This is not the latest version of Linkerd!
This documentation is for an older version of Linkerd. You may want the Linkerd 2.12 (current) documentation instead.

Upgrading Linkerd

In this guide, we’ll walk you through how to upgrade Linkerd.

Before starting, read through the version-specific upgrade notices below, which may contain important information you need to be aware of before commencing with the upgrade process:

There are three components that need to be upgraded:

Upgrade the CLI

This will upgrade your local CLI to the latest version. You will want to follow these instructions for anywhere that uses the Linkerd CLI. For Helm users feel free to skip to the Helm section.

To upgrade the CLI locally, run:

curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh

Alternatively, you can download the CLI directly via the Linkerd releases page.

Verify the CLI is installed and running correctly with:

linkerd version --client

Which should display:

Client version: L5D2_STABLE_VERSION

You are now ready to upgrade your control plane.

Upgrade the Control Plane

Now that you have upgraded the CLI, it is time to upgrade the Linkerd control plane on your Kubernetes cluster. Don’t worry, the existing data plane will continue to operate with a newer version of the control plane and your meshed services will not go down.

With Linkerd CLI

Use the linkerd upgrade command to upgrade the control plane. This command ensures that all of the control plane’s existing configuration and mTLS secrets are retained. Notice that we use the --prune flag to remove any Linkerd resources from the previous version which no longer exist in the new version.

linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

Next, run this command again with some --prune-whitelist flags added. This is necessary to make sure that certain cluster-scoped resources are correctly pruned.

linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd \
  --prune-whitelist=rbac.authorization.k8s.io/v1/clusterrole \
  --prune-whitelist=rbac.authorization.k8s.io/v1/clusterrolebinding \
  --prune-whitelist=apiregistration.k8s.io/v1/apiservice -f -

For upgrading a multi-stage installation setup, follow the instructions at Upgrading a multi-stage install.

Users who have previously saved the Linkerd control plane’s configuration to files can follow the instructions at Upgrading via manifests to ensure those configuration are retained by the linkerd upgrade command.

With Helm

For a Helm workflow, check out the instructions at Helm upgrade procedure.

Verify the control plane upgrade

Once the upgrade process completes, check to make sure everything is healthy by running:

linkerd check

This will run through a set of checks against your control plane and make sure that it is operating correctly.

To verify the Linkerd control plane version, run:

linkerd version

Which should display:

Client version: L5D2_STABLE_VERSION
Server version: L5D2_STABLE_VERSION

Next, we will upgrade your data plane.

Upgrade the Data Plane

With a fully up-to-date CLI running locally and Linkerd control plane running on your Kubernetes cluster, it is time to upgrade the data plane. The easiest way to do this is to run a rolling deploy on your services, allowing the proxy-injector to inject the latest version of the proxy as they come up.

With kubectl 1.15+, this can be as simple as using the kubectl rollout restart command to restart all your meshed services. For example,

kubectl -n <namespace> rollout restart deploy

Workloads that were previously injected using the linkerd inject --manual command can be upgraded by re-injecting the applications in-place. For example,

kubectl -n emojivoto get deploy -l linkerd.io/control-plane-ns=linkerd -oyaml \
  | linkerd inject --manual - \
  | kubectl apply -f -

Verify the data plane upgrade

Check to make sure everything is healthy by running:

linkerd check --proxy

This will run through a set of checks to verify that the data plane is operating correctly, and will list any pods that are still running older versions of the proxy.

Congratulation! You have successfully upgraded your Linkerd to the newer version. If you have any questions, feel free to raise them at the #linkerd2 channel in the Linkerd slack.

Upgrade notice: stable-2.9.3

Linkerd Repair

Due to a known issue in versions stable-2.9.0, stable-2.9.1, and stable-2.9.2, users who upgraded to one of those versions with the –prune flag (as described above) will have deleted the secret/linkerd-config-overrides resource which is necessary for performing any subsequent upgrades. Linkerd stable-2.9.3 includes a new linkerd repair command which restores this deleted resource. If you see unexpected error messages during upgrade such as “failed to read CA: not PEM-encoded”, please upgrade your CLI to stable-2.9.3 and run:

linkerd repair | kubectl apply -f -

This will restore the secret/linkerd-config-overrides resource and allow you to proceed with upgrading your control plane.

