Configuration

• 1: Authentication Strategies
• 1.1: Anonymous strategy
• 1.2: Header strategy
• 1.3: OpenID Connect strategy
• 1.4: OpenShift strategy
• 1.5: Token strategy
• 1.6: Session options
• 2: Console Customization
• 3: Custom Dashboards
• 4: Istio Environment
• 5: Kiali CR Reference
• 6: Namespace Management
• 7: Prometheus, Jaeger, Grafana
• 7.1: Grafana
• 7.2: Jaeger
• 7.3: Prometheus
• 8: Role-based access control
• 9: Traffic Health
• 10: Virtual Machine workloads

spec:
  auth:
    strategy: anonymous

apiVersion: kiali.io/v1alpha1
kind: Kiali
metadata:
  namespace: kiali-namespace
  name: kiali
spec:
  istio_namespace: istio-system
  deployment:
    namespace: kiali-namespace
  auth:
    strategy: anonymous

1 - Authentication Strategies

Kiali supports five authentication mechanisms:

• The default authentication strategy for OpenShift clusters is openshift.
• The default authentication strategy for all other Kubernetes clusters is token.

All mechanisms other than anonymous support Role-based access control.

Read the dedicated page of each authentication strategy to learn more.

1.1 - Anonymous strategy

Introduction

The anonymous strategy removes any authentication requirement. Users will have access to Kiali without providing any credentials.

Although the anonymous strategy doesn’t provide any access protection, it’s valid for some use-cases. Some examples known from the community:

• Exposing Kiali through a reverse proxy, where the reverse proxy is providing a custom authentication mechanism.
• Exposing Kiali on an already limited network of trusted users.
• When Kiali is accessed through kubectl port-forward or alike commands that allow usage of the cluster’s RBAC capabilities to limit access.
• When developing Kiali, where a developer has a private instance on his own machine.

Set-up

To use the anonymous strategy, use the following configuration in the Kiali CR:

spec:
  auth:
    strategy: anonymous

The anonymous strategy doesn’t have any additional configuration.

Access control

When using the anonymous strategy, the content displayed in Kiali is based on the permissions of the Kiali service account. By default, the Kiali service account has cluster wide access and will be able to display everything in the cluster.

OpenShift

If you are running Kiali in OpenShift, access can be customized by changing privileges to the Kiali ServiceAccount. For example, to reduce permissions to individual namespaces, first, remove the cluster-wide permissions granted by default:

  oc delete clusterrolebindings kiali

Then grant the kiali role only in needed namespaces. For example:

  oc adm policy add-role-to-user kiali system:serviceaccount:istio-system:kiali-service-account -n ${NAMESPACE}

View only

You can tell the Kiali Operator to install Kiali in “view only” mode (this does work for either OpenShift or Kubernetes). You do this by setting the view_only_mode to true in the Kiali CR, which allows Kiali to read service mesh resources found in the cluster, but it does not allow any change:

spec:
  deployment:
    view_only_mode: true

1.2 - Header strategy

Introduction

The header strategy assumes a reverse proxy is in front of Kiali, such as OpenUnison or OAuth2 Proxy, injecting the user’s identity into each request to Kiali as an Authorization header. This token can be an OpenID Connect token or any other token the cluster recognizes. All requests to Kubernetes will be made with this token, allowing Kiali to use the user’s own RBAC context.

In addition to a user token, the header strategy supports impersonation headers. If the impersonation headers are present in the request, then Kiali will act on behalf of the user specified by the impersonation (assuming the token supplied in the Authorization header is authorized to do so).

The header strategy takes advantage of the cluster’s RBAC. See the Role-based access control documentation for more details.

Set-up

The header strategy will work with any Kubernetes cluster. The token provided must be supported by that cluster. For instance, most “on-prem” clusters support OpenID Connect, but cloud hosted clusters do not. For clusters that don’t support a token, the impersonation headers can be injected by the reverse proxy.

spec:
  auth:
    strategy: header

The header strategy doesn’t have any additional configuration.

HTTP Header

The header strategy looks for a token in the Authorization HTTP header with the Bearer prefix. The HTTP header should look like:

Authorization: Bearer TOKEN

Where TOKEN is the appropriate token for your cluster. This TOKEN will be submitted to the API server via a TokenReview to validate the token ONLY on the first access to Kiali. On subsequent calls the TOKEN is passed through directly to the API server.

Security Considerations

Network Policies

A policy should be put in place to make sure that the only “client” for Kiali is the authenticating reverse proxy. This helps limit potential abuse and ensures that the authenticating reverse proxy is the source of truth for who accessed Kiali.

Short Lived Tokens

The authenticating reverse proxy should inject a short lived token in the Authorization header. A shorter lived token is less likely to be abused if leaked. Kiali will take whatever token is passed into the reqeuest, so as tokens are regenerated Kiali will use the new token.

Impersonation

TokenRequest API

The authenticating reverse proxy should use the TokenRequest API instead of static ServiceAccount tokens when possible while using impersonation. The ServiceAccount that can impersonate users and groups is privileged and having it be short lived cuts down on the possibility of a token being leaked while it’s being passed between different parts of the infrastructure.

Drop Incoming Impersonation Headers

The authenticating proxy MUST drop any headers it receives from a remote client that match the impersonation headers. Not only do you want to make sure that the authenticating proxy can’t be overriden on which user to authenticate, but also what groups they’re a member of.

1.3 - OpenID Connect strategy

Introduction

The openid authentication strategy lets you integrate Kiali to an external identity provider that implements OpenID Connect, and allows users to login to Kiali using their existing accounts of a third-party system.

If your Kubernetes cluster is also integrated with your OpenId provider, then Kiali’s openid strategy can offer role-based access control (RBAC) through the Kubernetes authorization mechanisms. See the RBAC documentation for more details.

Currently, Kiali supports the authorization code flow (preferred) and the implicit flow of the OpenId Connect spec.

Requirements

If you want to enable usage of the OpenId’s authorization code flow, make sure that the Kiali’s signing key is 16, 24 or 32 byte long. If you setup a signing key of a different size, Kiali will only be capable of using the implicit flow. If you install Kiali via the operator and don’t set a custom signing key, the operator should create a 16 byte long signing key.

If you don’t need RBAC support, the only requirement is to have a working OpenId Server where Kiali can be configured as a client application.

If you do need RBAC support, you need either:

• A Kubernetes cluster configured with OpenID connect integration, which results in the API server accepting tokens issued by your identity provider.
• A replacement or reverse proxy for the Kubernetes cluster API capable of handling the OIDC authentication.

The first option is preferred if you can manipulate your cluster API server startup flags, which will result in your cluster to also be integrated with the external OpenID provider.

The second option is provided for cases where you are using a managed Kubernetes and your cloud provider does not support configuring OpenID integration. Kiali assumes an implementation of a Kubernetes API server. For example, a community user has reported to successfully configure Kiali’s OpenID strategy by using kube-oidc-proxy which is a reverse proxy that handles the OpenID authentication and forwards the authenticated requests to the Kubernetes API.

Set-up with RBAC support

Assuming you already have a working Kubernetes cluster with OpenId integration (or a working alternative like kube-oidc-proxy), you should already had configured an application or a client in your OpenId server (some cloud providers configure this app/client automatically for you). You must re-use this existing application/client by adding the root path of your Kiali instance as an allowed/authorized callback URL. If the OpenID server provided you a client secret for the application/client, or if you had manually set a client secret, issue the following command to create a Kubernetes secret holding the OpenId client secret:

kubectl create secret generic kiali --from-literal="oidc-secret=$CLIENT_SECRET" -n $NAMESPACE

where $NAMESPACE is the namespace where you installed Kiali and $CLIENT_SECRET is the secret you configured or provided by your OpenId Server. If Kiali is already running, you may need to restart the Kiali pod so that the secret is mounted in Kiali.

Then, to enable the OpenID Connect strategy, the minimal configuration you need to set in the Kiali CR is like the following:

spec:
  auth:
    strategy: openid
    openid:
      client_id: "kiali-client"
      issuer_uri: "https://openid.issuer.com"

This assumes that your Kubernetes cluster is configured with OpenID Connect integration. In this case, the client-id and issuer_uri attributes must match the --oidc-client-id and --oidc-issuer-url flags used to start the cluster API server. If these values don’t match, users will fail to login to Kiali.

