{"id":1309,"date":"2021-01-28T10:55:35","date_gmt":"2021-01-28T10:55:35","guid":{"rendered":"https:\/\/ideaclan.com\/?p=1309"},"modified":"2021-02-02T07:16:49","modified_gmt":"2021-02-02T07:16:49","slug":"what-are-kubernetes-pods-and-how-to-run-it","status":"publish","type":"post","link":"https:\/\/ideaclan.com\/what-are-kubernetes-pods-and-how-to-run-it\/","title":{"rendered":"What Are Kubernetes Pods and How To Run It?"},"content":{"rendered":"

Pods are defined as the smallest, most basic deployable objects in Kubernetes<\/a>. It represents a single instance of a running process in your cluster. Pods can be considered to be a self-contained, isolated \u201clogical host\u201d that contains the systemic needs of the application it serves.<\/p>\n

In this article, you will learn about Kubernetes pods, use causes, and lifecycles, as well as how to use pods to deploy an application.<\/p>\n

What Is A K8s Pod?<\/strong><\/h2>\n

Among all the objects in Kubernetes, the pod is the smallest building block. Within a cluster, a pod represents a process that’s running. The internal of a pod can have one or more containers. Those within a single pod share:<\/p>\n

    \n
  • A unique network IP<\/li>\n
  • Network<\/li>\n
  • Storage<\/li>\n
  • Any additional specifications you\u2019ve applied to the pod<\/li>\n<\/ul>\n

    An alternative way to think of a pod is as a \u201clogical host\u201d that is specific to your application and holds one or more tightly-coupled containers. For example, we have an app-container and a logging-container in a pod. The only responsibility of the logging-container is to pull logs from the app-container. Locating your containers in a pod terminates extra communication setup because they’re co-located, so everything is local, and that they share all the resources. This is similar to execution on the same physical server in a pre-container world.<\/p>\n

    There are other things to do with pods as well. You might have an init container that initializes a second container. Once the second container is up and serving, the first container\u2019s job is done and hence it stops working.<\/p>\n

    Pod Model Types<\/strong><\/h2>\n

    You can create two types of pod models which are as follows:
    \n\u2022 One-container-per-pod:<\/strong> This is the most popular model. The post is the \u201cwrapper\u201d for a single container. Because pod is the smallest object that K8s, therefore, it manages the pods instead of directly managing the containers.
    \n\u2022 Multi-container-pod:<\/strong> In this model, a pod can hold multiple co-located containers that are closely coupled to share resources. These containers work as a one, cohesive unit of service. The pods then wraps these multi containers with storage resources into a single unit. Example use cases include sidecars, proxies, logging.<\/p>\n

    Each pod runs a single instance of your application. If you need to scale the app horizontally (such as running several replicas), you can use a pod per instance. This is contrasting from running multiple containers or the same app within a single pod.<\/p>\n

    It is important to highlight that pods are not intended as durable entities. In case a node fails or if you are maintaining nodes, the pods won\u2019t be able to survive. To overcome this issue, K8S has controllers- generally, a pod can be created with a type of controller.<\/p>\n

    Pod Lifecycle Phases<\/strong><\/h2>\n

    A pod status is responsible to tell us where the pod is in the lifecycle. It is meant to just give an idea not precisely, hence, it is a better practice to debug if the pod does not come up cleanly. The five phases of a pod lifecycle are:<\/p>\n

    \u2022 Pending:<\/strong> In this phase, the pod is accepted, but at least one container image has not been created.
    \n\u2022 Running:<\/strong> In this case, the pod is bound to a node and all containers are created. One container is running or within the process of starting or restarting
    \n\u2022 Succeeded:<\/strong> In the third phase, all containers in the pod successfully terminated and will not restart.
    \n\u2022 Failed:<\/strong> In this phase, all containers get terminated, with at least one container failing. The failed container exited with a non-zero status.
    \n\u2022 Unknown:<\/strong> In this phase, the state of the pod could not be obtained.<\/p>\n

    Pods In Practice
    \nNow that you know what a pod theoretically, it is time to know how it looks in practice. First, we will go over a simple pod manifest, later, we will deploy an example app showing how you can work with it.<\/p>\n

    The Manifest (YAML)<\/strong><\/h2>\n

    The manifest can be broken down into four parts:<\/p>\n

    \u2022 ApiVersion<\/strong>– It is the version of the Kubernetes API you are using.
    \n\u2022 Kind<\/strong>– It is the kind of object you wish to create.
    \n\u2022 Metadata<\/strong>– It is the information that uniquely identifies the object, such as name or namespace.
    \n\u2022 Spec<\/strong>– It is defined as a specified configuration of our pod, for example, image name, container name, volumes, etc.<\/p>\n