Upgrade notice: stable-2.9.0

Images are now hosted on ghcr.io

As of this version images are now hosted under ghcr.io instead of gcr.io. If you’re pulling images into a private repo please make the necessary changes.

Upgrading multicluster environments

Linkerd 2.9 changes the way that some of the multicluster components work and are installed compared to Linkerd 2.8.x. Users installing the multicluster components for the first time with Linkerd 2.9 can ignore these instructions and instead refer directly to the installing multicluster instructions.

Users who installed the multicluster component in Linkerd 2.8.x and wish to upgrade to Linkerd 2.9 should follow the upgrade multicluster instructions.

Ingress behavior changes

In previous versions when you injected your ingress controller (Nginx, Traefik, Ambassador, etc), then the ingress’ balancing/routing choices would be overridden with Linkerd’s (using service profiles, traffic splits, etc.).

As of 2.9 the ingress’s choices are honored instead, which allows preserving things like session-stickiness. Note however that this means per-route metrics are not collected, traffic splits will not be honored and retries/timeouts are not applied.

If you want to revert to the previous behavior, inject the proxy into the ingress controller using the annotation linkerd.io/inject: ingress, as explained in using ingress

Breaking changes in Helm charts

Some entries like controllerLogLevel and all the Prometheus config have changed their position in the settings hierarchy. To get a precise view of what has changed you can compare the stable-2.8.1 and stable-2.9.0 values.yaml files.

Post-upgrade cleanup

In order to better support cert-manager, the secrets linkerd-proxy-injector-tls, linkerd-sp-validator-tls and linkerd-tap-tls have been replaced by the secrets linkerd-proxy-injector-k8s-tls, linkerd-sp-validator-k8s-tls and linkerd-tap-k8s-tls respectively. If you upgraded through the CLI, please delete the old ones (if you upgraded through Helm the cleanup was automated).

Upgrade notice: stable-2.8.0

There are no version-specific notes for upgrading to this release. The upgrade process detailed above (upgrade the CLI, upgrade the control plane, then upgrade the data plane) should work.

Upgrade notice: stable-2.7.0

Checking whether any of your TLS certificates are approaching expiry

This version introduces a set of CLI flags and checks that help you rotate your TLS certificates. The new CLI checks will warn you if any of your certificates are expiring in the next 60 days. If you however want to check the expiration date of your certificates and determine for yourself whether you should be rotating them, you can execute the following commands. Note that this will require step 0.13.3 and jq 1.6.

Check your trust roots:

kubectl -n linkerd get cm linkerd-config -o=jsonpath="{.data.global}" |  \
jq -r .identityContext.trustAnchorsPem | \
step certificate inspect --short -

X.509v3 Root CA Certificate (ECDSA P-256) [Serial: 1]
  Subject:     identity.linkerd.cluster.local
  Issuer:      identity.linkerd.cluster.local
  Valid from:  2020-01-14T13:23:32Z
          to:  2021-01-13T13:23:52Z

Check your issuer certificate:

kubectl -n linkerd get secret linkerd-identity-issuer -o=jsonpath="{.data['crt\.pem']}" |  \
base64 --decode | \
step certificate inspect --short -

X.509v3 Root CA Certificate (ECDSA P-256) [Serial: 1]
  Subject:     identity.linkerd.cluster.local
  Issuer:      identity.linkerd.cluster.local
  Valid from:  2020-01-14T13:23:32Z
          to:  2021-01-13T13:23:52Z

If you determine that you wish to rotate your certificates you can follow the process outlined in Rotating your identity certificates. Note that this process uses functionality available in stable-2.7.0. So before you start your cert rotation, make sure to upgrade.

When ready, you can begin the upgrade process by installing the new CLI.

Breaking changes in Helm charts

As part of an effort to follow Helm’s best practices the Linkerd Helm chart has been restructured. As a result most of the keys have been changed. In order to ensure trouble-free upgrade of your Helm installation, please take a look at Helm upgrade procedure. To get a precise view of what has changed you can compare that stable-2.6.0 and stable-2.7.0 values.yaml files.

Upgrade notice: stable-2.6.0

The destination container is now deployed as its own Deployment workload. When you are planning the upgrade from one of the edge versions listed above, be sure to allocate time to restart the data plane once the control plane is successfully upgraded. This restart can be done at your convenience with the recommendation that it be done over the course of time appropriate for your application.