If you are using a replacement or a reverse proxy for the Kubernetes API server, the minimal configuration is like the following:

spec:
  auth:
    strategy: openid
    openid:
      api_proxy: "https://proxy.domain.com:port"
      api_proxy_ca_data: "..."
      client_id: "kiali-client"
      issuer_uri: "https://openid.issuer.com"

The value of client-id and issuer_uri must match the values of the configuration of your reverse proxy or cluster API replacement. The api_proxy attribute is the URI of the reverse proxy or cluster API replacement (only HTTPS is allowed). The api_proxy_ca_data is the public certificate authority file encoded in a base64 string, to trust the secure connection.

Set-up with no RBAC support

Register Kiali as a client application in your OpenId Server. Use the root path of your Kiali instance as the callback URL. If the OpenId Server provides you a client secret, or if you manually set a client secret, issue the following command to create a Kubernetes secret holding the OpenId client secret:

kubectl create secret generic kiali --from-literal="oidc-secret=$CLIENT_SECRET" -n $NAMESPACE

where $NAMESPACE is the namespace where you installed Kiali and $CLIENT_SECRET is the secret you configured or provided by your OpenId Server. If Kiali is already running, you may need to restart the Kiali pod so that the secret is mounted in Kiali.

Then, to enable the OpenID Connect strategy, the minimal configuration you need to set in the Kiali CR is like the following:

spec:
  auth:
    strategy: openid
    openid:
      client_id: "kiali-client"
      disable_rbac: true
      issuer_uri: "https://openid.issuer.com"

Additional configurations

Configuring the displayed user name

The Kiali front-end will, by default, retrieve the string of the sub claim of the OpenID token and display it as the user name. You can customize which field to display as the user name by setting the username_claim attribute of the Kiali CR. For example:

spec:
  auth:
    openid:
      username_claim: "email"

If you enabled RBAC, you will want the username_claim attribute to match the --oidc-username-claim flag used to start the Kubernetes API server, or the equivalent option if you are using a replacement or reverse proxy of the API server. Else, any user-friendly claim will be OK as it is purely informational.

Configuring requested scopes

By default, Kiali will request access to the openid, profile and email standard scopes. If you need a different set of scopes, you can set the scopes attribute in the Kiali CR. For example:

spec:
  auth:
    openid:
      scopes:
      - "openid"
      - "email"
      - "groups"

The openid scope is forced. If you don’t add it to the list of scopes to request, Kiali will still request it from the identity provider.

Configuring authentication timeout

When the user is redirected to the external authentication system, by default Kiali will wait at most 5 minutes for the user to authenticate. After that time has elapsed, Kiali will reject authentication. You can adjust this timeout by setting the authentication_timeout with the number of seconds that Kiali should wait at most. For example:

spec:
  auth:
    openid:
      authentication_timeout: 60 # Wait only one minute.

Configuring allowed domains

Some identity providers use a shared login and regardless of configuring your own application under your domain (or organization account), login can succeed even if the user that is logging in does not belong to your account or organization. Google is an example of this kind of provider.

To prevent foreign users from logging into your Kiali instance, you can configure a list of allowed domains:

spec:
  auth:
    openid:
      allowed_domains:
      - example.com
      - foo.com

The e-mail reported by the identity provider is used for the validation. Login will be allowed if the domain part of the e-mail is listed as an allowed domain; else, the user will be rejected. Naturally, you will need to configure the email scope to be requested.

There is a special case: some identity providers include a hd claim in the id_token. If this claim is present, this is used instead of extracting the domain from the user e-mail. For example, Google Workspace (aka G Suite) includes this hd claim for hosted domains.

Using an OpenID provider with a self-signed certificate

If your OpenID provider is using a self-signed certificate, you can disable certificate validation by setting the insecure_skip_verify_tls to true in the Kiali CR:

spec:
  auth:
    openid:
      insecure_skip_verify_tls: true

However, if your organization or internal network has an internal trusted certificate authority (CA), and your OpenID server is using a certificate issued by this CA, you can configure Kiali to trust certificates from this CA, rather than disabling verification. For this, create a ConfigMap named kiali-cabundle containing the root CA certificate (the public component) under the openid-server-ca.crt key:

apiVersion: v1
kind: ConfigMap
metadata:
  name: kiali-cabundle
  namespace: istio-system # This is Kiali's install namespace
data:
  openid-server-ca.crt: <the public component of your CA root certificate encoded in base64>

After restarting the Kiali pod, Kiali will trust this root certificate for all HTTPS requests related to OpenID authentication.

Using an HTTP/HTTPS Proxy

In some network configurations, there is the need to use proxies to connect to the outside world. OpenID requires outside world connections to get metadata and do key validation, so you can configure it by setting the http_proxy and https_proxy keys in the Kiali CR. They use the same format as the HTTP_PROXY and HTTPS_PROXY environment variables.

spec:
  auth:
    openid:
      http_proxy: http://USERNAME:PASSWORD@10.0.1.1:8080/
      https_proxy: https://USERNAME:PASSWORD@10.0.0.1:8080/

Passing additional options to the identity provider

When users click on the Login button on Kiali, a redirection occurs to the authentication page of the external identity provider. Kiali sends a fixed set of parameters to the identity provider to enable authentication. If you need to add an additional set of parameters to your identity provider, you can use the additional_request_params setting of the Kiali CR, which accepts key-value pairs. For example:

spec:
  auth:
    openid:
      additional_request_params:
        prompt: login

The prompt parameter is a standard OpenID parameter. When the login value is passed in this parameter, the identity provider is instructed to ask for user credentials regardless if the user already has an active session because of a previous login in some other system.

If your OpenId provider supports other non-standard parameters, you can specify the ones you need in this additional_request_params setting.

Take into account that you should not add the client_id, response_type, redirect_uri, scope, nonce nor state parameters to this list. These are already in use by Kiali and some already have a dedicated setting.

Provider-specific instructions

Using with Keycloak

When using OpenId with Keycloak, you will need to enable the Standard Flow Enabled option on the Client (in the Administration Console):

The Standard Flow described on the options is the same as the authorization code flow from the rest of the documentation.

If you get an error like Client is not allowed to initiate browser login with given response_type. Implicit flow is disabled for the client., it means that your signing key for Kiali is not a standard size (16, 24 or 32 bytes long).

Enabling the Implicit Flow Enabled option of the client will make the problem go away, but be aware that the implicit flow is less secure, and not recommended.

Using with Google Cloud Platform / GKE OAuth2

If you are using Google Cloud Platform (GCP) and its products such as Google Kubernetes Engine (GKE), it should be straightforward to configure Kiali’s OpenID strategy to authenticate using your Google credentials.

First, you’ll need to go to your GCP Project and to the Credentials screen which is available at (Menu Icon) > APIs & Services > Credentials.

On the Credentials screen you can select to create a new OAuth client ID.

On the Create OAuth client ID screen, set the Application type to Web Application and enter a name for your key.

Then enter in the Authorized Javascript origins and Authorized redirect URIs for your project. You can enter in localhost as appropriate during testing. You can also enter multiple URIs as appropriate.

After clicking Create you’ll be shown your newly minted client id and secret. These are important and needed for your Kiali CR yaml and Kiali secrets files.

You’ll need to update your Kiali CR file to include the following auth block.

spec:
  auth:
    strategy: "openid"
    openid:
      client_id: "<your client id from GCP>"
      disable_rbac: true
      issuer_uri: "https://accounts.google.com"
      scopes: ["openid", "email"]
      username_claim: "email"

Finally you will need to create a secret, if you don’t have one already, that sets the oidc-secret for the openid flow.

apiVersion: v1
kind: Secret
metadata:
  name: kiali
  namespace: istio-system
  labels:
    app: kiali
type: Opaque
data:
  oidc-secret: "<base64 encode your client secret from GCP and enter here>"

Once all these settings are complete just set your Kiali CR and the Kiali secret to your cluster. You may need to refresh your Kiali Pod to set the Secret if you add the Secret after the Kiali pod is created.

Using with Azure: AKS and AAD

AKS has support for a feature named AKS-managed Azure Active Directory, which enables integration between AKS and AAD. This has the advantage that users can use their AAD credentials to access AKS clusters and can also use Kubernetes RBAC features to assign privileges to AAD users.

However, Azure is implementing this integration via the Kubernetes Webhook Token Authentication rather than via the Kubernetes OpenID Connect Tokens authentication (see the Azure AD integration section in AKS Concepts documentation). Because of this difference, authentication in AKS behaves slightly different from a standard OpenID setup, but Kiali’s OpenID authentication strategy can still be used with full RBAC support by following the next steps.