    ApiVersion, kind, and metadata are the compulsory fields that are applicable to all Kubernetes objects, not just pods. The layout of the spec is also a requirement but it varies across objects. An example of manifest that is shown below shows what a single container pod spec looks like.<\/p>\n

    apiVersion<\/span>:<\/span> \"api version\"<\/span>              (<\/span>1<\/span>)<\/span>\r\nkind<\/span>:<\/span> \"object to create\"<\/span>                  (<\/span>2<\/span>)<\/span>\r\nMetadata<\/span>:<\/span>                   (<\/span>3<\/span>)<\/span>\r\n  Name<\/span>:<\/span> \"Pod name\"<\/span>\r\n  labels<\/span>:<\/span>\r\n    App<\/span>:<\/span> \"label value\"<\/span>\r\nSpec<\/span>:<\/span>                       (<\/span>4<\/span>)<\/span>\r\n  containers<\/span>:<\/span>\r\n  -<\/span> name<\/span>:<\/span> \"container name\"<\/span>\r\n    image<\/span>:<\/span> \"image to use for container\"<\/span><\/pre>\n
    After you have understood how the manifest looks, we will reveal both the models for creating a pod.<\/p>\n
    Single Container Pod<\/strong><\/h3>\n
    The pod-1.yaml is the manifest for our single container pod. It runs annginx pod that reflects something for us.<\/p>\n
    apiVersion<\/span>:<\/span> v1\r\nkind<\/span>:<\/span> Pod<\/span>\r\nmetadata<\/span>:<\/span>\r\n  name<\/span>:<\/span> firstpod\r\n  labels<\/span>:<\/span>\r\n    app<\/span>:<\/span> myapp\r\nspec<\/span>:<\/span>\r\n  containers<\/span>:<\/span>\r\n  -<\/span> name<\/span>:<\/span> my<\/span>-<\/span>first<\/span>-<\/span>pod\r\n    image<\/span>:<\/span> nginx<\/span><\/pre>\n
    Moving forward, we deploy this manifest into our local Kubernetes cluster by running Kubectl create -f pod-1.yaml. Afterwards, we run \u201ckubectl get pods\u201d to confirm if our pods is running as expected.<\/p>\n
    kubectl <\/span>get<\/span> pod\r\nNAME                                          READY     STATUS    RESTARTS   AGE\r\nfirstpod                                      <\/span>1<\/span>\/<\/span>1<\/span>       Running<\/span>   0<\/span>          45s<\/span><\/pre>\n
    You can see that it is now running. To confirm if it is actually running, run kubectl exec firstpod\u2014kubeconfig=kubeconfig \u2014 service nginx status. This runs a command inside our pod by passing in \u2014 service nginx status. (Note: this is similar to running docker exec.)<\/p>\n
    kubectl <\/span>exec<\/span> firstpod <\/span>\u2014<\/span> service nginx status\r\nnginx <\/span>is<\/span> running<\/span>.<\/span><\/pre>\n
    Now, we\u2019ll clean up by running kubectl delete pod firstpod.<\/p>\n
    kubectl <\/span>delete<\/span> pod firstpod\r\npod <\/span>\"firstpod\"<\/span> deleted<\/span><\/pre>\n
    Multi-Container Manifest<\/strong><\/h3>\n
    In this instance, we will deploy something more useful: a pod with multiple containers that work as a single entity. One container writes the present date to a file every 10 seconds while the opposite container serves the logs of us.<\/p>\n
    Move forward and deploy the pod-2.yaml manifest with kubectl create -f pod-2.yaml.<\/p>\n
    apiVersion<\/span>:<\/span> v1\r\nkind<\/span>:<\/span> Pod<\/span>\r\nmetadata<\/span>:<\/span>\r\n  name<\/span>:<\/span> multi<\/span>-<\/span>container<\/span>-<\/span>pod <\/span># Name of our pod<\/span>\r\nspec<\/span>:<\/span>\r\n  volumes<\/span>:<\/span>\r\n  -<\/span> name<\/span>:<\/span> shared<\/span>-<\/span>date<\/span>-<\/span>logs  <\/span># Creating a shared volume for my containers<\/span>\r\n    emptyDir<\/span>:<\/span> {}<\/span>\r\n  containers<\/span>:<\/span>\r\n  -<\/span> name<\/span>:<\/span> container<\/span>-<\/span>writing<\/span>-<\/span>dates <\/span># Name of first container<\/span>\r\n    image<\/span>:<\/span> alpine <\/span># Image to use for first container<\/span>\r\n    command<\/span>:<\/span> [<\/span>\"\/bin\/sh\"<\/span>]<\/span>\r\n    args<\/span>:<\/span> [<\/span>\"-c\"<\/span>,<\/span> \"while true; do date >> \/var\/log\/output.txt; sleep 10;done\"<\/span>]<\/span> # writing date every 10secs<\/span>\r\n    volumeMounts<\/span>:<\/span>\r\n    -<\/span> name<\/span>:<\/span> shared<\/span>-<\/span>date<\/span>-<\/span>logs\r\n      mountPath<\/span>:<\/span> \/var\/<\/span>log <\/span># Mounting log dir so app can write to it.