If you are upgrading from a previous stable version, restarting the data-plane is recommended as a best practice, although not necessary.

If you have previously labelled any of your namespaces with the linkerd.io/is-control-plane label so that their pod creation events are ignored by the HA proxy injector, you will need to update these namespaces to use the new config.linkerd.io/admission-webhooks: disabled label.

When ready, you can begin the upgrade process by installing the new CLI.

Upgrade notice: stable-2.5.0

This release supports Kubernetes 1.12+.

Upgrading from stable-2.4.x

Use the linkerd upgrade command to upgrade the control plane. This command ensures that all of the control plane’s existing configuration and mTLS secrets are retained.

# get the latest stable CLI
curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh
linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

The options --prune -l linkerd.io/control-plane-ns=linkerd above make sure that any resources that are removed from the linkerd upgrade output, are effectively removed from the system.

For upgrading a multi-stage installation setup, follow the instructions at Upgrading a multi-stage install.

Users who have previously saved the Linkerd control plane’s configuration to files can follow the instructions at Upgrading via manifests to ensure those configuration are retained by the linkerd upgrade command.

Once the upgrade command completes, use the linkerd check command to confirm the control plane is ready.

When ready, proceed to upgrading the data plane by following the instructions at Upgrade the data plane.

Upgrade notice: stable-2.4.0

This release supports Kubernetes 1.12+.

Upgrading from stable-2.3.x, edge-19.4.5, edge-19.5.x, edge-19.6.x, edge-19.7.x

Use the linkerd upgrade command to upgrade the control plane. This command ensures that all of the control plane’s existing configuration and mTLS secrets are retained.

# get the latest stable CLI
curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh

For Kubernetes 1.12+:

linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

For Kubernetes pre-1.12 where the mutating and validating webhook configurations’ sideEffects fields aren’t supported:

linkerd upgrade --omit-webhook-side-effects | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

The sideEffects field is added to the Linkerd webhook configurations to indicate that the webhooks have no side effects on other resources.

For HA setup, the linkerd upgrade command will also retain all previous HA configuration. Note that the mutating and validating webhook configurations are updated to set their failurePolicy fields to fail to ensure that un-injected workloads (as a result of unexpected errors) are rejected during the admission process. The HA mode has also been updated to schedule multiple replicas of the linkerd-proxy-injector and linkerd-sp-validator deployments.

For users upgrading from the edge-19.5.3 release, note that the upgrade process will fail with the following error message, due to a naming bug:

The ClusterRoleBinding "linkerd-linkerd-tap" is invalid: roleRef: Invalid value:
rbac.RoleRef{APIGroup:"rbac.authorization.k8s.io", Kind:"ClusterRole",
Name:"linkerd-linkerd-tap"}: cannot change roleRef

This can be resolved by simply deleting the linkerd-linkerd-tap cluster role binding resource, and re-running the linkerd upgrade command:

kubectl delete clusterrole/linkerd-linkerd-tap

For upgrading a multi-stage installation setup, follow the instructions at Upgrading a multi-stage install.

Users who have previously saved the Linkerd control plane’s configuration to files can follow the instructions at Upgrading via manifests to ensure those configuration are retained by the linkerd upgrade command.

Once the upgrade command completes, use the linkerd check command to confirm the control plane is ready.

When ready, proceed to upgrading the data plane by following the instructions at Upgrade the data plane.

Upgrading from stable-2.2.x

Follow the stable-2.3.0 upgrade instructions to upgrade the control plane to the stable-2.3.2 release first. Then follow these instructions to upgrade the stable-2.3.2 control plane to stable-2.4.0.

Upgrade notice: stable-2.3.0

stable-2.3.0 introduces a new upgrade command. This command only works for the edge-19.4.x and newer releases. When using the upgrade command from edge-19.2.x or edge-19.3.x, all the installation flags previously provided to the install command must also be added.

Upgrading from stable-2.2.x

To upgrade from the stable-2.2.x release, follow the Step-by-step instructions.

Note that if you had previously installed Linkerd with --tls=optional, delete the linkerd-ca deployment after successful Linkerd control plane upgrade:

kubectl -n linkerd delete deploy/linkerd-ca

Upgrading from edge-19.4.x

# get the latest stable
curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh

# upgrade the control plane
linkerd upgrade | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

Follow instructions for upgrading the data plane.

