operator/

README

Istio Operator

The istio/operator repo is part of istio/istio from 1.5 onwards. You can contribute by picking an unassigned open issue, creating a bug or feature request, or just coming to the weekly Environments Working Group meeting to share your ideas.

This document is an overview of how the operator works from a user perspective. For more details about the design and architecture and a code overview, see ARCHITECTURE.md.

Introduction

The operator uses the IstioOperator API, which has three main components:

• MeshConfig for runtime config consumed directly by Istio control plane components.
• Component configuration API, for managing K8s settings like resources, auto scaling, pod disruption budgets and others defined in the KubernetesResourceSpec for Istio core and addon components.
• The legacy Helm installation API for backwards compatibility.

Some parameters will temporarily exist both the component configuration and legacy Helm APIs - for example, K8s resources. However, the Istio community recommends using the first API as it is more consistent, is validated, and will naturally follow the graduation process for APIs while the same parameters in the configuration API are planned for deprecation.

Profiles, are provided as a starting point for an Istio install and can be customized by creating customization overlay files or passing parameters through the --set flag. For example, to select the minimal profile:

# minimal.yaml

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  profile: minimal

If you don't specify a configuration profile, Istio is installed using the default configuration profile. All profiles listed in istio.io are available by default, or profile: can point to a local file path to reference a custom profile base to use as a starting point for customization. See the API reference for details.

Developer quick start

The quick start describes how to install and use the operator mesh CLI command and/or controller.

CLI

To build the operator CLI, simply:

make build

Ensure the created binary is in your PATH to run the examples below.

Controller (in cluster)

Building a custom controller requires a Dockerhub (or similar) account. To build using the container based build:

HUB=docker.io/<your-account> TAG=latest make docker.operator

This builds the controller binary and docker file, and pushes the image to the specified hub with the latest tag. Once the images are pushed, configure kubectl to point to your cluster and install the controller.

Install the controller manifest:

istioctl operator init --hub docker.io/<your-account> --tag latest
kubectl create ns istio-system
kubectl apply -f operator/samples/default-install.yaml

This installs the controller into the cluster in the istio-operator namespace. The controller in turns installs the Istio control plane into the istio-system namespace by default.

Controller (running locally)

1. Set env $WATCH_NAMESPACE (default value is "istio-system") and $LEADER_ELECTION_NAMESPACE (default value is "istio-operator")

2. Create the WATCH_NAMESPACE and LEADER_ELECTION_NAMESPACE if they are not created yet.

kubectl create ns $WATCH_NAMESPACE --dry-run -o yaml | kubectl apply -f -
kubectl create ns $LEADER_ELECTION_NAMESPACE --dry-run -o yaml | kubectl apply -f -

1. From the istio repo root directory, run go run ./operator/cmd/operator/*.go server

To use Remote debugging with IntelliJ, replace above step 2 with following:

1. From ./operator/cmd/operator path run dlv debug --headless --listen=:2345 --api-version=2 -- server.

2. In IntelliJ, create a new Go Remote debug configuration with default settings.

3. Start debugging process and verify it is working. For example, try adding a breakpoint at Reconcile logic and apply a new CR.

Relationship between the CLI and controller

The CLI and controller share the same API and codebase for generating manifests from the API. You can think of the controller as the CLI command istioctl install running in a loop in a pod in the cluster and using the config from the in-cluster IstioOperator custom resource (CR). There are two major differences:

1. The controller does not accept any dynamic user config through flags. All user interaction is through the IstioOperator CR.
2. The controller has additional logic that mirrors istioctl commands like upgrade, but is driven from the declarative API rather than command line.

Quick tour of CLI commands

Flags

The istioctl command supports the following flags:

• dry-run: console output only, nothing applied to cluster or written to files.
• verbose: display entire manifest contents and other debug info (default is false).
• set: select profile or override profile defaults

Basic default manifest

The following command generates a manifest with the compiled-in default profile and charts:

istioctl manifest generate

You can see these sources for the compiled-in profiles and charts in the repo under manifests/. These profiles and charts are also included in the Istio release tar.

