Containers

• 1: Images
• 2: Container Environment
• 3: Runtime Class
• 4: Container Lifecycle Hooks

1 - Images

A container image represents binary data that encapsulates an application and all its software dependencies. Container images are executable software bundles that can run standalone and that make very well-defined assumptions about their runtime environment.

You typically create a container image of your application and push it to a registry before referring to it in a Pod.

This page provides an outline of the container image concept.

Image names

Container images are usually given a name such as pause, example/mycontainer, or kube-apiserver. Images can also include a registry hostname; for example: fictional.registry.example/imagename, and possibly a port number as well; for example: fictional.registry.example:10443/imagename.

If you don't specify a registry hostname, Kubernetes assumes that you mean the Docker public registry. You can change this behavior by setting a default image registry in the container runtime configuration.

After the image name part you can add a tag or digest (in the same way you would when using with commands like docker or podman). Tags let you identify different versions of the same series of images. Digests are a unique identifier for a specific version of an image. Digests are hashes of the image's content, and are immutable. Tags can be moved to point to different images, but digests are fixed.

Image tags consist of lowercase and uppercase letters, digits, underscores (_), periods (.), and dashes (-). A tag can be up to 128 characters long, and must conform to the following regex pattern: [a-zA-Z0-9_][a-zA-Z0-9._-]{0,127}. You can read more about it and find the validation regex in the OCI Distribution Specification. If you don't specify a tag, Kubernetes assumes you mean the tag latest.

Image digests consists of a hash algorithm (such as sha256) and a hash value. For example: sha256:1ff6c18fbef2045af6b9c16bf034cc421a29027b800e4f9b68ae9b1cb3e9ae07. You can find more information about the digest format in the OCI Image Specification.

Some image name examples that Kubernetes can use are:

• busybox — Image name only, no tag or digest. Kubernetes will use the Docker public registry and latest tag. Equivalent to docker.io/library/busybox:latest.
• busybox:1.32.0 — Image name with tag. Kubernetes will use the Docker public registry. Equivalent to docker.io/library/busybox:1.32.0.
• registry.k8s.io/pause:latest — Image name with a custom registry and latest tag.
• registry.k8s.io/pause:3.5 — Image name with a custom registry and non-latest tag.
• registry.k8s.io/pause@sha256:1ff6c18fbef2045af6b9c16bf034cc421a29027b800e4f9b68ae9b1cb3e9ae07 — Image name with digest.
• registry.k8s.io/pause:3.5@sha256:1ff6c18fbef2045af6b9c16bf034cc421a29027b800e4f9b68ae9b1cb3e9ae07 — Image name with tag and digest. Only the digest will be used for pulling.

Updating images

When you first create a Deployment, StatefulSet, Pod, or other object that includes a PodTemplate, and a pull policy was not explicitly specified, then by default the pull policy of all containers in that Pod will be set to IfNotPresent. This policy causes the kubelet to skip pulling an image if it already exists.

Image pull policy

The imagePullPolicy for a container and the tag of the image both affect when the kubelet attempts to pull (download) the specified image.

Here's a list of the values you can set for imagePullPolicy and the effects these values have:

IfNotPresent

the image is pulled only if it is not already present locally.

Always

every time the kubelet launches a container, the kubelet queries the container image registry to resolve the name to an image digest. If the kubelet has a container image with that exact digest cached locally, the kubelet uses its cached image; otherwise, the kubelet pulls the image with the resolved digest, and uses that image to launch the container.

Never

the kubelet does not try fetching the image. If the image is somehow already present locally, the kubelet attempts to start the container; otherwise, startup fails. See pre-pulled images for more details.

The caching semantics of the underlying image provider make even imagePullPolicy: Always efficient, as long as the registry is reliably accessible. Your container runtime can notice that the image layers already exist on the node so that they don't need to be downloaded again.

To make sure the Pod always uses the same version of a container image, you can specify the image's digest; replace <image-name>:<tag> with <image-name>@<digest> (for example, image@sha256:45b23dee08af5e43a7fea6c4cf9c25ccf269ee113168c19722f87876677c5cb2).