<\/span>\r\n  -<\/span> name<\/span>:<\/span> container<\/span>-<\/span>serving<\/span>-<\/span>dates <\/span># Name of second container<\/span>\r\n    image<\/span>:<\/span> nginx<\/span>:<\/span>1.7<\/span>.<\/span>9<\/span> # Image for second container<\/span>\r\n    ports<\/span>:<\/span>\r\n      -<\/span> containerPort<\/span>:<\/span> 80<\/span> # Defining what port to use.<\/span>\r\n    volumeMounts<\/span>:<\/span>\r\n    -<\/span> name<\/span>:<\/span> shared<\/span>-<\/span>date<\/span>-<\/span>logs\r\n      mountPath<\/span>:<\/span> \/usr\/<\/span>share<\/span>\/<\/span>nginx<\/span>\/<\/span>html <\/span># Where nginx will serve the written file<\/span><\/pre>\n
    Here, take a brief pause and understand the volumes in pods. In the example given above, volumes provide how for the containers to communicate during the pod\u2019s life. If you delete a pod and recreate it again, any stored data in the shared volume is lost. Persistent volumes object rescues this issue so that your data exists beyond pod loss. This multi-container example is used not only to demonstrate how to create two container pods but to also show the way both containers share resources.<\/p>\n
    kubectl create <\/span>-<\/span>f pod<\/span>-<\/span>2.yaml<\/span>\r\npod <\/span>\"multi-container-pod\"<\/span> created<\/span><\/pre>\n
    Afterward, we confirm that it is actually deployed.<\/p>\n
    Now, you can see that it running. It is time to make sure that our second container is serving the dates.<\/p>\n
    Check to ensure two containers are in our pod by running kubectl describe pod \u201cpod name\u201d. This command shows what the created object seems like.<\/p>\n
    Containers<\/span>:<\/span>\r\n  container<\/span>-<\/span>writing<\/span>-<\/span>dates<\/span>:<\/span>\r\n    Container<\/span> ID<\/span>:<\/span>  docker<\/span>:<\/span>\/\/e5274fb901cf276ed5d94b625b36f240e3ca7f1a89cbe74b3c492347e98c7a5b<\/span>\r\n    Image<\/span>:<\/span>         alpine\r\n    <\/span>Image<\/span> ID<\/span>:<\/span>      docker<\/span>-<\/span>pullable<\/span>:<\/span>\/\/alpine@sha256:621c2f39f8133acb8e64023a94dbdf0d5ca81896102b9e57c0dc184cadaf5528<\/span>\r\n    Port<\/span>:<\/span>          \r\n    Host<\/span> Port<\/span>:<\/span>     \r\n    Command<\/span>:<\/span>\r\n      \/bin\/<\/span>sh\r\n    <\/span>Args<\/span>:<\/span>\r\n      -<\/span>c\r\n      <\/span>while<\/span> true<\/span>;<\/span> do<\/span> date <\/span>>><\/span> \/var\/<\/span>log<\/span>\/<\/span>output<\/span>.<\/span>txt<\/span>;<\/span> sleep <\/span>10<\/span>;<\/span>done<\/span>\r\n    State<\/span>:<\/span>          Running<\/span>\r\n      Started<\/span>:<\/span>      Fri<\/span>,<\/span> 16<\/span> Nov<\/span> 2018<\/span> 11<\/span>:<\/span>31<\/span>:<\/span>44<\/span> -<\/span>0700<\/span>\r\n    Ready<\/span>:<\/span>          True<\/span>\r\n    Restart<\/span> Count<\/span>:<\/span>  0<\/span>\r\n    Environment<\/span>:<\/span>    \r\n    Mounts<\/span>:<\/span>\r\n      \/var\/<\/span>log <\/span>from<\/span> shared<\/span>-<\/span>date<\/span>-<\/span>logs <\/span>(<\/span>rw<\/span>)<\/span>\r\n      \/<\/span>var<\/span>\/<\/span>run<\/span>\/<\/span>secrets<\/span>\/<\/span>Kubernetes<\/span>.