Docker Hub username>' \
  --docker-password='<Docker Hub password>'

namespace/werf-guide-app created
secret/registrysecret created

Preparing the repository

cd ~/werf-guide/app

# To see what changes we will make later in this chapter, let's replace all the application files
# in the repository with new, modified files containing the changes described below.
git rm -r .
cp -Recurse -Force ~/werf-guide/guides/examples/rails/080_configuration/* .
git add .
git commit -m WIP

# Enter the command below to show the files we are going to change.
git show --stat
# Enter the command below to show the changes that will be made.
git show

cd ~/werf-guide/app

# To see what changes we will make later in this chapter, let's replace all the application files
# in the repository with new, modified files containing the changes described below.
git rm -r .
cp -rf ~/werf-guide/guides/examples/rails/080_configuration/. .
git add .
git commit -m WIP

# Enter the command below to show files we are going to change.
git show --stat
# Enter the command below to show the changes that will be made.
git show

# Clone the example repository to ~/werf-guide/guides (if you have not cloned it yet).
if (-not (Test-Path ~/werf-guide/guides)) {
  git clone https://github.com/werf/website $env:HOMEPATH/werf-guide/guides
}

# Copy the (unchanged) application files to ~/werf-guide/app.
rm -Recurse -Force ~/werf-guide/app
cp -Recurse -Force ~/werf-guide/guides/examples/rails/050_s3 ~/werf-guide/app

# Make the ~/werf-guide/app directory a git repository.
cd ~/werf-guide/app
git init
git add .
git commit -m initial

# To see what changes we will make later in this chapter, let's replace all the application files
# in the repository with new, modified files containing the changes described below.
git rm -r .
cp -Recurse -Force ~/werf-guide/guides/examples/rails/080_configuration/* .
git add .
git commit -m WIP

# Enter the command below to show the files we are going to change.
git show --stat
# Enter the command below to show the changes that will be made.
git show

# Clone the example repository to ~/werf-guide/guides (if you have not cloned it yet).
test -e ~/werf-guide/guides || git clone https://github.com/werf/website ~/werf-guide/guides

# Copy the (unchanged) application files to ~/werf-guide/app.
rm -rf ~/werf-guide/app
cp -rf ~/werf-guide/guides/examples/rails/050_s3 ~/werf-guide/app

# Make the ~/werf-guide/app directory a git repository.
cd ~/werf-guide/app
git init
git add .
git commit -m initial

# To see what changes we will make later in this chapter, let's replace all the application files
# in the repository with new, modified files containing the changes described below.
git rm -r .
cp -rf ~/werf-guide/guides/examples/rails/080_configuration/. .
git add .
git commit -m WIP

# Enter the command below to show files we are going to change.
git show --stat
# Enter the command below to show the changes that will be made.
git show

ConfigMap and Secret

ConfigMap and Secret Kubernetes resources allow you to separate environment-dependent and context-specific configuration from container images.

Both objects store data in key-value pairs and provide them to Pods as environment variables, command-line arguments, or configuration files mounted in a selected container.

ConfigMap stores non-confidential data, while Secret stores confidential data (various secret types).

You can learn more about these resource types in the official Kubernetes documentation (ConfigMaps, Secrets). This chapter discusses various use cases for ConfigMaps and Secrets using our application as an example.

Storing application’s config files in the ConfigMap

Currently, the nginx.conf config file gets copied into the image during the build. As a result, the image is rebuilt, and Pods are restarted each time the file is modified. In addition, there is no easy way to benefit from templating in our nginx.conf.

You can get around these obstacles by moving .werf/nginx.conf into a dedicated ConfigMap. This way, you can mount nginx.conf when deploying instead of adding the file during the build:

apiVersion: v1
kind: ConfigMap
metadata:
  name: nginx-conf
data:
  nginx.conf: |
    user nginx;
    worker_processes 1;
    pid /run/nginx.pid;

    events {
      worker_connections 1024;
    }

    http {
      include /etc/nginx/mime.types;
      default_type application/octet-stream;

      upstream backend {
        server 127.0.0.1:3000 fail_timeout=0;
      }

      server {
        listen 80;
        server_name _;

        root /www;

        client_max_body_size 100M;
        keepalive_timeout 10s;