When using image tags, if the image registry were to change the code that the tag on that image represents, you might end up with a mix of Pods running the old and new code. An image digest uniquely identifies a specific version of the image, so Kubernetes runs the same code every time it starts a container with that image name and digest specified. Specifying an image by digest pins the code that you run so that a change at the registry cannot lead to that mix of versions.

There are third-party admission controllers that mutate Pods (and PodTemplates) when they are created, so that the running workload is defined based on an image digest rather than a tag. That might be useful if you want to make sure that your entire workload is running the same code no matter what tag changes happen at the registry.

Default image pull policy

When you (or a controller) submit a new Pod to the API server, your cluster sets the imagePullPolicy field when specific conditions are met:

• if you omit the imagePullPolicy field, and you specify the digest for the container image, the imagePullPolicy is automatically set to IfNotPresent.
• if you omit the imagePullPolicy field, and the tag for the container image is :latest, imagePullPolicy is automatically set to Always.
• if you omit the imagePullPolicy field, and you don't specify the tag for the container image, imagePullPolicy is automatically set to Always.
• if you omit the imagePullPolicy field, and you specify a tag for the container image that isn't :latest, the imagePullPolicy is automatically set to IfNotPresent.

Required image pull

If you would like to always force a pull, you can do one of the following:

• Set the imagePullPolicy of the container to Always.
• Omit the imagePullPolicy and use :latest as the tag for the image to use; Kubernetes will set the policy to Always when you submit the Pod.
• Omit the imagePullPolicy and the tag for the image to use; Kubernetes will set the policy to Always when you submit the Pod.
• Enable the AlwaysPullImages admission controller.

ImagePullBackOff

When a kubelet starts creating containers for a Pod using a container runtime, it might be possible the container is in Waiting state because of ImagePullBackOff.

The status ImagePullBackOff means that a container could not start because Kubernetes could not pull a container image (for reasons such as invalid image name, or pulling from a private registry without imagePullSecret). The BackOff part indicates that Kubernetes will keep trying to pull the image, with an increasing back-off delay.

Kubernetes raises the delay between each attempt until it reaches a compiled-in limit, which is 300 seconds (5 minutes).

Image pull per runtime class

If you enable the RuntimeClassInImageCriApi feature gate, the kubelet references container images by a tuple of image name and runtime handler rather than just the image name or digest. Your container runtime may adapt its behavior based on the selected runtime handler. Pulling images based on runtime class is useful for VM-based containers, such as Windows Hyper-V containers.

Serial and parallel image pulls

By default, the kubelet pulls images serially. In other words, the kubelet sends only one image pull request to the image service at a time. Other image pull requests have to wait until the one being processed is complete.

Nodes make image pull decisions in isolation. Even when you use serialized image pulls, two different nodes can pull the same image in parallel.

If you would like to enable parallel image pulls, you can set the field serializeImagePulls to false in the kubelet configuration. With serializeImagePulls set to false, image pull requests will be sent to the image service immediately, and multiple images will be pulled at the same time.

When enabling parallel image pulls, ensure that the image service of your container runtime can handle parallel image pulls.

The kubelet never pulls multiple images in parallel on behalf of one Pod. For example, if you have a Pod that has an init container and an application container, the image pulls for the two containers will not be parallelized. However, if you have two Pods that use different images, and the parallel image pull feature is enabled, the kubelet will pull the images in parallel on behalf of the two different Pods.

Maximum parallel image pulls

When serializeImagePulls is set to false, the kubelet defaults to no limit on the maximum number of images being pulled at the same time. If you would like to limit the number of parallel image pulls, you can set the field maxParallelImagePulls in the kubelet configuration. With maxParallelImagePulls set to n, only n images can be pulled at the same time, and any image pull beyond n will have to wait until at least one ongoing image pull is complete.

Limiting the number of parallel image pulls prevents image pulling from consuming too much network bandwidth or disk I/O, when parallel image pulling is enabled.

You can set maxParallelImagePulls to a positive number that is greater than or equal to 1. If you set maxParallelImagePulls to be greater than or equal to 2, you must set serializeImagePulls to false. The kubelet will fail to start with an invalid maxParallelImagePulls setting.