Output to dirs

The output of the manifest is concatenated into a single file. To generate a directory hierarchy with subdirectory levels representing a child dependency, use the following command:

istioctl manifest generate -o istio_manifests

Use depth first search to traverse the created directory hierarchy when applying your YAML files. This is needed for correct sequencing of dependencies. Child manifest directories must wait for their parent directory to be fully applied, but not their sibling manifest directories.

Just apply it for me

The following command generates the manifests and applies them in the correct dependency order, waiting for the dependencies to have the needed CRDs available:

istioctl install

Review the values of a configuration profile

The following commands show the values of a configuration profile:

# show available profiles
istioctl profile list

# show the values in demo profile
istioctl profile dump demo

# show the values after a customization file is applied
istioctl profile dump -f samples/pilot-k8s.yaml

# show differences between the default and demo profiles
istioctl profile dump default > 1.yaml
istioctl profile dump demo > 2.yaml
istioctl profile diff 1.yaml 2.yaml

# show the differences in the generated manifests between the default profile and a customized install
istioctl manifest generate > 1.yaml
istioctl manifest generate -f samples/pilot-k8s.yaml > 2.yaml
istioctl manifest diff 1.yam1 2.yaml

The profile dump sub-command supports a couple of useful flags:

• config-path: select the root for the configuration subtree you want to see e.g. just show Pilot:

istioctl profile dump --config-path components.pilot

• filename: set parameters in the configuration file before dumping the resulting profile e.g. show the pilot k8s overlay settings:

istioctl profile dump --filename samples/pilot-k8s.yaml

Select a specific configuration profile

The simplest customization is to select a profile different to default e.g. minimal. See manifests/profiles/minimal.yaml:

# minimal-install.yaml
apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  profile: minimal

Use istioctl to generate the manifests for the new configuration profile:

istioctl manifest generate -f manifests/profiles/minimal.yaml

After running the command, the Helm charts are rendered using manifests/profiles/minimal.yaml.

--set syntax

The CLI --set option can be used to override settings within the profile.

For example, to enable auto mTLS, use istioctl manifest generate --set values.global.mtls.auto=true --set values.global.controlPlaneSecurityEnabled=true

To override a setting that includes dots, escape them with a backslash (). Your shell may require enclosing quotes.

istioctl manifest generate --set "values.sidecarInjectorWebhook.injectedAnnotations.container\.apparmor\.security\.beta\.kubernetes\.io/istio-proxy=runtime/default"

To override a setting that is part of a list, use brackets.

istioctl manifest generate --set values.gateways.istio-ingressgateway.enabled=false \
--set values.gateways.istio-egressgateway.enabled=true \
--set 'values.gateways.istio-egressgateway.secretVolumes[0].name'=egressgateway-certs \
--set 'values.gateways.istio-egressgateway.secretVolumes[0].secretName'=istio-egressgateway-certs \
--set 'values.gateways.istio-egressgateway.secretVolumes[0].mountPath'=/etc/istio/egressgateway-certs

Install from file path

The compiled in charts and profiles are used by default, but you can specify a file path, for example:

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  profile: /usr/home/bob/go/src/github.com/ostromart/istio-installer/data/profiles/default.yaml
  installPackagePath: /usr/home/bob/go/src/github.com/ostromart/istio-installer/data/charts/

You can mix and match these approaches. For example, you can use a compiled-in configuration profile with charts in your local file system.