Upgrading a multi-stage install

edge-19.4.5 introduced a Multi-stage install feature. If you previously installed Linkerd via a multi-stage install process, you can upgrade each stage, analogous to the original multi-stage installation process.

Stage 1, for the cluster owner:

linkerd upgrade config | kubectl apply -f -

Stage 2, for the service owner:

linkerd upgrade control-plane | kubectl apply -f -

Upgrading via manifests

By default, the linkerd upgrade command reuses the existing linkerd-config config map and the linkerd-identity-issuer secret, by fetching them via the the Kubernetes API. edge-19.4.5 introduced a new --from-manifests flag to allow the upgrade command to read the linkerd-config config map and the linkerd-identity-issuer secret from a static YAML file. This option is relevant to CI/CD workflows where the Linkerd configuration is managed by a configuration repository.

For release after edge-20.10.1/stable-2.9.0, you need to add secret/linkerd-config-overrides to the linkerd-manifest.yaml by running command:

kubectl -n linkerd get \
  secret/linkerd-identity-issuer \
  configmap/linkerd-config \
  secret/linkerd-config-overrides \
  -oyaml > linkerd-manifests.yaml

linkerd upgrade --from-manifests linkerd-manifests.yaml | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

For release after stable-2.6.0 and prior to edge-20.10.1/stable-2.9.0, you can use this command:

kubectl -n linkerd get \
  secret/linkerd-identity-issuer \
  configmap/linkerd-config \
  -oyaml > linkerd-manifests.yaml

linkerd upgrade --from-manifests linkerd-manifests.yaml | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

For releases prior to edge-19.8.1/stable-2.5.0, and after stable-2.6.0, you may pipe a full linkerd install manifest into the upgrade command:

linkerd install > linkerd-install.yaml

# deploy Linkerd
cat linkerd-install.yaml | kubectl apply -f -

# upgrade Linkerd via manifests
cat linkerd-install.yaml | linkerd upgrade --from-manifests -

Upgrading from edge-19.2.x or edge-19.3.x

# get the latest stable
curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh

# Install stable control plane, using flags previously supplied during
# installation.
# For example, if the previous installation was:
# linkerd install --proxy-log-level=warn --proxy-auto-inject | kubectl apply -f -
# The upgrade command would be:
linkerd upgrade --proxy-log-level=warn --proxy-auto-inject | kubectl apply --prune -l linkerd.io/control-plane-ns=linkerd -f -

Follow instructions for upgrading the data plane.

Upgrade notice: stable-2.2.0

There are two breaking changes in stable-2.2.0. One relates to Service Profiles, the other relates to Automatic Proxy Injection. If you are not using either of these features, you may skip directly to the full upgrade instructions.

Service Profile namespace location

Service Profiles, previously defined in the control plane namespace in stable-2.1.0, are now defined in their respective client and server namespaces. Service Profiles defined in the client namespace take priority over ones defined in the server namespace.

Automatic Proxy Injection opt-in

The linkerd.io/inject annotation, previously opt-out in stable-2.1.0, is now opt-in.

To enable automation proxy injection for a namespace, you must enable the linkerd.io/inject annotation on either the namespace or the pod spec. For more details, see the Automatic Proxy Injection doc.

A note about application updates

Also note that auto-injection only works during resource creation, not update. To update the data plane proxies of a deployment that was auto-injected, do one of the following:

  • Manually re-inject the application via linkerd inject (more info below under Upgrade the data plane)
  • Delete and redeploy the application

Auto-inject support for application updates is tracked on github

Step-by-step instructions (stable-2.2.x)

Upgrade the 2.2.x CLI

This will upgrade your local CLI to the latest version. You will want to follow these instructions for anywhere that uses the linkerd CLI.

To upgrade the CLI locally, run:

curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh

Alternatively, you can download the CLI directly via the Linkerd releases page.

Verify the CLI is installed and running correctly with:

linkerd version

Which should display:

Client version: L5D2_STABLE_VERSION
Server version: stable-2.1.0

It is expected that the Client and Server versions won’t match at this point in the process. Nothing has been changed on the cluster, only the local CLI has been updated.