Multi-architecture images with image indexes

As well as providing binary images, a container registry can also serve a container image index. An image index can point to multiple image manifests for architecture-specific versions of a container. The idea is that you can have a name for an image (for example: pause, example/mycontainer, kube-apiserver) and allow different systems to fetch the right binary image for the machine architecture they are using.

The Kubernetes project typically creates container images for its releases with names that include the suffix -$(ARCH). For backward compatibility, generate older images with suffixes. For instance, an image named as pause would be a multi-architecture image containing manifests for all supported architectures, while pause-amd64 would be a backward-compatible version for older configurations, or for YAML files with hardcoded image names containing suffixes.

Using a private registry

Private registries may require authentication to be able to discover and/or pull images from them. Credentials can be provided in several ways:

• Specifying imagePullSecrets when you define a Pod

  Only Pods which provide their own keys can access the private registry.

• Configuring Nodes to Authenticate to a Private Registry

  • All Pods can read any configured private registries.
  • Requires node configuration by cluster administrator.

• Using a kubelet credential provider plugin to dynamically fetch credentials for private registries

  The kubelet can be configured to use credential provider exec plugin for the respective private registry.

• Pre-pulled Images

  • All Pods can use any images cached on a node.
  • Requires root access to all nodes to set up.

• Vendor-specific or local extensions

  If you're using a custom node configuration, you (or your cloud provider) can implement your mechanism for authenticating the node to the container registry.

These options are explained in more detail below.

Specifying imagePullSecrets on a Pod

Kubernetes supports specifying container image registry keys on a Pod. All imagePullSecrets must be Secrets that exist in the same Namespace as the Pod. These Secrets must be of type kubernetes.io/dockercfg or kubernetes.io/dockerconfigjson.

Configuring nodes to authenticate to a private registry

Specific instructions for setting credentials depends on the container runtime and registry you chose to use. You should refer to your solution's documentation for the most accurate information.

For an example of configuring a private container image registry, see the Pull an Image from a Private Registry task. That example uses a private registry in Docker Hub.

Kubelet credential provider for authenticated image pulls

You can configure the kubelet to invoke a plugin binary to dynamically fetch registry credentials for a container image. This is the most robust and versatile way to fetch credentials for private registries, but also requires kubelet-level configuration to enable.

This technique can be especially useful for running static Pods that require container images hosted in a private registry. Using a ServiceAccount or a Secret to provide private registry credentials is not possible in the specification of a static Pod, because it cannot have references to other API resources in its specification.

See Configure a kubelet image credential provider for more details.

Interpretation of config.json

The interpretation of config.json varies between the original Docker implementation and the Kubernetes interpretation. In Docker, the auths keys can only specify root URLs, whereas Kubernetes allows glob URLs as well as prefix-matched paths. The only limitation is that glob patterns (*) have to include the dot (.) for each subdomain. The amount of matched subdomains has to be equal to the amount of glob patterns (*.), for example:

• *.kubernetes.io will not match kubernetes.io, but will match abc.kubernetes.io.
• *.*.kubernetes.io will not match abc.kubernetes.io, but will match abc.def.kubernetes.io.
• prefix.*.io will match prefix.kubernetes.io.
• *-good.kubernetes.io will match prefix-good.kubernetes.io.

This means that a config.json like this is valid:

{     "auths": {         "my-registry.example/images": { "auth": "…" },         "*.my-registry.example/images": { "auth": "…" }     } } 

Image pull operations pass the credentials to the CRI container runtime for every valid pattern. For example, the following container image names would match successfully:

• my-registry.example/images
• my-registry.example/images/my-image
• my-registry.example/images/another-image
• sub.my-registry.example/images/my-image

However, these container image names would not match:

• a.sub.my-registry.example/images/my-image
• a.b.sub.my-registry.example/images/my-image

The kubelet performs image pulls sequentially for every found credential. This means that multiple entries in config.json for different paths are possible, too:

{     "auths": {         "my-registry.example/images": {             "auth": "…"         },         "my-registry.example/images/subpath": {             "auth": "…"         }     } } 

If now a container specifies an image my-registry.example/images/subpath/my-image to be pulled, then the kubelet will try to download it using both authentication sources if one of them fails.