Check diffs of manifests

The following command takes two manifests and output the differences in a readable way. It can be used to compare between the manifests generated by operator API and helm directly:

istioctl manifest diff ./out/helm-template/manifest.yaml ./out/mesh-manifest/manifest.yaml

New API customization

The new platform level installation API defines install time parameters like component and enablement and namespace, and K8s settings like resources, HPA spec etc. in a structured way. The simplest customization is to turn components on and off. For example, to turn on cni (samples/cni-on.yaml:

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  components:
    cni:
      enabled: true

The operator validates the configuration and automatically detects syntax errors. If you are using Helm values that are incompatible, the schema validation used in the operator may reject input that is valid for Helm. Each Istio component has K8s settings, and these can be overridden from the defaults using official K8s APIs rather than Istio defined schemas (samples/pilot-k8s.yaml):

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  components:
    pilot:
      k8s:
        resources:
          requests:
            cpu: 1000m # override from default 500m
            memory: 4096Mi # ... default 2048Mi
        hpaSpec:
          maxReplicas: 10 # ... default 5
          minReplicas: 2  # ... default 1
        nodeSelector: # ... default empty
          master: "true"
        tolerations: # ... default empty
        - key: dedicated
          operator: Exists
          effect: NoSchedule
        - key: CriticalAddonsOnly
          operator: Exists

The K8s settings are defined in detail in the operator API. The settings are the same for all components, so a user can configure pilot K8s settings in exactly the same, consistent way as galley settings. Supported K8s settings currently include:

• resources
• readiness probes
• replica count
• HorizontalPodAutoscaler
• PodDisruptionBudget
• pod annotations
• container environment variables
• ImagePullPolicy
• priority class name
• node selector
• toleration
• affinity and anti-affinity
• deployment strategy
• service annotations
• service spec
• pod securityContext

All of these K8s settings use the K8s API definitions, so K8s documentation can be used for reference. All K8s overlay values are also validated in the operator.

Customizing the old values.yaml API

The new platform install API above deals with K8s level settings. The remaining values.yaml parameters deal with Istio control plane operation rather than installation. For the time being, the operator just passes these through to the Helm charts unmodified (but validated through a schema). Values.yaml settings are overridden the same way as the new API, though a customized CR overlaid over default values for the selected profile. Here's an example of overriding some global level default values (samples/values-global.yaml):

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  profile: sds
  values:
    global:
      logging:
        level: "default:warning" # override from info

Values overrides can also be specified for a particular component (samples/values-pilot.yaml):

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  values:
    pilot:
      traceSampling: 0.1 # override from 1.0

Advanced K8s resource overlays

Advanced users may occasionally have the need to customize parameters (like container command line flags) which are not exposed through either of the installation or configuration APIs described in this document. For such cases, it's possible to overlay the generated K8s resources before they are applied with user-defined overlays. For example, to override some container level values in the Pilot container (samples/pilot-advanced-override.yaml):

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  components:
    pilot:
      k8s:
        overlays:
        - kind: Deployment
          name: istio-pilot
          patches:
          - path: spec.template.spec.containers.[name:discovery].args.[30m]
            value: "60m" # OVERRIDDEN
          - path: spec.template.spec.containers.[name:discovery].ports.[containerPort:8080].containerPort
            value: 8090 # OVERRIDDEN
          - path: 'spec.template.spec.volumes[100]' #push to the list
            value:
              configMap:
                name: my-config-map
              name: my-volume-name
          - path: 'spec.template.spec.containers[0].volumeMounts[100]'
            value:
              mountPath: /mnt/path1
              name: my-volume-name
        - kind: Service
          name: istio-pilot
          patches:
          - path: spec.ports.[name:grpc-xds].port
            value: 15099 # OVERRIDDEN

The user-defined overlay uses a path spec that includes the ability to select list items by key. In the example above, the container with the key-value "name: discovery" is selected from the list of containers, and the command line parameter with value "30m" is selected to be modified. The advanced overlay capability is described in more detail in the spec.

Interaction with controller

The controller shares the same API as the operator CLI, so it's possible to install any of the above examples as a CR in the cluster in the istio-system namespace and the controller will react to it with the same outcome as running istioctl install -f <path-to-custom-resource-file>.

Architecture

See ARCHITECTURE.md