First, enable the AAD integration on your AKS cluster. See the official AKS documentation to learn how. Once it is enabled, your AKS panel should show the following:

Create a web application for Kiali in your Azure AD panel:

1. Go to AAD > App Registration, create an application with a redirect url like \https://<your-kiali-url>
2. Go to Certificates & secrets and create a client secret.
   1. After creating the client secret, take note of the provided secret. Create a Kubernetes secret in your cluster as mentioned in the Set-up with RBAC support section. Please, note that the suggested name for the Kubernetes Secret is kiali. If you want to customize the secret name, you will have to specify your custom name in the Kiali CR. See: secret_name in Kial CR Reference.
3. Go to API Permissions and press the Add a permission button. In the new page that appears, switch to the APIs my organization uses tab.
4. Type the following ID in the search field: 6dae42f8-4368-4678-94ff-3960e28e3630 (this is a shared ID for all Azure clusters). And select the resulting entry.
5. Select the Delegated permissions square.
6. Select the user.read permission.
7. Go to Authentication and make sure that the Access tokens checkbox is ticked.

Then, create or modify your Kiali CR and include the following settings:

spec:
  auth:
    strategy: "openid"
    openid:
      client_id: "<your Kiali application client id from Azure>"
      issuer_uri: "https://sts.windows.net/<your AAD tenant id>/"
      username_claim: preferred_username
      api_token: access_token
      additional_request_params:
        resource: "6dae42f8-4368-4678-94ff-3960e28e3630"

You can find your client_id and tenant_id in the Overview page of the Kiali App registration that you just created. See this documentation for more information.

1.4 - OpenShift strategy

Introduction

The openshift authentication strategy is the preferred and default strategy when Kiali is deployed on an OpenShift cluster.

When using the openshift strategy, a user logging into Kiali will be redirected to the login page of the OpenShift console. Once the user provides his OpenShift credentials, he will be redireted back to Kiali and will be logged in if the user has enough privileges.

The openshift strategy takes advantage of the cluster’s RBAC. See the Role-based access control documentation for more details.

Set-up

Since openshift is the default strategy when deploying Kiali in OpenShift, you shouldn’t need to configure anything. If you want to be verbose, use the following configuration in the Kiali CR:

spec:
  auth:
    strategy: openshift

The openshift strategy doesn’t have any additional configuration. The Kiali operator will make sure to setup the needed OpenShift OAuth resources to register Kiali as a client.

1.5 - Token strategy

Introduction

The token authentication strategy allows a user to login to Kiali using the token of a Kubernetes ServiceAccount. This is similar to the login view of Kubernetes Dashboard.

The token strategy takes advantage of the cluster’s RBAC. See the Role-based access control documentation for more details.

Set-up

Since token is the default strategy when deploying Kiali in Kubernetes, you shouldn’t need to configure anything, unless your cluster is OpenShift. If you want to be verbose or if you need to enable the token strategy in OpenShift, use the following configuration in the Kiali CR:

spec:
  auth:
    strategy: token

The token strategy doesn’t have any additional configuration other than the session expiration time.

1.6 - Session options

There are two settings that are available for the user’s session. The first one is the session expiration time, which is only applicable to token and header authentication strategies:

spec:
  login_token:
    # By default, users session expires in 24 hours.
    expiration_seconds: 86400

The session expiration time is the amount of time before the user is asked to extend his session by another cycle. It does not matter if the user is actively using Kiali, the user will be asked if the session should be extended.

The second available option is the signing key configuration, which is unset by default, meaning that a random 16-character signing key will be generated and stored to a secret named kiali-signing-key, in Kiali’s installation namespace:

spec:
  login_token:
    # By default, create a random signing key and store it in
    # a secret named "kiali-signing-key".
    signing_key: ""

If the secret already exists (which may mean a previous Kiali installation was present), then the secret is reused.

The signing key is used on security sensitive data. For example, one of the usages is to sign HTTP cookies related to the user session to prevent session forgery.

If you need to set a custom fixed key, you can pre-create or modify the kiali-signing-key secret:

apiVersion: v1
kind: Secret
metadata:
  namespace: "kiali-installation-namespace"
  name: kiali-signing-key
type: Opaque
data:
  key: "<your signing key encoded in base64>"

If you prefer a different secret name for the signing key and/or a different key-value pair of the secret, you can specify your preferred names in the Kiali CR:

spec:
  login_token:
    signing_key: "secret:<secretName>:<secretDataKey>"

2 - Console Customization

Custom metric aggregations

The inbound and outbound metric pages, in the Metrics settings drop-down, provides an opinionated set of groupings that work both for filtering out metric data that does not match the selection and for aggregating data into series. Each option is backed by a label on the collected Istio telemetry.

It is possible to add custom aggregations, like in the following example:

spec:
  kiali_feature_flags:
    ui_defaults:
      metrics_inbound:
        aggregations:
        - display_name: Istio Network
          label: topology_istio_io_network
        - display_name: Istio Revision
          label: istio_io_rev
      metrics_outbound:
        aggregations:
        - display_name: Istio Revision
          label: istio_io_rev

Notice that custom aggregations for inbound and outbound metrics are defined separately.

You can find some screenshots in Kiali v1.40 feature update blog post.

Default metrics duration and refresh interval

Most Kiali pages show metrics per refresh and refresh interval drop-downs. These are located at the top-right of the page.

Metrics per refresh specifies the time range back from the current instant to fetch metrics and/or distributed tracing data. By default, a 1-minute time range is selected.

Refresh interval specifies how often Kiali will automatically refresh the data shown. By default, Kiali refreshes data every 15 seconds.

spec:
  kiali_feature_flags:
    ui_defaults:
      # Valid values: 1m, 5m, 10m, 30m, 1h, 3h, 6h, 12h, 1d, 7d, 30d
      metrics_per_refresh: "1m"

      # Valid values: pause, 10s, 15s, 30s, 1m, 5m, 15m
      refresh_interval: "15s"

User selections won’t persist a reload.

Default namespace selection

By default, when Kiali is accessed by the first time, on most Kiali pages users will need to use the namespace drop-down to choose namespaces they want to view data from. The selection will be persisted on reloads.

However, it is possible to configure a predefined selection of namespaces, like in the following example:

spec:
  kiali_feature_flags:
    ui_defaults:
      namespaces:
      - istio-system
      - bookinfo

Namespace selection will reset to the predefined set on reloads. Also, if for some reason a namespace becomes deleted, Kiali will simply ignore it from the list.

Graph find and hide presets

In the toolbar of the topology graph, the Find and Hide textboxes can be configured with presets for your most used criteria. You can find screenshots and a brief description of this feature in the feature update blog post for versions 1.31 to 1.33.

The following are the default presets:

spec:
  kiali_feature_flags:
    ui_defaults:
      graph:
        find_options:
        - description: "Find: slow edges (> 1s)"
          expression: "rt > 1000"
        - description: "Find: unhealthy nodes"
          expression:  "! healthy"
        - description: "Find: unknown nodes"
          expression:  "name = unknown"
        hide_options:
        - description: "Hide: healthy nodes"
          expression: "healthy"
        - description: "Hide: unknown nodes"
          expression:  "name = unknown"

Hopefully, the attributes to configure this feature are self-explanatory.

Note that by providing your own presets, you will be overriding the default configuration. Make sure to include any default presets that you need in case you provide your own configuration.

Graph default traffic rates

Traffic rates in the graph are fetched from Istio telemetry and there are several metric sources that can be used.

In the graph page, you can select the traffic rate metrics using the Traffic drop-down (next to the Namespaces drop-down). By default, Requests is selected for GRPC and HTTP protocols, and Sent bytes is selected for the TCP protocol, but you can change the default selection:

spec:
  kiali_feature_flags:
    ui_defaults:
      graph:
        traffic:
          grpc: "requests" # Valid values: none, requests, sent, received and total
          http: "requests" # Valid values: none and requests
          tcp:  "sent"     # Valid values: none, sent, received and total

Note that only requests provide response codes and will allow health to be calculated. Also, the resulting topology graph may be different for each source.

3 - Custom Dashboards

Custom Dashboards require some configuration to work properly.