Pre-pulled images

By default, the kubelet tries to pull each image from the specified registry. However, if the imagePullPolicy property of the container is set to IfNotPresent or Never, then a local image is used (preferentially or exclusively, respectively).

If you want to rely on pre-pulled images as a substitute for registry authentication, you must ensure all nodes in the cluster have the same pre-pulled images.

This can be used to preload certain images for speed or as an alternative to authenticating to a private registry.

Similar to the usage of the kubelet credential provider, pre-pulled images are also suitable for launching static Pods that depend on images hosted in a private registry.

Ensure image pull credential verification

If the KubeletEnsureSecretPulledImages feature gate is enabled for your cluster, Kubernetes will validate image credentials for every image that requires credentials to be pulled, even if that image is already present on the node. This validation ensures that images in a Pod request which have not been successfully pulled with the provided credentials must re-pull the images from the registry. Additionally, image pulls that re-use the same credentials which previously resulted in a successful image pull will not need to re-pull from the registry and are instead validated locally without accessing the registry (provided the image is available locally). This is controlled by theimagePullCredentialsVerificationPolicy field in the Kubelet configuration.

This configuration controls when image pull credentials must be verified if the image is already present on the node:

• NeverVerify: Mimics the behavior of having this feature gate disabled. If the image is present locally, image pull credentials are not verified.
• NeverVerifyPreloadedImages: Images pulled outside the kubelet are not verified, but all other images will have their credentials verified. This is the default behavior.
• NeverVerifyAllowListedImages: Images pulled outside the kubelet and mentioned within the preloadedImagesVerificationAllowlist specified in the kubelet config are not verified.
• AlwaysVerify: All images will have their credentials verified before they can be used.

This verification applies to pre-pulled images, images pulled using node-wide secrets, and images pulled using Pod-level secrets.

Creating a Secret with a Docker config

You need to know the username, registry password and client email address for authenticating to the registry, as well as its hostname. Run the following command, substituting placeholders with the appropriate values:

kubectl create secret docker-registry <name> \   --docker-server=<docker-registry-server> \   --docker-username=<docker-user> \   --docker-password=<docker-password> \   --docker-email=<docker-email> 

If you already have a Docker credentials file then, rather than using the above command, you can import the credentials file as a Kubernetes Secret. Create a Secret based on existing Docker credentials explains how to set this up.

This is particularly useful if you are using multiple private container registries, as kubectl create secret docker-registry creates a Secret that only works with a single private registry.

Referring to imagePullSecrets on a Pod

Now, you can create pods which reference that secret by adding the imagePullSecrets section to a Pod definition. Each item in the imagePullSecrets array can only reference one Secret in the same namespace.

For example:

cat <<EOF > pod.yaml apiVersion: v1 kind: Pod metadata:   name: foo   namespace: awesomeapps spec:   containers:     - name: foo       image: janedoe/awesomeapp:v1   imagePullSecrets:     - name: myregistrykey EOF  cat <<EOF >> ./kustomization.yaml resources: - pod.yaml EOF 

This needs to be done for each Pod that is using a private registry.

However, you can automate this process by specifying the imagePullSecrets section in a ServiceAccount resource. See Add ImagePullSecrets to a Service Account for detailed instructions.

You can use this in conjunction with a per-node .docker/config.json. The credentials will be merged.

Use cases

There are a number of solutions for configuring private registries. Here are some common use cases and suggested solutions.

1. Cluster running only non-proprietary (e.g. open-source) images. No need to hide images.
   • Use public images from a public registry
     • No configuration required.
     • Some cloud providers automatically cache or mirror public images, which improves availability and reduces the time to pull images.
2. Cluster running some proprietary images which should be hidden to those outside the company, but visible to all cluster users.
   • Use a hosted private registry
     • Manual configuration may be required on the nodes that need to access to private registry.
   • Or, run an internal private registry behind your firewall with open read access.
     • No Kubernetes configuration is required.
   • Use a hosted container image registry service that controls image access
     • It will work better with Node autoscaling than manual node configuration.
   • Or, on a cluster where changing the node configuration is inconvenient, use imagePullSecrets.
3. Cluster with proprietary images, a few of which require stricter access control.
   • Ensure AlwaysPullImages admission controller is active. Otherwise, all Pods potentially have access to all images.
   • Move sensitive data into a Secret resource, instead of packaging it in an image.
4. A multi-tenant cluster where each tenant needs own private registry.
   • Ensure AlwaysPullImages admission controller is active. Otherwise, all Pods of all tenants potentially have access to all images.
   • Run a private registry with authorization required.
   • Generate registry credentials for each tenant, store into a Secret, and propagate the Secret to every tenant namespace.
   • The tenant then adds that Secret to imagePullSecrets of each namespace.