<\/span>io<\/span>\/<\/span>serviceaccount <\/span>from<\/span> default<\/span>-<\/span>token<\/span>-<\/span>8dl5j<\/span> (<\/span>ro<\/span>)<\/span>\r\n  container<\/span>-<\/span>serving<\/span>-<\/span>dates<\/span>:<\/span>\r\n    Container<\/span> ID<\/span>:<\/span>   docker<\/span>:<\/span>\/\/f9c85f3fe398c3197644fb117dc1681635268903b3bba43aa0a1d151fab6ad22<\/span>\r\n    Image<\/span>:<\/span>          nginx<\/span>:<\/span>1.7<\/span>.<\/span>9<\/span>\r\n    Image<\/span> ID<\/span>:<\/span>       docker<\/span>-<\/span>pullable<\/span>:<\/span>\/\/nginx@sha256:e3456c851a152494c3e4ff5fcc26f240206abac0c9d794affb40e0714846c451<\/span>\r\n    Port<\/span>:<\/span>           80<\/span>\/<\/span>TCP\r\n    <\/span>Host<\/span> Port<\/span>:<\/span>      0<\/span>\/<\/span>TCP\r\n    <\/span>State<\/span>:<\/span>          Running<\/span>\r\n      Started<\/span>:<\/span>      Fri<\/span>,<\/span> 16<\/span> Nov<\/span> 2018<\/span> 11<\/span>:<\/span>31<\/span>:<\/span>44<\/span> -<\/span>0700<\/span>\r\n    Ready<\/span>:<\/span>          True<\/span>\r\n    Restart<\/span> Count<\/span>:<\/span>  0<\/span>\r\n    Environment<\/span>:<\/span>    \r\n    Mounts<\/span>:<\/span>\r\n      \/usr\/<\/span>share<\/span>\/<\/span>nginx<\/span>\/<\/span>html <\/span>from<\/span> shared<\/span>-<\/span>date<\/span>-<\/span>logs <\/span>(<\/span>rw<\/span>)<\/span>\r\n      \/<\/span>var<\/span>\/<\/span>run<\/span>\/<\/span>secrets<\/span>\/<\/span>Kubernetes<\/span>.<\/span>io<\/span>\/<\/span>serviceaccount <\/span>from<\/span> default<\/span>-<\/span>token<\/span>-<\/span>8dl5j<\/span> (<\/span>ro<\/span>)<\/span><\/pre>\n
    Now that both the containers are running, it is time to make sure both are doing their assigned jobs. Connect to the container by running kubectl exec -ti multi-container-pod -c container-serving-dates \u2014kubeconfig=kubeconfig bash. Now we are inside the container.<\/p>\n
    Finally, we run curl \u2018http:\/\/localhost:80\/output.txt\u2019 inside the container and it should serve our file. (If you don\u2019t have curl installed in the container, first run apt-get update && apt-get install curl then run curl \u2018http:\/\/localhost:80\/output.txt\u2019 again.)<\/p>\n
    curl <\/span>'http:\/\/localhost:80\/app.txt'<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>31<\/span>:<\/span>44<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>31<\/span>:<\/span>54<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>04<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>14<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>24<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>34<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>44<\/span> UTC <\/span>2018<\/span>\r\nFri<\/span> Nov<\/span> 16<\/span> 18<\/span>:<\/span>32<\/span>:<\/span>54<\/span> UTC <\/span>2018<\/span><\/pre>\n","protected":false},"excerpt":{"rendered":"
    Pods are defined as the smallest, most basic deployable objects in Kubernetes. It represents a single instance of a running process in your cluster. Pods can be considered to be…<\/p>\n","protected":false},"author":3,"featured_media":1322,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0},"categories":[9],"tags":[],"post_mailing_queue_ids":[],"_links":{"self":[{"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/posts\/1309"}],"collection":[{"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/comments?post=1309"}],"version-history":[{"count":0,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/posts\/1309\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/media\/1322"}],"wp:attachment":[{"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/media?parent=1309"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/categories?post=1309"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ideaclan.com\/wp-json\/wp\/v2\/tags?post=1309"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}