        # For the /packs path, assets are served directly from the NGINX container’s file system.
        location /packs/ {
          # Due to the nature of the Webpacker's bundling mechanism, the client can keep an asset cache
          # for as long as needed without worrying about invalidation.
          expires 1y;
          add_header Cache-Control public;
          add_header Last-Modified "";
          add_header ETag "";

          # Serve pre-compressed files if possible (instead of compressing them on the fly).
          gzip_static on;

          access_log off;

          try_files $uri =404;
        }

        # The media assets (pictures, etc.) will also be retrieved from the NGINX container’s file system,
        # however, we will disable gzip compression for them.
        location /packs/media {
          expires 1y;
          add_header Cache-Control public;
          add_header Last-Modified "";
          add_header ETag "";

          access_log off;

          try_files $uri =404;
        }

        # All requests, except for asset ones, are routed to the Puma backend.
        location / {
          proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
          proxy_set_header X-Real-IP $remote_addr;
          proxy_set_header X-Forwarded-Proto $scheme;
          proxy_set_header Host $http_host;
          proxy_redirect off;

          proxy_pass http://backend;
        }
      }
    }

Now, let’s add our ConfigMap to the Deployment — we will mount it as a file in the frontend container:

...
- name: frontend
  image: {{ .Values.werf.image.frontend }}
  ports:
  - containerPort: 80
  volumeMounts:
  - mountPath: /etc/nginx/nginx.conf
    subPath: nginx.conf
    name: nginx-conf
volumes:
- name: nginx-conf
  configMap:
    name: nginx-conf

We also need to delete the no longer needed .werf/nginx.conf as well as a command to copy this file to an image during the build. The Dockerfile for the frontend image will look as follows:

...
# Add an NIGINX image with the pre-built assets.
FROM nginx:stable-alpine as frontend
WORKDIR /www

# Copy the pre-built assets from the above image.
COPY --from=assets /app/public /www

Handling ConfigMap and Secret changes

Note that by default, changes in ConfigMaps or Secrets mounted in a Deployment, StatefulSet, or DaemonSet will not trigger Pod restarts for the new configuration to take effect. To trigger Pod restart, you need to add annotations with checksums of all ConfigMaps and Secrets used by this Pod. In this case, the annotations will change in response to ConfigMap and Secret modifications, leading to the Pod restart. Below is an example of an annotation that includes the checksum of the nginx.conf ConfigMap:

...
template:
  metadata:
    labels:
      app: werf-guide-app
    annotations:
      checksum/configmap-nginx: '{{ include (print $.Template.BasePath "/configmap-nginx.yaml") . | sha256sum }}'

Note that you will need a unique annotation for each mounted ConfigMap/Secret.

The authors of this tutorial prefer to use operators like stakater/Reloader instead of checksum-containing annotations because they are simpler, more flexible, and easier to work with.

Values

Using a Helm chart to deploy an application has many advantages — for example, you can create manifest templates with Helm charts. Values is the key built-in templating object. With it, you can access values passed to the chart from various sources.

When using werf, all data passed to the chart can be grouped into several categories:

• User-defined regular values: values.yaml parameters and corresponding options.
• User-defined secret values: parameters from the secret-values.yaml file.
• Service values: information about the project, release, built images, etc.

Reusing configuration with Values and ConfigMaps

Frequently used parameters can be put into .helm/values.yaml:

...
mysql:
  storageSize: "100Mi"

… to make manifest configuration easier:

...
volumeClaimTemplates:
- metadata:
    name: mysql-data
  spec:
    accessModes: ["ReadWriteOnce"]
    resources:
      requests:
        storage: "{{ .Values.mysql.storageSize }}"

This way, you can adjust the parameters to the specific environment (we will discuss this in more detail in the following tutorial chapters).

Probably, the most useful feature is the ability to place repetitive configuration snippets into .helm/values.yaml. For example, you can store in this file environment variables used in multiple places:

...
app:
  envs:
    RAILS_ENV: "production"

The environment variables from .Values.app.envs can be inserted either into the container manifest as env values (as we did before), or you can put them into the ConfigMap and load it into the container via envFrom.