If you need access to multiple registries, you can create one Secret per registry.

Legacy built-in kubelet credential provider

In older versions of Kubernetes, the kubelet had a direct integration with cloud provider credentials. This provided the ability to dynamically fetch credentials for image registries.

There were three built-in implementations of the kubelet credential provider integration: ACR (Azure Container Registry), ECR (Elastic Container Registry), and GCR (Google Container Registry).

Starting with version 1.26 of Kubernetes, the legacy mechanism has been removed, so you would need to either:

• configure a kubelet image credential provider on each node; or
• specify image pull credentials using imagePullSecrets and at least one Secret.

What's next

• Read the OCI Image Manifest Specification.
• Learn about container image garbage collection.
• Learn more about pulling an Image from a Private Registry.

2 - Container Environment

This page describes the resources available to Containers in the Container environment.

Container environment

The Kubernetes Container environment provides several important resources to Containers:

• A filesystem, which is a combination of an image and one or more volumes.
• Information about the Container itself.
• Information about other objects in the cluster.

Container information

The hostname of a Container is the name of the Pod in which the Container is running. It is available through the hostname command or the gethostname function call in libc.

The Pod name and namespace are available as environment variables through the downward API.

User defined environment variables from the Pod definition are also available to the Container, as are any environment variables specified statically in the container image.

Cluster information

A list of all services that were running when a Container was created is available to that Container as environment variables. This list is limited to services within the same namespace as the new Container's Pod and Kubernetes control plane services.

For a service named foo that maps to a Container named bar, the following variables are defined:

FOO_SERVICE_HOST=<the host the service is running on> FOO_SERVICE_PORT=<the port the service is running on> 

Services have dedicated IP addresses and are available to the Container via DNS, if DNS addon is enabled. 

What's next

• Learn more about Container lifecycle hooks.
• Get hands-on experience attaching handlers to Container lifecycle events.

3 - Runtime Class

This page describes the RuntimeClass resource and runtime selection mechanism.

RuntimeClass is a feature for selecting the container runtime configuration. The container runtime configuration is used to run a Pod's containers.

Motivation

You can set a different RuntimeClass between different Pods to provide a balance of performance versus security. For example, if part of your workload deserves a high level of information security assurance, you might choose to schedule those Pods so that they run in a container runtime that uses hardware virtualization. You'd then benefit from the extra isolation of the alternative runtime, at the expense of some additional overhead.

You can also use RuntimeClass to run different Pods with the same container runtime but with different settings.

Setup

1. Configure the CRI implementation on nodes (runtime dependent)
2. Create the corresponding RuntimeClass resources

1. Configure the CRI implementation on nodes

The configurations available through RuntimeClass are Container Runtime Interface (CRI) implementation dependent. See the corresponding documentation (below) for your CRI implementation for how to configure.

The configurations have a corresponding handler name, referenced by the RuntimeClass. The handler must be a valid DNS label name.

2. Create the corresponding RuntimeClass resources

The configurations setup in step 1 should each have an associated handler name, which identifies the configuration. For each handler, create a corresponding RuntimeClass object.

The RuntimeClass resource currently only has 2 significant fields: the RuntimeClass name (metadata.name) and the handler (handler). The object definition looks like this:

# RuntimeClass is defined in the node.k8s.io API group apiVersion: node.k8s.io/v1 kind: RuntimeClass metadata:   # The name the RuntimeClass will be referenced by.   # RuntimeClass is a non-namespaced resource.   name: myclass  # The name of the corresponding CRI configuration handler: myconfiguration  

The name of a RuntimeClass object must be a valid DNS subdomain name.