Declaring a custom dashboard

When installing Kiali, you define your own custom dashboards in the Kiali CR spec.custom_dashboards field. Here’s an example of what it looks like:

custom_dashboards:
- name: vertx-custom
  title: Vert.x Metrics
  runtime: Vert.x
  discoverOn: "vertx_http_server_connections"
  items:
  - chart:
      name: "Server response time"
      unit: "seconds"
      spans: 6
      metrics:
      - metricName: "vertx_http_server_responseTime_seconds"
        displayName: "Server response time"
      dataType: "histogram"
      aggregations:
      - label: "path"
        displayName: "Path"
      - label: "method"
        displayName: "Method"
  - chart:
      name: "Server active connections"
      unit: ""
      spans: 6
      metricName: "vertx_http_server_connections"
      dataType: "raw"
  - include: "micrometer-1.1-jvm"
  externalLinks:
  - name: "My custom Grafana dashboard"
    type: "grafana"
    variables:
      app: var-app
      namespace: var-namespace
      version: var-version

The name field corresponds to what you can set in the pod annotation kiali.io/dashboards.

The rest of the field definitions are:

• runtime: optional, name of the related runtime. It will be displayed on the corresponding Workload Details page. If omitted no name is displayed.
• title: dashboard title, displayed as a tab in Application or Workloads Details
• discoverOn: metric name to match for auto-discovery. If omitted, the dashboard won’t be discovered automatically, but can still be used via pods annotation.
• items: a list of items, that can be either chart, to define a new chart, or include to reference another dashboard
  • chart: new chart object
    • name: name of the chart
    • chartType: type of the chart, can be one of line (default), area, bar or scatter
    • unit: unit for Y-axis. Free-text field to provide any unit suffix. It can eventually be scaled on display. See specific section below.
    • unitScale: in case the unit needs to be scaled by some factor, set that factor here. For instance, if your data is in milliseconds, set 0.001 as scale and seconds as unit.
    • spans: number of “spans” taken by the chart, from 1 to 12, using bootstrap convention
    • metrics: a list of metrics to display on this single chart:
      • metricName: the metric name in Prometheus
      • displayName: name to display on chart
    • dataType: type of data to be displayed in the chart. Can be one of raw, rate or histogram. Raw data will be queried without transformation. Rate data will be queried using promQL rate() function. And histogram with histogram_quantile() function.
    • min and max: domain for Y-values. When unset, charts implementations should usually automatically adapt the domain with the displayed data.
    • xAxis: type of the X-axis, can be one of time (default) or series. When set to series, only one datapoint per series will be displayed, and the chart type then defaults to bar.
    • aggregator: defines how the time-series are aggregated when several are returned for a given metric and label set. For example, if a Deployment creates a ReplicaSet of several Pods, you will have at least one time-series per Pod. Since Kiali shows the dashboards at the workload (ReplicaSet) level or at the application level, they will have to be aggregated. This field can be used to fix the aggregator, with values such as sum or avg (full list available in Prometheus documentation). However, if omitted the aggregator will default to sum and can be changed from the dashboard UI.
    • aggregations: list of labels eligible for aggregations / groupings (they will be displayed in Kiali through a dropdown list)
      • label: Prometheus label name
      • displayName: name to display in Kiali
      • singleSelection: boolean flag to switch between single-selection and multi-selection modes on the values of this label. Defaults to false.
    • groupLabels: a list of Prometheus labels to be used for grouping. Similar to aggregations, except this grouping will be always turned on.
    • sortLabel: Prometheus label to be used for the metrics display order.
    • sortLabelParseAs: set to int if sortLabel needs to be parsed and compared as an integer instead of string.
  • include: to include another dashboard, or a specific chart from another dashboard. Typically used to compose with generic dashboards such as the ones about MicroProfile Metrics or Micrometer-based JVM metrics. To reference a full dashboard, set the name of that dashboard. To reference a specific chart of another dashboard, set the name of the dashboard followed by $ and the name of the chart (ex: include: "microprofile-1.1$Thread count").
• externalLinks: a list of related external links (e.g. to Grafana dashboards)
  • name: name of the related dashboard in the external system (e.g. name of a Grafana dashboard)
  • type: link type, currently only grafana is allowed
  • variables: a set of variables that can be injected in the URL. For instance, with something like namespace: var-namespace and app: var-app, an URL to a Grafana dashboard that manages namespace and app variables would look like: http://grafana-server:3000/d/xyz/my-grafana-dashboard?var-namespace=some-namespace&var-app=some-app. The available variables in this context are namespace, app and version.

In Kiali, labels for grouping are aggregated in the top toolbar, so if the same label refers to different things depending on the metric, you wouldn’t be able to distinguish them in the UI. For that reason, ideally, labels should not have too generic names in Prometheus. For instance labels named “id” for both memory spaces and buffer pools would better be named “space_id” and “pool_id”. If you have control on label names, it’s an important aspect to take into consideration. Else, it is up to you to organize dashboards with that in mind, eventually splitting them into smaller ones to resolve clashes.

Dashboard scope

The custom dashboards defined in the Kiali CR are available for all workloads in all namespaces.

Additionally, new custom dashboards can be created for a given namespace or workload, using the dashboards.kiali.io/templates annotation.

This is an example where a “Custom Envoy” dashboard will be available for all applications and workloads for the default namespace:

apiVersion: v1
kind: Namespace
metadata:
  name: default
  annotations:
    dashboards.kiali.io/templates: |
      - name: custom_envoy
        title: Custom Envoy
        discoverOn: "envoy_server_uptime"
        items:
          - chart:
              name: "Pods uptime"
              spans: 12
              metricName: "envoy_server_uptime"
              dataType: "raw"

This other example will create an additional “Active Listeners” dashboard only on details-v1 workload:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: details-v1
  labels:
    app: details
    version: v1
spec:
  replicas: 1
  selector:
    matchLabels:
      app: details
      version: v1
  template:
    metadata:
      labels:
        app: details
        version: v1
      annotations:
        dashboards.kiali.io/templates: |
          - name: envoy_listeners
            title: Active Listeners
            discoverOn: "envoy_listener_manager_total_listeners_active"
            items:
              - chart:
                  name: "Total Listeners"
                  spans: 12
                  metricName: "envoy_listener_manager_total_listeners_active"
                  dataType: "raw"
    spec:
      serviceAccountName: bookinfo-details
      containers:
      - name: details
        image: docker.io/istio/examples-bookinfo-details-v1:1.16.2
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 9080
        securityContext:
          runAsUser: 1000

Units

Some units are recognized in Kiali and scaled appropriately when displayed on charts:

• unit: "seconds" can be scaled down to ms, µs, etc.
• unit: "bytes-si" and unit: "bitrate-si" can be scaled up to kB, MB (etc.) using SI / metric system. The aliases unit: "bytes" and unit: "bitrate" can be used instead.
• unit: "bytes-iec" and unit: "bitrate-iec" can be scaled up to KiB, MiB (etc.) using IEC standard / IEEE 1541-2002 (scale by powers of 2).

Other units will fall into the default case and be scaled using SI standard. For instance, unit: "m" for meter can be scaled up to km.

Prometheus Configuration

Kiali custom dashboards work exclusively with Prometheus, so it must be configured correctly to pull your application metrics.

If you are using the demo Istio installation with addons, your Prometheus instance should already be correctly configured and you can skip to the next section; with the exception of Istio 1.6.x where you need customize the ConfigMap, or install Istio with the flag --set meshConfig.enablePrometheusMerge=true.

Using another Prometheus instance

You can use a different instance of Prometheus for these metrics, as opposed to Istio metrics. This second Prometheus instance can be configured from the Kiali CR when using the Kiali operator, or ConfigMap otherwise:

# ...
external_services:
  custom_dashboards:
    prometheus:
      url: URL_TO_PROMETHEUS_SERVER_FOR_CUSTOM_DASHBOARDS
    namespace_label: kubernetes_namespace
  prometheus:
    url: URL_TO_PROMETHEUS_SERVER_FOR_ISTIO_METRICS
# ...

For more details on this configuration, such as Prometheus authentication options, check the Kiali CR Reference page.

You must make sure that this Prometheus instance is correctly configured to scrape your application pods and generates labels that Kiali will understand. Please refer to this documentation to setup the kubernetes_sd_config section. As a reference, here is how it is configured in Istio.