The first option is easier to do, but the ConfigMap option is more convenient if you have a large number of shared environment variables. This way, you will avoid duplicating them between controllers. Your ConfigMap may look like this:

apiVersion: v1
kind: ConfigMap
metadata:
  name: app-envs
data:
  {{- range $key, $val := .Values.app.envs }}
  "{{ $key }}": "{{ $val }}"
  {{- end }}

Now you can use envFrom in the Deployment to define ConfigMap’s data as container environment variables:

...
containers:
- name: backend
  image: {{ .Values.werf.image.backend }}
  command: ["bundle", "exec", "rails", "server"]
  ports:
  - containerPort: 3000
  envFrom:
  - configMapRef:
      name: app-envs

Don’t forget to add checksum-containing annotations to trigger Pod re-deployment in response to ConfigMap changes:

...
template:
  metadata:
    labels:
      app: werf-guide-app
    annotations:
      checksum/configmap-nginx: '{{ include (print $.Template.BasePath "/configmap-nginx.yaml") . | sha256sum }}'
      checksum/configmap-app-envs: '{{ include (print $.Template.BasePath "/configmap-app-envs.yaml") . | sha256sum }}'

Similarly, you can load this ConfigMap into other resources that use the same environment variables.

Handling confidential data using Values and Secrets

To start working with secrets, you first need to generate a symmetric encryption key. You can use the following command to do this: werf helm secret generate-secret-key. However, since we have already prepared/encrypted the secrets for you, the encryption key that will be used in this chapter is generated as well. This encryption key is stored in the repository in the .werf_secret_key file; werf uses it automatically.

Caution! You MUST NOT store the key in the repository when working with real-life applications. We recommend keeping it in a safe place and passing via the WERF_SECRET_KEY environment variable. You can read more about using encryption keys in the werf documentation.

The application configuration contains data that should not be stored in unencrypted form in the repository (e.g., the database login and password). Thus, we won’t store the login and password in plain text in the configuration file (as we currently do) database.yml. Instead, we will store them encrypted in the .helm/secret-values.yaml file along with other confidential parameters:

app:
  secretEnvs:
    S3_USERNAME: 10008f28d41aa1f99901bd9581863d4668cd9ffde5e9262d5495cf7acab027903721
    S3_PASSWORD: 10007146b848eb0c24116ef18377394efceb096ee2718208dd7a9c209fe3a8f95c07
    DB_USERNAME: 1000bcd85fafffbaaec5bc11152307b7503e9764e737264ab7bc44240652394e374d
    DB_PASSWORD: 1000687e9b8472b98dc7395b09b7fdc628cdff5b6dc1e1a548ba6bffe76c30874f4b
    SECRET_KEY_BASE: 1000921032b2b78973963c1a2d0b6bc5ec98bfbde55cbd4c49c32ebaed73f8d6a736
mysql:
  secretEnvs:
    MYSQL_ROOT_PASSWORD: 1000f29b2bd6877c2f690d7fe82822a590bfd154db2d947102eb4341722cf564f635
minio:
  secretEnvs:
    MINIO_ROOT_USER: 1000ee40b96582f838ebbf4fb245ea15d912f0a1e76e3ba8461078a3078c06764315
    MINIO_ROOT_PASSWORD: 1000b7467033b6227b7a2052ac60e16e34382f26dee78f4015276b01b1b5c25a887a

Use the command below to decrypt the secrets stored in .helm/secret-values.yaml:

werf helm secret values decrypt .helm/secret-values.yaml

You should see the following output:

app:
  secretEnvs:
    S3_USERNAME: minioadmin
    S3_PASSWORD: minioadmin
    DB_USERNAME: root
    DB_PASSWORD: password
    SECRET_KEY_BASE: something
mysql:
  secretEnvs:
    MYSQL_ROOT_PASSWORD: password
minio:
  secretEnvs:
    MINIO_ROOT_USER: minioadmin
    MINIO_ROOT_PASSWORD: minioadmin

Now, you need to pass the secrets contained in .helm/secret-values.yaml back to the application config. To do this, let’s first pass them to the Secret resource:

apiVersion: v1
kind: Secret
metadata:
  name: app-envs
type: Opaque
data:
  {{- range $key, $val := .Values.app.secretEnvs }}
  "{{ $key }}": "{{ $val | b64enc }}"
  {{- end }}

… and then load this resource’s data as a set of environment variables by mounting it in containers:

...
containers:
- name: backend
  image: {{ .Values.werf.image.backend }}
  command: ["bundle", "exec", "rails", "server"]
  ports:
  - containerPort: 3000
  envFrom:
  - configMapRef:
      name: app-envs
  - secretRef:
      name: app-envs

You have to make similar changes to the rest of the .helm/templates files (we won’t delve into details here; you can take a look at their contents in the repository.)