Usage

Once RuntimeClasses are configured for the cluster, you can specify a runtimeClassName in the Pod spec to use it. For example:

apiVersion: v1 kind: Pod metadata:   name: mypod spec:   runtimeClassName: myclass   # ... 

This will instruct the kubelet to use the named RuntimeClass to run this pod. If the named RuntimeClass does not exist, or the CRI cannot run the corresponding handler, the pod will enter the Failed terminal phase. Look for a corresponding event for an error message.

If no runtimeClassName is specified, the default RuntimeHandler will be used, which is equivalent to the behavior when the RuntimeClass feature is disabled.

CRI Configuration

For more details on setting up CRI runtimes, see CRI installation.

containerd

Runtime handlers are configured through containerd's configuration at /etc/containerd/config.toml. Valid handlers are configured under the runtimes section:

[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.${HANDLER_NAME}] 

See containerd's config documentation for more details:

CRI-O

Runtime handlers are configured through CRI-O's configuration at /etc/crio/crio.conf. Valid handlers are configured under the crio.runtime table:

[crio.runtime.runtimes.${HANDLER_NAME}]   runtime_path = "${PATH_TO_BINARY}" 

See CRI-O's config documentation for more details.

Scheduling

By specifying the scheduling field for a RuntimeClass, you can set constraints to ensure that Pods running with this RuntimeClass are scheduled to nodes that support it. If scheduling is not set, this RuntimeClass is assumed to be supported by all nodes.

To ensure pods land on nodes supporting a specific RuntimeClass, that set of nodes should have a common label which is then selected by the runtimeclass.scheduling.nodeSelector field. The RuntimeClass's nodeSelector is merged with the pod's nodeSelector in admission, effectively taking the intersection of the set of nodes selected by each. If there is a conflict, the pod will be rejected.

If the supported nodes are tainted to prevent other RuntimeClass pods from running on the node, you can add tolerations to the RuntimeClass. As with the nodeSelector, the tolerations are merged with the pod's tolerations in admission, effectively taking the union of the set of nodes tolerated by each.

To learn more about configuring the node selector and tolerations, see Assigning Pods to Nodes.

Pod Overhead

You can specify overhead resources that are associated with running a Pod. Declaring overhead allows the cluster (including the scheduler) to account for it when making decisions about Pods and resources.

Pod overhead is defined in RuntimeClass through the overhead field. Through the use of this field, you can specify the overhead of running pods utilizing this RuntimeClass and ensure these overheads are accounted for in Kubernetes.

What's next

• RuntimeClass Design
• RuntimeClass Scheduling Design
• Read about the Pod Overhead concept
• PodOverhead Feature Design

4 - Container Lifecycle Hooks

This page describes how kubelet managed Containers can use the Container lifecycle hook framework to run code triggered by events during their management lifecycle.

Overview

Analogous to many programming language frameworks that have component lifecycle hooks, such as Angular, Kubernetes provides Containers with lifecycle hooks. The hooks enable Containers to be aware of events in their management lifecycle and run code implemented in a handler when the corresponding lifecycle hook is executed.

Container hooks

There are two hooks that are exposed to Containers:

PostStart

This hook is executed immediately after a container is created. However, there is no guarantee that the hook will execute before the container ENTRYPOINT. No parameters are passed to the handler.

PreStop

This hook is called immediately before a container is terminated due to an API request or management event such as a liveness/startup probe failure, preemption, resource contention and others. A call to the PreStop hook fails if the container is already in a terminated or completed state and the hook must complete before the TERM signal to stop the container can be sent. The Pod's termination grace period countdown begins before the PreStop hook is executed, so regardless of the outcome of the handler, the container will eventually terminate within the Pod's termination grace period. No parameters are passed to the handler.

A more detailed description of the termination behavior can be found in Termination of Pods.

StopSignal

The StopSignal lifecycle can be used to define a stop signal which would be sent to the container when it is stopped. If you set this, it overrides any STOPSIGNAL instruction defined within the container image.

A more detailed description of termination behaviour with custom stop signals can be found in Stop Signals.