It is important to preserve label mapping, so that Kiali can filter by app and version, and to have the same namespace label as defined per Kiali config. Here’s a relabel_configs that allows this:

      relabel_configs:
      - action: labelmap
        regex: __meta_kubernetes_pod_label_(.+)
      - source_labels: [__meta_kubernetes_namespace]
        action: replace
        target_label: kubernetes_namespace

Pod Annotations and Auto-discovery

Application pods must be annotated for the Prometheus scraper, for example, within a Deployment definition:

spec:
  template:
    metadata:
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/port: "8080"
        prometheus.io/path: "/metrics"

• prometheus.io/scrape tells Prometheus to fetch these metrics or not
• prometheus.io/port is the port under which metrics are exposed
• prometheus.io/path is the endpoint path where metrics are exposed, default is /metrics

Kiali will try to discover automatically dashboards that are relevant for a given Application or Workload. To do so, it reads their metrics and try to match them with the discoverOn field defined on dashboards.

But if you can’t rely on automatic discovery, you can explicitly annotate the pods to associate them with Kiali dashboards.

spec:
  template:
    metadata:
      annotations:
        # (prometheus annotations...)
        kiali.io/dashboards: vertx-server

kiali.io/dashboards is a comma-separated list of dashboard names that Kiali will look for. Each name in the list must match the name of a built-in dashboard or the name of a custom dashboard as defined in the Kial CR’s spec.custom_dashboards.

4 - Istio Environment

Labels and resource names

Istio recommends adding app and version labels to pods to attach this information to telemetry. Kiali relies on correctness of these labels for several features.

In Istio, it is possible to use a different set of labels, like app.kubernetes.io/name and app.kubernetes.io/version, however you must configure Kiali to the labels you are using. By default, Kiali uses Istio’s recommended labels:

spec:
  istio_labels:
    app_label_name: "app"
    version_label_name: "version"

Monitoring port of the IstioD pod

Kiali connects directly to the IstioD pod (not the Service) to check for its health. By default, the connection is done to port 15014 which is the default monitoring port of the IstioD pod.

Under some circumstances, you may need to change the monitoring port of the IstioD pod to something else. For example, when running IstioD in host network mode the network is shared between several pods, requiring to change listening ports of some pods to prevent conflicts.

It is possible to map the newly chosen monitoring port of the IstioD pod in the related Service to let other services continue working normally. However, since Kiali connects directly to the IstioD pod, you need to configure the assigned monitoring port in the Kiali CR:

spec:
  external_services:
    istio:
      istiod_pod_monitoring_port: 15014

Multi-cluster support

Kiali has experimental support for Istio multi-cluster installations using the multi-primary on different networks pattern. This support is enabled by default, but requires the Kiali ServiceAccount to have read access to secrets in the Istio namespace. If you don’t have a multi-cluster setup or don’t want Kiali to have read access to secrets in the Istio namespace, you can disable clustering support:

spec:
  kiali_feature_flags:
    clustering:
      enabled: false

Root namespace

Istio’s root namespace is the namespace where you can create some resources to define default Istio configurations and adapt Istio behavior to your environment. For more information on this Istio configuration, check the Istio docs Global Mesh options page and search for “rootNamespace”.

Kiali uses the root namespace for some of the validations of Istio resources. If you customized the Istio root namespace, you will need to replicate that configuration in Kiali. By default, it is unset:

spec:
  external_services:
    istio:
      root_namespace: ""

Sidecar injection, canary upgrade management and Istio revisions

Kiali can assist into configuring automatic sidecar injection, and also can assist when you are migrating workloads from an old Istio version to a newer one using the canary upgrade method. Kiali uses the standard Istio labels to control sidecar injection policy and canary upgrades.

Management of sidecar injection is enabled by default. If you don’t want this feature, you can disable it with the following configuration:

spec:
  kiali_feature_flags:
    istio_injection_action: false

Assistance for migrating workloads between Istio revisions when doing a canary upgrade is turned off by default. This is because it is required to know what is the revision name that was used when installing each Istio control plane. You can enable and configure the canary upgrade support with the following configuration:

spec:
  external_services:
    istio:
      istio_canary_revision:
        # Revision string of old Istio version
        current: "1-10-3"
        # Revision string of new Istio version
        upgrade: "1-11-0"
  kiali_feature_flags:
    # Turns on canary upgrade support
    istio_upgrade_action: true

It is important to note that canary upgrades require adding a revision name during the installation of control planes. You will notice that the revision name will be appended to the name of Istio resources. Thus, once/if you are using Kiali with an Istio control plane that has a revision name you will need to specify what is the name of a few Istio resources that Kiali uses. For example, if your control plane has a revision name 1-11-0 you would need to set these configurations:

spec:
  external_services:
    istio:
      config_map_name: "istio-1-11-0"
      istio_sidecar_injector_config_map_name: "istio-sidecar-injector-1-11-0"
      istiod_deployment_name: "istiod-1-11-0"

There following are links to sections of Kiali blogs posts that briefly explains these features:

• Sidecar auto-injection control description
• Istio’s canary upgrade assistance description

5 - Kiali CR Reference

apiVersion: kiali.io/v1alpha1
kind: Kiali
metadata:
  name: kiali
  annotations:
    ansible.sdk.operatorframework.io/verbosity: "1"
spec:
  additional_display_details:
  - title: "API Documentation"
    annotation: "kiali.io/api-spec"
    icon_annotation: "kiali.io/api-type"

  installation_tag: ""

  istio_namespace: ""

  version: "default"

  api:
    namespaces:
      exclude:
      - "^istio-operator"
      - "^kube-.*"
      - "^openshift.*"
      - "^ibm.*"
      - "^kiali-operator"
      # default: label_selector is undefined
      label_selector: "kiali.io/member-of=istio-system"

  auth:
    strategy: ""
    openid:
      # default: additional_request_params is empty
      additional_request_params:
        openIdReqParam: "openIdReqParamValue"
      # default: allowed_domains is an empty list
      allowed_domains: ["allowed.domain"]
      api_proxy: ""
      api_proxy_ca_data: ""
      api_token: "id_token"
      authentication_timeout: 300
      authorization_endpoint: ""
      client_id: ""
      disable_rbac: false
      http_proxy: ""
      https_proxy: ""
      insecure_skip_verify_tls: false
      issuer_uri: ""
      scopes: ["openid", "profile", "email"]
      username_claim: "sub"
    openshift:
      client_id_prefix: "kiali"

  # default: custom_dashboards is an empty list
  custom_dashboards:
  - name: "envoy"

  deployment:
    accessible_namespaces: ["^((?!(istio-operator|kube-.*|openshift.*|ibm.*|kiali-operator)).)*$"]
    # default: additional_service_yaml is empty
    additional_service_yaml:
      externalName: "kiali.example.com"
    affinity:
      # default: node is empty
      node:
        requiredDuringSchedulingIgnoredDuringExecution:
          nodeSelectorTerms:
          - matchExpressions:
            - key: kubernetes.io/e2e-az-name
              operator: In
              values:
              - e2e-az1
              - e2e-az2
      # default: pod is empty
      pod:
        requiredDuringSchedulingIgnoredDuringExecution:
        - labelSelector:
            matchExpressions:
            - key: security
              operator: In
              values:
              - S1
          topologyKey: topology.kubernetes.io/zone
      # default: pod_anti is empty
      pod_anti:
        preferredDuringSchedulingIgnoredDuringExecution:
        - weight: 100
          podAffinityTerm:
            labelSelector:
              matchExpressions:
              - key: security
                operator: In
                values:
                - S2
            topologyKey: topology.kubernetes.io/zone
    # default: configmap_annotations is empty
    configmap_annotations:
      strategy.spinnaker.io/versioned: "false"
    # default: custom_secrets is an empty list
    custom_secrets:
    - name: "a-custom-secret"
      mount: "/a-custom-secret-path"
      optional: true
    hpa:
      api_version: ""
      # default: spec is empty
      spec:
        maxReplicas: 2
        minReplicas: 1
        metrics:
        - type: Resource
          resource:
            name: cpu
            target:
              type: Utilization
              averageUtilization: 50
    # default: host_aliases is an empty list
    host_aliases:
    - ip: "192.168.1.100"
      hostnames:
      - "foo.local"
      - "bar.local"
    image_digest: ""
    image_name: ""
    image_pull_policy: "IfNotPresent"
    # default: image_pull_secrets is an empty list
    image_pull_secrets: ["image.pull.secret"]
    image_version: ""
    ingress:
      # default: additional_labels is empty
      additional_labels:
        ingressAdditionalLabel: "ingressAdditionalLabelValue"
      class_name: "nginx"
      # default: enabled is undefined
      enabled: false
      # default: override_yaml is undefined
      override_yaml:
        metadata:
          annotations:
            nginx.ingress.kubernetes.io/secure-backends: "true"
            nginx.ingress.kubernetes.io/backend-protocol: "HTTPS"
        spec:
          rules:
          - http:
              paths:
              - path: "/kiali"
                pathType: Prefix
                backend:
                  service:
                    name: "kiali"
                    port: 
                      number: 20001
    instance_name: "kiali"
    logger:
      log_level: "info"
      log_format: "text"
      sampler_rate: "1"
      time_field_format: "2006-01-02T15:04:05Z07:00"
    namespace: "istio-system"
    # default: node_selector is empty
    node_selector:
      nodeSelector: "nodeSelectorValue"
    # default: pod_annotations is empty
    pod_annotations:
      podAnnotation: "podAnnotationValue"
    # default: pod_labels is empty
    pod_labels:
      sidecar.istio.io/inject: "true"
    priority_class_name: ""
    replicas: 1
    # default: resources is undefined
    resources:
      requests:
        cpu: "10m"
        memory: "64Mi"
      limits:
        memory: "1Gi"
    secret_name: "kiali"
    security_context: {}
    # default: service_annotations is empty
    service_annotations:
      svcAnnotation: "svcAnnotationValue"
    # default: service_type is undefined
    service_type: "NodePort"
    # default: tolerations is an empty list
    tolerations:
    - key: "example-key"
      operator: "Exists"
      effect: "NoSchedule"
    verbose_mode: "3"
    version_label: ""
    view_only_mode: false