Upgrade the 2.2.x control plane

Now that you have upgraded the CLI running locally, it is time to upgrade the Linkerd control plane on your Kubernetes cluster. Don’t worry, the existing data plane will continue to operate with a newer version of the control plane and your meshed services will not go down.

To upgrade the control plane in your environment, run the following command. This will cause a rolling deploy of the control plane components that have changed.

linkerd install | kubectl apply -f -

The output will be:

namespace/linkerd configured
configmap/linkerd-config created
serviceaccount/linkerd-identity created
clusterrole.rbac.authorization.k8s.io/linkerd-linkerd-identity configured
clusterrolebinding.rbac.authorization.k8s.io/linkerd-linkerd-identity configured
service/linkerd-identity created
secret/linkerd-identity-issuer created
deployment.extensions/linkerd-identity created
serviceaccount/linkerd-controller unchanged
clusterrole.rbac.authorization.k8s.io/linkerd-linkerd-controller configured
clusterrolebinding.rbac.authorization.k8s.io/linkerd-linkerd-controller configured
service/linkerd-controller-api configured
service/linkerd-destination created
deployment.extensions/linkerd-controller configured
customresourcedefinition.apiextensions.k8s.io/serviceprofiles.linkerd.io configured
serviceaccount/linkerd-web unchanged
service/linkerd-web configured
deployment.extensions/linkerd-web configured
serviceaccount/linkerd-prometheus unchanged
clusterrole.rbac.authorization.k8s.io/linkerd-linkerd-prometheus configured
clusterrolebinding.rbac.authorization.k8s.io/linkerd-linkerd-prometheus configured
service/linkerd-prometheus configured
deployment.extensions/linkerd-prometheus configured
configmap/linkerd-prometheus-config configured
serviceaccount/linkerd-grafana unchanged
service/linkerd-grafana configured
deployment.extensions/linkerd-grafana configured
configmap/linkerd-grafana-config configured
serviceaccount/linkerd-sp-validator created
clusterrole.rbac.authorization.k8s.io/linkerd-linkerd-sp-validator configured
clusterrolebinding.rbac.authorization.k8s.io/linkerd-linkerd-sp-validator configured
service/linkerd-sp-validator created
deployment.extensions/linkerd-sp-validator created

Check to make sure everything is healthy by running:

linkerd check

This will run through a set of checks against your control plane and make sure that it is operating correctly.

To verify the Linkerd control plane version, run:

linkerd version

Which should display:

Client version: L5D2_STABLE_VERSION
Server version: L5D2_STABLE_VERSION

Upgrade the 2.2.x data plane

With a fully up-to-date CLI running locally and Linkerd control plane running on your Kubernetes cluster, it is time to upgrade the data plane. This will change the version of the linkerd-proxy sidecar container and run a rolling deploy on your service.

For stable-2.3.0+, if your workloads are annotated with the auto-inject linkerd.io/inject: enabled annotation, then you can just restart your pods using your Kubernetes cluster management tools (helm, kubectl etc.).

As the pods are being re-created, the proxy injector will auto-inject the new version of the proxy into the pods.

If auto-injection is not part of your workflow, you can still manually upgrade your meshed services by re-injecting your applications in-place.

Begin by retrieving your YAML resources via kubectl, and pass them through the linkerd inject command. This will update the pod spec with the linkerd.io/inject: enabled annotation. This annotation will be picked up by Linkerd’s proxy injector during the admission phase where the Linkerd proxy will be injected into the workload. By using kubectl apply, Kubernetes will do a rolling deploy of your service and update the running pods to the latest version.

Example command to upgrade an application in the emojivoto namespace, composed of deployments:

kubectl -n emojivoto get deploy -l linkerd.io/control-plane-ns=linkerd -oyaml \
  | linkerd inject - \
  | kubectl apply -f -

Check to make sure everything is healthy by running:

linkerd check --proxy

This will run through a set of checks against both your control plane and data plane to verify that it is operating correctly.

You can make sure that you’ve fully upgraded all the data plane by running:

kubectl get po --all-namespaces -o yaml \
  | grep linkerd.io/proxy-version

The output will look something like:

linkerd.io/proxy-version: L5D2_STABLE_VERSION
linkerd.io/proxy-version: L5D2_STABLE_VERSION

If there are any older versions listed, you will want to upgrade them as well.