After passing environment variables to the container, you have to substitute them in the application’s configuration files:

production:
  adapter: mysql2
  pool: 5
  host: mysql
  port: 3306
  username: <%= ENV["DB_USERNAME"] %>
  password: <%= ENV["DB_PASSWORD"] %>
  database: werf-guide-app
  encoding: utf8

minio:
  service: S3
  endpoint: http://minio:9000
  region: us-east-1
  bucket: werf-guide-app
  access_key_id: <%= ENV["S3_USERNAME"] %>
  secret_access_key: <%= ENV["S3_PASSWORD"] %>
  force_path_style: true

secret_key_base: <%= ENV["SECRET_KEY_BASE"] %>

You can also use Secrets to store and mount complete secret configuration files. This is similar to using ConfigMap to mount non-secret configuration files (as described above), except that the Secret contents must be stored encrypted in .helm/secret/... or .helm/secret-values.yaml. You can read more about this in the werf documentation.

Note: learn more about working with secrets in the werf documentation.

Checking whether the application runs as expected

Let’s make sure the configuration changes did not affect the application performance:

werf converge --repo <DOCKER HUB USERNAME>/werf-guide-app

You should see the following output:

...
┌ ⛵ image backend
│ Use cache image for backend/dockerfile
│      name: <DOCKER HUB USERNAME>/werf-guide-app:8df39a540816b5851314934d932f0152a8fa409369a58dbfeccce4be-1632852281567
│        id: 0825950bfd70
│   created: 2022-09-28 21:04:41 +0000 UTC
│      size: 366.3 MiB
└ ⛵ image backend (7.19 seconds)

┌ ⛵ image frontend
│ Use cache image for frontend/dockerfile
│      name: <DOCKER HUB USERNAME>/werf-guide-app:18b3a25a107798890200664b7de037c998bef8ed9d9bbe4d66efaf53-1632852281351
│        id: 7d029b84655e
│   created: 2022-09-28 21:04:41 +0000 UTC
│      size: 9.5 MiB
└ ⛵ image frontend (7.60 seconds)

┌ Waiting for release resources to become ready
...
│ ┌ Status progress
│ │ DEPLOYMENT      REPLICAS  AVAILABLE  UP-TO-DATE
│ │ werf-guide-app  1/1       1          1
│ │ │    POD                         READY  RESTARTS  STATUS
│ │ ├──  guide-app-85bd8c68f5-w9nff  2/2    0         Running
│ │ └──  guide-app-865b6d68bc-bkt9l  2/2    0         Terminating
│ │ STATEFULSET  REPLICAS  READY  UP-TO-DATE
│ │ minio        1/1       1      1
│ │ │    POD  READY  RESTARTS  STATUS
│ │ └──  0    1/1    0         Running
│ │ mysql        1/1       1      1
│ │ │    POD  READY  RESTARTS  STATUS
│ │ └──  0    1/1    0         Running
│ │ JOB                        ACTIVE  DURATION  SUCCEEDED/FAILED
│ │ setup-and-migrate-db-rev2  0       24s       1/0
│ │ │    POD                        READY  RESTARTS  STATUS
│ │ └──  and-migrate-db-rev2-lhnq7  0/1    0         Completed
│ │ setup-minio-rev2           0       33s       0->1/0
│ │ │    POD               READY  RESTARTS  STATUS
│ │ └──  minio-rev2-wr7tw  0/1    0         Running  ->
│ │ Completed
│ └ Status progress
└ Waiting for release resources to become ready (33.05 seconds)

Release "werf-guide-app" has been upgraded. Happy Helming!
NAME: werf-guide-app
LAST DEPLOYED: Tue Sep 28 21:26:48 2022
NAMESPACE: werf-guide-app
STATUS: deployed
REVISION: 2
TEST SUITE: None
Running time 43.88 seconds

Check if the application is available:

curl http://werf-guide-app.test/ping

You should see the following output:

pong

In this chapter, you learned how to securely store the confidential application parameters, minimize duplicate configuration, and turn application configs into templates (thus, no re-builds are necessary when changes are made).