  external_services:
    custom_dashboards:
      discovery_auto_threshold: 10
      discovery_enabled: "auto"
      enabled: true
      is_core: false
      namespace_label: "namespace"
      prometheus:
        auth:
          ca_file: ""
          insecure_skip_verify: false
          password: ""
          token: ""
          type: "none"
          use_kiali_token: false
          username: ""
        cache_duration: 10
        cache_enabled: true
        cache_expiration: 300
        # default: custom_headers is empty
        custom_headers:
          customHeader1: "customHeader1Value"
        health_check_url: ""
        is_core: true
        # default: query_scope is empty
        query_scope:
          mesh_id: "mesh-1"
          cluster: "cluster-east"
        thanos_proxy:
          enabled: false
          retention_period: "7d"
          scrape_interval: "30s"
        url: ""
    grafana:
      auth:
        ca_file: ""
        insecure_skip_verify: false
        password: ""
        token: ""
        type: "none"
        use_kiali_token: false
        username: ""
      dashboards:
      - name: "Istio Service Dashboard"
        variables:
          namespace: "var-namespace"
          service: "var-service"
      - name: "Istio Workload Dashboard"
        variables:
          namespace: "var-namespace"
          workload: "var-workload"
      - name: "Istio Mesh Dashboard"
      - name: "Istio Control Plane Dashboard"
      - name: "Istio Performance Dashboard"
      - name: "Istio Wasm Extension Dashboard"
      enabled: true
      health_check_url: ""
      # default: in_cluster_url is undefined
      in_cluster_url: ""
      is_core: false
      url: ""
    istio:
      component_status:
        components:
        - app_label: "istiod"
          is_core: true
          is_proxy: false
        - app_label: "istio-ingressgateway"
          is_core: true
          is_proxy: true
          # default: namespace is undefined
          namespace: istio-system
        - app_label: "istio-egressgateway"
          is_core: false
          is_proxy: true
          # default: namespace is undefined
          namespace: istio-system
        enabled: true
      config_map_name: "istio"
      envoy_admin_local_port: 15000
      # default: istio_canary_revision is undefined
      istio_canary_revision:
        current: "1-9-9"
        upgrade: "1-10-2"
      istio_identity_domain: "svc.cluster.local"
      istio_injection_annotation: "sidecar.istio.io/inject"
      istio_sidecar_annotation: "sidecar.istio.io/status"
      istio_sidecar_injector_config_map_name: "istio-sidecar-injector"
      istiod_deployment_name: "istiod"
      istiod_pod_monitoring_port: 15014
      root_namespace: ""
      url_service_version: ""
    prometheus:
      auth:
        ca_file: ""
        insecure_skip_verify: false
        password: ""
        token: ""
        type: "none"
        use_kiali_token: false
        username: ""
      cache_duration: 10
      cache_enabled: true
      cache_expiration: 300
      # default: custom_headers is empty
      custom_headers:
        customHeader1: "customHeader1Value"
      health_check_url: ""
      is_core: true
      # default: query_scope is empty
      query_scope:
        mesh_id: "mesh-1"
        cluster: "cluster-east"
      thanos_proxy:
        enabled: false
        retention_period: "7d"
        scrape_interval: "30s"
      url: ""
    tracing:
      auth:
        ca_file: ""
        insecure_skip_verify: false
        password: ""
        token: ""
        type: "none"
        use_kiali_token: false
        username: ""
      enabled: true
      in_cluster_url: ""
      is_core: false
      namespace_selector: true
      # default: query_scope is empty
      query_scope:
        mesh_id: "mesh-1"
        cluster: "cluster-east"
      url: ""
      use_grpc: true
      whitelist_istio_system: ["jaeger-query", "istio-ingressgateway"]

  health_config:
    # default: rate is an empty list
    rate:
    - namespace: ".*"
      kind: ".*"
      name: ".*"
      tolerance:
      - protocol: "http"
        direction: ".*"
        code: "[1234]00"
        degraded: 5
        failure: 10

  identity:
    # default: cert_file is undefined
    cert_file: ""
    # default: private_key_file is undefined
    private_key_file: ""

  istio_labels:
    app_label_name: "app"
    injection_label_name: "istio-injection"
    injection_label_rev:  "istio.io/rev"
    version_label_name: "version"

  kiali_feature_flags:
    certificates_information_indicators:
      enabled: true
      secrets:
      - "cacerts"
      - "istio-ca-secret"
    clustering:
      enabled: true
    disabled_features: []
    istio_injection_action: true
    istio_upgrade_action: false
    ui_defaults:
      graph:
        find_options:
        - description: "Find: slow edges (> 1s)"
          expression: "rt > 1000"
        - description: "Find: unhealthy nodes"
          expression:  "! healthy"
        - description: "Find: unknown nodes"
          expression:  "name = unknown"
        hide_options:
        - description: "Hide: healthy nodes"
          expression: "healthy"
        - description: "Hide: unknown nodes"
          expression:  "name = unknown"
        traffic:
          grpc: "requests"
          http: "requests"
          tcp:  "sent"
      metrics_per_refresh: "1m"
      # default: metrics_inbound is undefined
      metrics_inbound:
        aggregations:
        - display_name: "Istio Network"
          label: "topology_istio_io_network"
        - display_name: "Istio Revision"
          label: "istio_io_rev"
      # default: metrics_outbound is undefined
      metrics_outbound:
        aggregations:
        - display_name: "Istio Network"
          label: "topology_istio_io_network"
        - display_name: "Istio Revision"
          label: "istio_io_rev"
      # default: namespaces is an empty list
      namespaces: ["istio-system"]
      refresh_interval: "60s"
    validations:
      ignore: ["KIA1201"]

  kubernetes_config:
    burst: 200
    cache_duration: 300
    cache_enabled: true
    cache_istio_types:
    - "AuthorizationPolicy"
    - "DestinationRule"
    - "EnvoyFilter"
    - "Gateway"
    - "PeerAuthentication"
    - "RequestAuthentication"
    - "ServiceEntry"
    - "Sidecar"
    - "VirtualService"
    - "WorkloadEntry"
    - "WorkloadGroup"
    cache_namespaces:
    - ".*"
    cache_token_namespace_duration: 10
    excluded_workloads:
    - "CronJob"
    - "DeploymentConfig"
    - "Job"
    - "ReplicationController"
    qps: 175

  login_token:
    expiration_seconds: 86400
    signing_key: ""

  server:
    address: ""
    audit_log: true
    cors_allow_all: false
    gzip_enabled: true
    observability:
      metrics:
        enabled: true
        port: 9090
      tracing:
        collector_url: "http://jaeger-collector.istio-system:14268/api/traces"
        enabled: false
    port: 20001
    web_fqdn: ""
    web_history_mode: ""
    web_port: ""
    web_root: ""
    web_schema: ""

bash <(curl -sL https://raw.githubusercontent.com/kiali/kiali-operator/master/crd-docs/bin/validate-kiali-cr.sh) \
  -crd https://raw.githubusercontent.com/kiali/kiali-operator/master/crd-docs/crd/kiali.io_kialis.yaml \
  --kiali-cr-name kiali \
  -n istio-system

annotations:
  kiali.io/api-spec: http://list/to/my/api/doc
  kiali.io/api-type: rest