Hook handler implementations

Containers can access a hook by implementing and registering a handler for that hook. There are three types of hook handlers that can be implemented for Containers:

• Exec - Executes a specific command, such as pre-stop.sh, inside the cgroups and namespaces of the Container. Resources consumed by the command are counted against the Container.
• HTTP - Executes an HTTP request against a specific endpoint on the Container.
• Sleep - Pauses the container for a specified duration. This is a beta-level feature default enabled by the PodLifecycleSleepAction feature gate.

Hook handler execution

When a Container lifecycle management hook is called, the Kubernetes management system executes the handler according to the hook action, httpGet, tcpSocket (deprecated) and sleep are executed by the kubelet process, and exec is executed in the container.

The PostStart hook handler call is initiated when a container is created, meaning the container ENTRYPOINT and the PostStart hook are triggered simultaneously. However, if the PostStart hook takes too long to execute or if it hangs, it can prevent the container from transitioning to a running state.

PreStop hooks are not executed asynchronously from the signal to stop the Container; the hook must complete its execution before the TERM signal can be sent. If a PreStop hook hangs during execution, the Pod's phase will be Terminating and remain there until the Pod is killed after its terminationGracePeriodSeconds expires. This grace period applies to the total time it takes for both the PreStop hook to execute and for the Container to stop normally. If, for example, terminationGracePeriodSeconds is 60, and the hook takes 55 seconds to complete, and the Container takes 10 seconds to stop normally after receiving the signal, then the Container will be killed before it can stop normally, since terminationGracePeriodSeconds is less than the total time (55+10) it takes for these two things to happen.

If either a PostStart or PreStop hook fails, it kills the Container.

Users should make their hook handlers as lightweight as possible. There are cases, however, when long running commands make sense, such as when saving state prior to stopping a Container.

Hook delivery guarantees

Hook delivery is intended to be at least once, which means that a hook may be called multiple times for any given event, such as for PostStart or PreStop. It is up to the hook implementation to handle this correctly.

Generally, only single deliveries are made. If, for example, an HTTP hook receiver is down and is unable to take traffic, there is no attempt to resend. In some rare cases, however, double delivery may occur. For instance, if a kubelet restarts in the middle of sending a hook, the hook might be resent after the kubelet comes back up.

Debugging Hook handlers

The logs for a Hook handler are not exposed in Pod events. If a handler fails for some reason, it broadcasts an event. For PostStart, this is the FailedPostStartHook event, and for PreStop, this is the FailedPreStopHook event. To generate a failed FailedPostStartHook event yourself, modify the lifecycle-events.yaml file to change the postStart command to "badcommand" and apply it. Here is some example output of the resulting events you see from running kubectl describe pod lifecycle-demo:

Events:   Type     Reason               Age              From               Message   ----     ------               ----             ----               -------   Normal   Scheduled            7s               default-scheduler  Successfully assigned default/lifecycle-demo to ip-XXX-XXX-XX-XX.us-east-2...   Normal   Pulled               6s               kubelet            Successfully pulled image "nginx" in 229.604315ms   Normal   Pulling              4s (x2 over 6s)  kubelet            Pulling image "nginx"   Normal   Created              4s (x2 over 5s)  kubelet            Created container lifecycle-demo-container   Normal   Started              4s (x2 over 5s)  kubelet            Started container lifecycle-demo-container   Warning  FailedPostStartHook  4s (x2 over 5s)  kubelet            Exec lifecycle hook ([badcommand]) for Container "lifecycle-demo-container" in Pod "lifecycle-demo_default(30229739-9651-4e5a-9a32-a8f1688862db)" failed - error: command 'badcommand' exited with 126: , message: "OCI runtime exec failed: exec failed: container_linux.go:380: starting container process caused: exec: \"badcommand\": executable file not found in $PATH: unknown\r\n"   Normal   Killing              4s (x2 over 5s)  kubelet            FailedPostStartHook   Normal   Pulled               4s               kubelet            Successfully pulled image "nginx" in 215.66395ms   Warning  BackOff              2s (x2 over 3s)  kubelet            Back-off restarting failed container 

What's next

• Learn more about the Container environment.
• Get hands-on experience attaching handlers to Container lifecycle events.