- name: myapp
  title: My App Metrics
  items:
  - chart:
      name: "Thread Count"
      spans: 4
      metricName: "thread-count"
      dataType: "raw"

- name: envoy

custom_secrets:
- name: mysecret
  mount: /mysecret-path
- name: my-other-secret
  mount: /my-other-secret-location
  optional: true

host_aliases:
- ip: 192.168.1.100
  hostnames:
  - "foo.local"
  - "bar.local"

hpa:
  api_version: "autoscaling/v2"
  spec:
    maxReplicas: 2
    minReplicas: 1
    metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 50

override_yaml:
  metadata:
    annotations:
      nginx.ingress.kubernetes.io/secure-backends: "true"
      nginx.ingress.kubernetes.io/backend-protocol: "HTTPS"
  spec:
    rules:
    - http:
        paths:
        - path: /kiali
          pathType: Prefix
          backend:
            service
              name: "kiali"
              port:
                number: 20001

sidecar.istio.io/inject: "true"

requests:
  cpu: "10m"
  memory: "64Mi"
limits:
  memory: "1Gi"

metrics_inbound:
  aggregations:
  - display_name: Istio Network
    label: topology_istio_io_network
  - display_name: Istio Revision
    label: istio_io_rev

metrics_outbound:
  aggregations:
  - display_name: Istio Network
    label: topology_istio_io_network
  - display_name: Istio Revision
    label: istio_io_rev

6 - Namespace Management

Introduction

The default Kiali installation (as mentioned in the Installation guide) gives Kiali access to all namespaces available in the cluster.

It is possible to restrict Kiali to a set of desired namespaces by providing a list of the ones you want, excluding the ones you don’t want, or filtering by a label selector. You can use a combination of these options.

Accessible Namespaces

You can configure which namespaces are accessible and observable through Kiali. You can use regex expressions which will be matched against the operator’s visible namespaces. If not set in the Kiali CR, the default makes accessible all cluster namespaces, with the exception of a predefined set of the cluster’s system workloads.

The list of accessible namespaces is specified in the Kiali CR via the accessible_namespaces setting, under the main deployment section. As an example, if Kiali is to be installed in the istio-system namespace, and is expected to monitor all namespaces prefixed with mycorp_, the setting would be:

spec:
  deployment:
    accessible_namespaces:
    - istio-system
    - mycorp_.*

Note that the namespaces declared here (including any regex expressions) are evaluated and discovered at install time. Namespaces that do not exist at the time of install but are created later in the future will not be accessible by Kiali. For Kiali to be given access to namespaces created in the future, you must edit the Kiali CR and update the accessible_namespaces setting to include the new namespaces. The one exception is if you set accessible_namespaces to the special value ["**"] - in that case, all namespaces (including any namespaces created in the future) will be accessible to Kiali.

This configuration accepts the special pattern accessible_namespaces: ["**"] which denotes that Kiali is given access to all namespaces in the cluster, including any namespaces created in the future.

When installing multiple Kiali instances into a single cluster, accessible_namespaces must be mutually exclusive. In other words, a namespace set must be matched by only one Kiali CR. Regular expressions must not have overlapping patterns.

Maistra supports multi-tenancy and the accessible_namespaces extends that feature to Kiali. However, explicit naming of accessible namespaces can benefit non-Maistra installations as well - with it Kiali does not need cluster roles and the Kiali Operator does not need permissions to create cluster roles.

Excluded Namespaces

The Kiali CR tells the Kiali Operator which accessible namespaces should be excluded from the list of namespaces provided by the API and UI. This can be useful if wildcards are used when specifying Accessible Namespaces. This setting has no effect on namespace accessibility. It is only a filter, not security-related.

For example, if the accessible_namespaces configuration includes mycorp_.* but it is not desirable to see test namespaces, the following configuration can be used:

api:
  namespaces:
    exclude:
      - mycorp_test.*

Namespace Selectors

To fetch a subset of the available namespaces, Kiali supports an optional Kubernetes label selector. This selector is especially useful when spec.deployment.accessible_namespaces is set to ["**"] but you want to reduce the namespaces presented in the UI’s namespace list.

The label selector is defined in the Kiali CR setting spec.api.namespaces.label_selector.

The example below selects all namespaces that have a label kiali-enabled: true:

api:
  namespaces:
    label_selector: kiali-enabled=true

For further information on how this api.namespaces.label_selector interacts with spec.deployment.accessible_namespaces read the Kiali CR Reference documentation.

To label a namespace you can use the following command. For more information see the Kubernete’s official documentation.

  kubectl label namespace my-namespace kiali-enabled=true

Note that when deploying multiple control planes in the same cluster, you will want to set the label selector’s value unique to each control plane. This allows each Kiali instance to select only the namespaces relevant to each control plane. Because in this “soft-multitenancy” mode spec.deployment.accessible_namespaces is typically set to an explicit set of namespaces (i.e. not ["**"]), you do not have to do anything with this label_selector. This is because the default value of label_selector is kiali.io/member-of: <spec.istio_namespace> when spec.deployment.accessible_namespaces is not set to the “all namespaces” value ["**"]. This allows you to have multiple control planes in the same cluster, with each control plane having its own Kiali instance. If you set your own Kiali instance name in the Kiali CR (i.e. you set spec.deployment.instance_name to something other than kiali), then the default label will be kiali.io/<spec.deployment.instance_name>.member-of: <spec.istio_namespace>.

7 - Prometheus, Jaeger, Grafana

Prometheus is a required telemetry data source for Kiali. Jaeger is a highly recommended tracing data source. Kiali also offers a simple Grafana add-on integration. This page describes how to configure Kiali to communicate with these dependencies.

Read the dedicated configuration page to learn more.

7.1 - Grafana

Grafana configuration

Istio provides preconfigured Grafana dashboards for the most relevant metrics of the mesh. Although Kiali offers similar views in its metrics dashboards, it is not in Kiali’s goals to provide the advanced querying options, nor the highly customizable settings, that are available in Grafana. Thus, it is recommended that you use Grafana if you need those advanced options.

Kiali can provide a direct link from its metric dashboards to the equivalent or most similar Grafana dashboard, which is convenient if you need the powerful Grafana options. For these links to appear in Kiali you need to manually configure the Grafana URL, like in the following example:

spec:
  external_services:
    grafana:
      enabled: true
      # Grafana service name is "grafana" and is in the "telemetry" namespace.
      in_cluster_url: 'http://grafana.telemetry:3000/'
      # Public facing URL of Grafana
      url: 'http://my-ingress-host/grafana'

7.2 - Jaeger

Jaeger configuration

Jaeger is a highly recommended service because Kiali uses distributed tracing data for several features, providing an enhanced experience.

By default, Kiali will try to reach Jaeger at the GRPC-enabled URL of the form http://tracing.<istio_namespace_name>:16685/jaeger, which is the usual case if you are using the Jaeger Istio add-on. If this endpoint is unreachable, Kiali will disable features that use distributed tracing data.

If your Jaeger instance has a different service name or is installed to a different namespace, you must manually provide the endpoint where it is available, like in the following example:

spec:
  external_services:
    tracing:
      # Enabled by default. Kiali will anyway fallback to disabled if
      # Jaeger is unreachable.
      enabled: true
      # Jaeger service name is "tracing" and is in the "telemetry" namespace.
      # Make sure the URL you provide corresponds to the non-GRPC enabled endpoint
      # if you set "use_grpc" to false.
      in_cluster_url: 'http://tracing.telemetry:16685/jaeger'
      use_grpc: true
      # Public facing URL of Jaeger
      url: 'http://my-jaeger-host/jaeger'

Minimally, you must provide spec.external_services.tracing.in_cluster_url to enable Kiali features that use distributed tracing data. However, Kiali can provide contextual links that users can use to jump to the Jaeger console to inspect tracing data more in depth. For these links to be available you need to set the spec.external_services.tracing.url which may mean that you should expose Jaeger outside the cluster.

7.3 - Prometheus

Prometheus configuration

Kiali requires Prometheus to generate the topology graph, show metrics, calculate health and for several other features. If Prometheus is missing or Kiali can’t reach it, Kiali won’t work properly.

By default, Kiali assumes that Prometheus is available at the URL of the form http://prometheus.<istio_namespace_name>:9090, which is the usual case if you are using the Prometheus Istio add-on. If your Prometheus instance has a different service name or is installed in a different namespace, you must manually provide the endpoint where it is available, like in the following example:

spec:
  external_services:
    prometheus:
      # Prometheus service name is "metrics" and is in the "telemetry" namespace
      url: "http://metrics.telemetry:9090/"

Kiali maintains an internal cache of some Prometheus queries to improve performance (mainly, the queries to calculate Health indicators). It would be very rare to see data delays, but should you notice any delays you may tune caching parameters to values that work better for your environment.

See the [Kiali CR reference page]("/docs/configuration/kialis.kiali.io/#example-cr" >}}) for the current default values.

Compatibility with Prometheus-like servers

Although Kiali assumes a Prometheus server and is tested against it, there are TSDBs that can be used as a Prometheus replacement despite not implementing the full Prometheus API.

Community users have faced two issues when using Prometheus-like TSDBs:

• Kiali may report that the TSDB is unreachable, and/or
• Kiali may show empty metrics if the TSBD does not implement the /api/v1/status/config.

To fix these issues, you may need to provide a custom health check endpoint for the TSDB and/or manually provide the configurations that Kiali reads from the /api/v1/status/config API endpoint:

spec:
  external_services:
    prometheus:
      # Fix the "Unreachable" metrics server warning.
      health_check_url: "http://custom-tsdb-health-check-url"
      # Fix for the empty metrics dashboards
      thanos_proxy:
        enabled: true
        retention_period: "7d"
        scrape_interval: "30s"

Prometheus Tuning

Production environments should not be using the Istio Prometheus add-on, or carrying over its configuration settings. That is useful only for small, or demo installations. Instead, Prometheus should have been installed in a production-oriented way, following the Prometheus documentation.

This section is primarily for users where Prometheus is being used specifically for Kiali, and possible optimizations that can be made knowing that Kiali does not utilize all of the default Istio and Envoy telemetry.

Metric Thinning

Istio and Envoy generate a large amount of telemetry for analysis and troubleshooting. This can result in significant resources being required to ingest and store the telemetry, and to support queries into the data. If you use the telemetry specifically to support Kiali, it is possible to drop unnecessary metrics and unnecessary labels on required metrics. This FAQ Entry displays the metrics and attributes required for Kiali to operate.

To reduce the default telemetry to only what is needed by Kiali¹ users can add the following snippet to their Prometheus configuration. Because things can change with different versions, it is recommended to ensure you use the correct version of this documentation based on your Kiali/Istio version.

The metric_relabel_configs: attribute should be added under each job name defined to scrape Istio or Envoy metrics. Below we show it under the kubernetes-pods job, but you should adapt as needed. Be careful of indentation.

    - job_name: kubernetes-pods
      metric_relabel_configs:
      - action: drop
        source_labels: [__name__]
        regex: istio_agent_.*|istiod_.*|istio_build|citadel_.*|galley_.*|pilot_[^p].*|envoy_cluster_[^u].*|envoy_cluster_update.*|envoy_listener_[^dh].*|envoy_server_[^mu].*|envoy_wasm_.*
      - action: labeldrop
        regex: chart|destination_app|destination_version|heritage|.*operator.*|istio.*|release|security_istio_io_.*|service_istio_io_.*|sidecar_istio_io_inject|source_app|source_version

Applying this configuration should reduce the number of stored metrics by about 20%, as well as reducing the number of attributes stored on many remaining metrics.

Metric Thinning with Crippling

The section above drops metrics unused by Kiali. As such, making those configuration changes should not negatively impact Kiali behavior in any way. But some very heavy metrics remain. These metrics can also be dropped, but their removal will impact the behavior of Kiali. This may be OK if you don’t use the affected features of Kiali, or if you are willing to sacrifice the feature for the associated metric savings. In particular, these are “Histogram” metrics. Istio is planning to make some improvements to help users better configure these metrics, but as of this writing they are still defined with fairly inefficient default “buckets”, making the number of associated time-series quite large, and the overhead of maintaining and querying the metrics, intensive. Each histogram actually is comprised of 3 stored metrics. For example, a histogram named xxx would result in the following metrics stored into Prometheus:

• xxx_bucket
  • The most intensive metric, and is required to calculate percentile values.
• xxx_count
  • Required to calculate ‘avg’ values.
• xxx_sum
  • Required to calculate rates over time, and for ‘avg’ values.

When considering whether to thin the Histogram metrics, one of the following three approaches is recommended:

1. If the relevant Kiali reporting is needed, keep the histogram as-is.
2. If the relevant Kiali reporting is not needed, or not worth the additional metric overhead, drop the entire histogram.
3. If the metric chart percentiles are not required, drop only the xxx_bucket metric. This removes the majority of the histogram overhead while keeping rate and average (non-percentile) values in Kiali.

These are the relevant Histogram metrics:

istio_request_bytes

This metric is used to produce the Request Size chart on the metric tabs. It also supports Request Throughput edge labels on the graph.

• Appending |istio_request_bytes_.* to the drop regex above would drop all associated metrics and would prevent any request size/throughput reporting in Kiali.
• Appending |istio_request_bytes_bucket to the drop regex above, would prevent any request size percentile reporting in the Kiali metric charts.

istio_response_bytes

This metric is used to produce the Response Size chart on the metric tabs. And also supports Response Throughput edge labels on the graph

• Appending |istio_response_bytes_.* to the drop regex above would drop all associated metrics and would prevent any response size/throughput reporting in Kiali.
• Appending |istio_response_bytes_bucket to the drop regex above would prevent any response size percentile reporting in the Kiali metric charts.

istio_request_duration_milliseconds

This metric is used to produce the Request Duration chart on the metric tabs. It also supports Response Time edge labels on the graph.

• Appending |istio_request_duration_milliseconds_.* to the drop regex above would drop all associated metrics and would prevent any request duration/response time reporting in Kiali.
• Appending |istio_request_duration_milliseconds_bucket to the drop regex above would prevent any request duration/response time percentile reporting in the Kiali metric charts or graph edge labels.

Scrape Interval

The Prometheus globalScrapeInterval is an important configuration option². The scrape interval can have a significant effect on metrics collection overhead as it takes effort to pull all of those configured metrics and update the relevant time-series. And although it doesn’t affect time-series cardinality, it does affect storage for the data-points, as well as having impact when computing query results (the more data-points, the more processing and aggregation).

Users should think carefully about their configured scrape interval. Note that the Istio addon for prometheus configures it to 15s. This is great for demos but may be too frequent for production scenarios. The prometheus helm charts set a default of 1m, which is more reasonable for most installations, but may not be the desired frequency for any particular setup.

The recommendation for Kiali is to set the longest interval possible, while still providing a useful granularity. The longer the interval the less data points scraped, thus reducing processing, storage, and computational overhead. But the impact on Kiali should be understood. It is important to realize that request rates (or byte rates, message rates, etc) require a minumum of two data points:

rate = (dp2 - dp1) / timePeriod

That means for Kiali to show anything useful in the graph, or anywhere rates are used (many places), the minimum duration must be >= 2 x globalScrapeInterval. Kiali will eliminate invalid Duration options given the globalScrapeInterval.

Kiali does a lot of aggregation and querying over time periods. As such, the number of data points will affect query performance, especially for larger time periods.

For more information, see the Prometheus documentation.

TSDB retention time

The Prometheus tsdbRetentionTime is an important configuration option. It has a significant effect on metrics storage, as Prometheus will keep each reported data-point for that period of time, performing compaction as needed. The larger the retention time, the larger the required storage. Note also that Kiali queries against large time periods, and very large data-sets, may result in poor performance or timeouts.

The recommendation for Kiali is to set the shortest retention time that meets your needs and/or operational limits. In some cases users may want to offload older data to a secondary store. Kiali will eliminate invalid Duration options given the tsdbRetentionTime.

For more information, see the Prometheus documentation.

8 - Role-based access control

Introduction

Kiali supports role-based access control (RBAC) when using any authentication strategy other than anonymous.

Although the anonymous strategy does not support RBAC, you can still limit privileges if using an OpenShift cluster. See the access control section in Anonymous strategy.

Kiali uses the RBAC capabilities of the underlying cluster. Thus, RBAC is accomplished by using the standard RBAC features of the cluster, which is through ClusterRoles, ClusterRoleBindings, Roles and RoleBindings resources. Read the Kubernetes RBAC documentation for details. If you are using OpenShift, read the OpenShift RBAC documentation.

In general, Kiali will give access to the resources granted to the account used to login. Specifically, depending on the authentication strategy, this translates to: