When creating applications which needs to Watch the kubernetes api, there is no way of creating a new KubeConfig object created by credentials provided by kubernetes (or openshift).

Basically the Config::fromCluster() is what we need, but then available on the KubeConfig class like

const kc = new k8s.KubeConfig();
kc.fromCluster();  // <-- this is missing and so required!

Or am I doing something completely wrong here?
It's just that the application which needs to query the kubernetes api, is running in kubernetes itself, so .kube/config files are not working in this usecase.

The package.json currently has a dependency on the bluebird module, but doesn't include the typings for it. The means that any Promise<...> objects returned from the Core_V1Api are typed as any.

The fix is to include a dependency on @types/bluebird.

A temporary workaround is for the user to install @types/bluebird.

This fix for insecure-skip-tls-verify config is committed before the update of package version 0.3.0, but in the published version 0.3.0, it still compares with 'true' string instead of boolean true value in file dist/config_types.js.

I saw the examples in the main page refer to the package name as typescript and I assume we will have them in other docs/code. This issue is to track the progress of renaming all of them to javascript.

Dear team, I looked for a way to create a pod but could not find while searching code.
Is there a doc/manual page?

New versions of kubernetes have been released, this requires that the clients be regenerated for v1.10 and v1.11

There are a few tsconfig settings you might consider adopting in this project.

• esModuleInterop: make TypeScript's ES6 module import/export semantics more in line with standards, and enables other tools to understand the static module graph for things like tree-shaking.
• allowSyntheticDefaultImports: allow importing a commonjs module as the default export
• strict: make TypeScript extra pedantic about types. This ensures TypeScript is doing the best job possible enforcing the correctness of your code. It also ensures that generated d.ts files are free of loose definitions, allowing strict consumers to benefit from tighter type definitions.
• forceConsistentCasingInFileNames: ensure we always warn when casing differs because some file systems (cough NTFS cough) aren't strict about casing by default. This helps ensure Windows contributors don't get failures in CI due to casing issues.
• importHelpers: keep package size down by pulling out common helpers into external dependencies (these can be bundled into the final artifact by a build tool like rollup if you desire).
• declarationMap: if you're ok including the original source in the package (IMO very nice for debugging and go-to-definition for inspecting implementation details), declarationMap is needed to make it work.

For anyone with the the following error:

ode:16396) UnhandledPromiseRejectionWarning: Error: Server responded with a non-101 status: 500 Internal Server Error
Response Headers Follow:
date: Mon, 24 Sep 2018 16:01:57 GMT
content-length: 75
content-type: text/plain; charset=utf-8

    at WebSocketClient.failHandshake (../nodes/node_modules/websocket/lib/WebSocketClient.js:339:32)
    at ClientRequest.<anonymous> (../nodes/node_modules/websocket/lib/WebSocketClient.js:278:18)
    at ClientRequest.emit (events.js:182:13)
    at HTTPParser.parserOnIncomingClient (_http_client.js:555:21)
    at HTTPParser.parserOnHeadersComplete (_http_common.js:109:17)
    at TLSSocket.socketOnData (_http_client.js:441:20)
    at TLSSocket.emit (events.js:182:13)
    at addChunk (_stream_readable.js:283:12)
    at readableAddChunk (_stream_readable.js:264:11)
    at TLSSocket.Readable.push (_stream_readable.js:219:10)
    at TLSWrap.onread (net.js:639:20)
(node:16396) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 1)
(node:16396) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.

just simply override the following code in file node_modules/@kubernetes/client-node/dist/attach.js,

"use strict";
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
    return new (P || (P = Promise))(function (resolve, reject) {
        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
        function step(result) { result.done ? resolve(result.value) : new P(function (resolve) { resolve(result.value); }).then(fulfilled, rejected); }
        step((generator = generator.apply(thisArg, _arguments || [])).next());
    });
};
Object.defineProperty(exports, "__esModule", { value: true });
const querystring = require("querystring");
const web_socket_handler_1 = require("./web-socket-handler");
class Attach {
    constructor(config) {
        this.handler = new web_socket_handler_1.WebSocketHandler(config);
    }
    attach(namespace, podName, containerName, stdout, stderr, stdin, tty) {
        return __awaiter(this, void 0, void 0, function* () {
            const query = {
                stdout: stdout != null,
                stderr: stderr != null,
                stdin: stdin != null,
                tty,
                container: containerName,
            };
            const queryStr = querystring.stringify(query);
            const path = `/api/v1/namespaces/${namespace}/pods/${podName}/attach?${queryStr}`;
            const conn = yield this.handler.connect(path, null, (streamNum, buff) => {
                web_socket_handler_1.WebSocketHandler.handleStandardStreams(streamNum, buff, stdout, stderr);
            });
            if (stdin != null) {
                web_socket_handler_1.WebSocketHandler.handleStandardInput(conn, stdin);
            }
            return conn;
        });
    }
}
exports.Attach = Attach;
//# sourceMappingURL=attach.js.map

Hope it works for you too!

Hello,

I was trying to use the Apps API instead of the Core API and noticed there aren't helpers for creating clients from a config. I ended up writing my own that look almost identical to the config ones provided for Core. Should we add helpers for Apps as well? I can just submit a PR.

Thanks,
Sheldon

ref: kubernetes/community#1721

This repo is listed in sigs.yaml as belonging to sig-api-machinery, but it lacks an OWNERS file. Can you please add one with the appropriate list of approvers and reviewers? eg:

• robust example: approvers, reviewers, aliases
• simple example: only approvers, no aliases

For more information see the OWNERS docs

/sig api-machinery

/assign @lavalamp @deads2k
sig-api-machinery chairs

/assign @brendanburns
as the person merging PR's recently

As per the email sent to kubernetes-dev[1], please create a SECURITY_CONTACTS
file.

The template for the file can be found in the kubernetes-template repository[2].
A description for the file is in the steering-committee docs[3], you might need
to search that page for "Security Contacts".

Please feel free to ping me on the PR when you make it, otherwise I will see when
you close this issue. :)

Thanks so much, let me know if you have any questions.

(This issue was generated from a tool, apologies for any weirdness.)

[1] https://groups.google.com/forum/#!topic/kubernetes-dev/codeiIoQ6QE
[2] https://github.com/kubernetes/kubernetes-template-project/blob/master/SECURITY_CONTACTS
[3] https://github.com/kubernetes/community/blob/master/committee-steering/governance/sig-governance-template-short.md

Hi,
Everytime I try to patch a node, I hit a HTTP 415
the server responded with the status code 415 but did not return more information.

After some researches, I found that the API wait for Content-Type: application/strategic-merge-patch+json header for PATCH operations.
But this header is not set.

Is there any way I can override headers before sending the request ?

As part of kubernetes-client/gen#59, I am trying to generate Javascript client for Voyager. I tried to follow the regeneration instructions here: https://github.com/kubernetes-client/javascript#re-generating-code . But those instruction are outdated. How do I generate javascript client ?

#65 has instructions for a local install from git, however this is inconvenient for CI/CD processes as we need to download and build a dependency which results in longer build times, and increased complexity.

NPM supports installs from github via npm install kubernetes-client/javascript, however the current directory structure prevents this from working.

To solve these issues:

• flatten the project structure so package.json is in the root of the repo
• stick to one project per repo, if there is a jquery client in the future it should have it's own repo

I dug through the api file and could not find anything that hit the https://docs.openshift.org/latest/rest_api/api/v1.Namespace.html#Get-api-v1-namespaces endpoint. Is there a different way this is exposed in this library or is it just missing?

I just installed the package and encountered a build error because the modules were not found in index.ts. At face value seems that the problem is that index.ts should be index.d.ts but delving a bit deeper into the problem, it seems that the npm package does not contain the contents of the dist directory generated by an npm build (of the kubernetes-client project).

Currently the watch function will exit if the stream is empty, we should have a flag for continuous monitoring. Or is there a better way to be watching for resources?

var kc = new k8s.KubeConfig();
kc.loadFromDefault();

var k8sApi = kc.makeApiClient(k8s.AppsV1beta1Deployment);
//var k8sApi = kc.makeApiClient(k8s.AppsV1beta1DeploymentList);

k8sApi.listNamespacedDeployment.then( (res) => {
    return res.body;
});

i get this error:

"TypeError: apiClient.setDefaultAuthentication is not a function"

Originally posted by @scher200 in #103 (comment)

Where can I find the documentation for this client?

I am trying to scale my deployment, but the client doesn't appear to have the scale nor deployment command.
https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.9/#deployment-v1-apps

In the following example

import k8s = require('@kubernetes/client-node');
var client: k8s.Core_v1Api;
client = k8s.Config.defaultClient();

typescript produces the following error

error TS2694: Namespace '"<path>/@kubernetes/client-node/index"' has no exported member 'Core_v1Api'.

This could be related to #23 but I also did not see it exported in the distribution. This example is overly simplified to demonstrate the problem.

This appears to be more of an upstream issue with swagger-codegen (ref swagger-api/swagger-codegen#7930) but I wanted to flag it here.

Basically as it stands, adding a new option is likely to break compatibility, and the usability isn't great as it is... To illustrate I just wrote this call: client.listNamespacedPod(namespace, undefined, undefined, undefined, undefined, labelSelector) - who knows what sequence the parameters will be in for the next k8s version :)

I'd at least suggest setting the appropriate configuration flag if the above issue gets resolved.

I propose the following changes to the API object model:

1. Make non-required fields (and fields that aren't apiVersion or kind) in all API objects optional e.g., containers: V1Container -> containers?: V1Container.
2. Automatically generate type guards to make it easy to condition on the types of API objects you're interacting with in TypeScript.
3. Change all API objects in api.ts to be interfaces rather than classes.
4. Add a utility type that represent the set of top-level API objects (e.g., type CoreTypes = AppsV1beta1Deployment | AppsV1beta1DeploymentList | [...]), so that you can deal with a well-typed collection of objects for some API group.

Justification for (1)

Suppose I manually construct an object so that I can later kubectl it to the server:

        const svc = <k8s.V1Service>{
          "kind": "Service",
          "apiVersion": "v1",
          "metadata": {
            name: name,
          },
          "spec": {}
        };

This will actually not compile because k8s.V1Service requires a status field. This field need not exist for our purposes, though. It is possible to work around this by casting to any first (e.g., const svc = <k8s.V1Service><any>{ [...] }, but this is not a particularly compelling "blessed pattern", because status actually is an optional field.

This approach will make apps using --strictNullCheck a little more verbose, but really not checking null here is just hiding from the real issue, which is that it can very well be null.

Justification for (2)

Using the API objects in api.ts is pretty painful in TypeScript.

Let's say you are writing an app in TypeScript, and have some object const foo: AppsV1beta1Deployment | AppsV1beta2Deployment | ExtensionsV1beta1Deployment;.

If you want to figure out which of these types foo really has at runtime, you would ordinarily do something like:

if (foo instanceof AppsV1beta1Deployment) {` 

But if you have built up an API object using an object literal (as we did in (1)), then that object literal will not be an instanceof any of them, because it will be of type object. This means that you have to write your own type guard that checks kind and apiVersion, but really this should just be exposed by the library itself.

Furthermore, if you did want to use the official constructors in api.ts, you'd need to do something like the below:

const svc = new k8s.V1Service();
svc.metadata = new k8s.V1ObjectMeta();
svc.metadata.name = name;
svc.spec = new k8s.V1ServiceSpec();

Needless to say, this is not ideal.

Justification for (3)

In addition to points (1) and (2), we're not exposing any methods or custom constructors on the API object classes in api.ts, so we're not getting any benefit from them being classes, and there are significant drawbacks.

Is there a way to get the beta API to use it for making requests? I want to make a createNamespacedDeployment request, but it only exists on Extensions_v1beta1Api.

Hi there,

I'm trying to use this on the OTC (open telecom cloud).

They provide a forked version for kubectl here:
https://github.com/Cusfomer-kubectl/kubernetes

And in the end I have a kube config file that looks like this:

apiVersion: v1
clusters:
- cluster:
    cluster-uuid: 4f65a35c-a9c8-4c98-87a4-xxxxxxxxxxxx
    server: https://cce.eu-de.otc.t-systems.com
  name: fg-dev
contexts:
- context:
    cluster: fg-dev
    user: frederik_berg
  name: fg-dev
current-context: fg-dev
kind: Config
preferences: {}
users:
- name: frederik_berg
  user:
    access-key: <access-key consisting of uppercase alphanum chars>
    region-id: eu-de
    secret-key: <secret-key consisting of mixed case alphanum chars>

the config_types.js looks like it dosn't support this:

function userIterator() {
    return function (elt, i, list) {
        if (!elt.name) {
            throw new Error(`users[${i}].name is missing`);
        }
        let token = null;
        if (elt.user["token"]) {
            token = elt.user["token"];
        }
        if (elt.user["token-file"]) {
            token = fs.readFileSync(elt.user["token-file"]);
        }
        return {
            name: elt.name,
            certData: elt.user["client-certificate-data"],
            certFile: elt.user["client-certificate"],
            keyData: elt.user["client-key-data"],
            keyFile: elt.user["client-key"],
            authProvider: elt.user["auth-provider"],
            token: token,
        };
    };
}

Is there any way I can make this work?

P.S. this is using the newest version available from npm/yarn not from git directly

There is no package.json.

Command-line: kubectl log <pod-name>
Documentation: https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.11/#read-log-pod-v1-core

From what I read so far on kubernetes/kubernetes#13885, the /log endpoints only expects binary.k8s.io as protocol. Without it (but still sending the *.channel.k8s.io ones), a 403 Forbidden status code is returned.

I'll try to submit a PR for this in the upcoming days, suggestions are welcome.

Right now I'm trying to fit this into WebSocketHandler, but there is already a binary handler that expects the first byte of the message to be the channel number, which is not the case for the /log endpoint. See https://github.com/kubernetes-client/javascript/blob/master/node-client/src/web-socket-handler.ts#L89-L94

When running nsp against a project with @kubernetes/client-node 0.2.1 The following vulnerable dependency is highlighted:

> nsp check

(+) 1 vulnerability found
┌────────────┬────────────────────────────────────────────────────────────────────┐
│            │ Sandbox Breakout / Arbitrary Code Execution                        │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ Name       │ static-eval                                                        │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ CVSS       │ 6.2 (Medium)                                                       │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ Installed  │ 0.2.3                                                              │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ Vulnerable │ <=1.1.1                                                            │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ Patched    │ >=2.0.0                                                            │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ Path       │ [email protected] > @kubernetes/[email protected] > [email protected] >  │
│            │ [email protected]                                                  │
├────────────┼────────────────────────────────────────────────────────────────────┤
│ More Info  │ https://nodesecurity.io/advisories/548                             │
└────────────┴────────────────────────────────────────────────────────────────────┘

The static-eval package should be upgraded to version 2.0.0.

Hi! I'm using 0.2.1 and i've noticed that createNamespacedDeployment is not available on Core_v1Api
It is available only under Apps_v1Api but it is not clear for me how to use it
Can you point me in some direction please?

When using Google Kubernetes Engine on Windows, the Kubernetes client
sometimes fails with the following error:

C:\...\node_modules\@kubernetes\client-node\dist\config.js:99
                            throw new Error('Failed to refresh token: ' + result);
                            ^

Error: Failed to refresh token:
    at KubeConfig.applyAuthorizationHeader (C:\...\node_modules\@kubernetes\client-node\dist\config.js:99:35)
    at KubeConfig.applyOptions (C:\...\node_modules\@kubernetes\client-node\dist\config.js:126:14)
    at KubeConfig.applyToRequest (C:\...\node_modules\@kubernetes\client-node\dist\config.js:138:14)
    at Core_v1Api.createNamespace (C:\...\node_modules\@kubernetes\client-node\dist\api.js:27561:38)
    at Object.<anonymous> (C:\...\create-namespace.js:19:5)
    at Module._compile (module.js:643:30)
    at Object.Module._extensions..js (module.js:654:10)
    at Module.load (module.js:556:32)
    at tryModuleLoad (module.js:499:12)
    at Function.Module._load (module.js:491:3)
    at Function.Module.runMain (module.js:684:10)
    at startup (bootstrap_node.js:187:16)
    at bootstrap_node.js:608:3

The content of result.stderr:

'C:\\Users\\cbuchacher\\AppData\\Local\\Google\\Cloud\' is not recognized as an internal or external command,\r\noperable program or batch file.\r\n'

The content of user.authProvider.config['cmd-path']:

C:\Users\cbuchacher\AppData\Local\Google\Cloud SDK\google-cloud-sdk\bin\gcloud.cmd

Is it supposed to handle transient faults like 504 by retrying, or have users handle?
It seems to me the fix could be as simple as using one of request wrappers like node-request-retry in place of request, for example.
Any thoughts would be appreciated.

It seems the conversion from Swagger to TypeScript incorrectly makes every parameter on request body types required. In a lot of cases the request parameters contain all of the properties that you'd expect when fetching objects, which ofter you can't use in the request body, so the only feasible workaround is to cast to with <any> or Partial<T>. For example, status is required in the interface for the body for createNamespace, which of course doesn't make sense when creating a namespace.

I'm not sure if this is an upstream issue with swagger-codegen or the configuration, but for the library to work as expected, there probably needs to be a separation between types returned from read/list calls, and the types used for create/patch calls.

Let me know if we can help with this, we're likely to be heavy users of the library so we're happy to contribute, but would probably need some pointers to get started .)

I'm trying to use the node client from a Typescript file, but I can get the compiler to be happy when including it. Doing:
import { Config } from '@kubernetes/client-node/dist/src/config';
the compiler complains that it can't find the definition file for bluebird. This is because it's in the devDependencies of the client's package.json, so it doesn't get installed when doing npm install.

What is the recommended way to use the package from Typescript? An example would be appreciated.

When using var k8sApi = k8s.Config.defaultClient(); to get the default client, if the users .kube/config file includes a context with insecure-skip-tls-verify the certificate-authority is not necessarily going to be present. See example below:

- cluster:
    insecure-skip-tls-verify: true
    server: https://127.0.0.1:443
  name: my-cluster

In these cases the call to defaultClient() fails with the following error:

Uncaught Error: clusters[0].cluster.[certificate-authority-data, certificate-authority] is missing

This is trigged by Config Types validation code which tests for a certificate authority even if insecure-skip-tls-verify is set to true. Source

This prevents using this library even if the currently set context does provide a context-authority since all context configurations are validated during the iteration.

I would like to be able to connect to some clusters using the insecure flag.

Error : UnhandledPromiseRejectionWarning: TypeError: Cannot set property 'Authorization' of undefined

Raised from here (opts.headers does not exists) https://github.com/kubernetes-client/javascript/blob/master/node-client/src/config.ts#L156

When I patch it :

if (token) {
            if(!opts.headers) {
                opts.headers = {};
            }
            opts.headers['Authorization'] = token;
 }

I get : Error: Server responded with a non-101 status: 400 Bad Request at WebSocketClient.failHandshake (audit-id: 8c520bd1-f6b8-4c98-9175-b77e23724fc8)

I'm using GCE with Kubernetes 1.8 with .kube/config's user :

name: kubectl-exec
  user:
    token: e...Q

my script :

import k8s = require('@kubernetes/client-node');
let command = process.argv[2];

let kc = new k8s.KubeConfig();
kc.loadFromFile(process.env['HOME'] + '/.kube/config');

let exec = new k8s.Exec(kc);
exec.exec('default', 'dep-yolo-6476bc5465-9s6bm', 'yolo', command, process.stdout, process.stderr, process.stdin, true /* tty */);

Other commands (list namespaces, get pod ...) are working fine

It seems the watch class only accepts a KubeConfig object, the example shows loading from a file.

However within a pod we should be using token and crt. How do we apply these to the Watch object?

See #21

We have a use-case of loading config params from another Json Object, loaded from persistence layer - as the target cluster is dynamic in our case.

It would be helpful to have Config read from Buffer or by invoking parameterised constructor:

{
  kubernetesPassword: string;
  kubernetesUrl: string;
  kubernetesUserName: string;
  registrySecret: string;
  caCert: string;
  cert: string;
  key: string;
}

71f0ac8

const kc = new k8s.KubeConfig();
kc.loadFromCluster();

const k8sApi = kc.makeApiClient(k8s.Core_v1Api);

k8sApi.listNamespacedPod('default')
    .then((res) => {
	console.log(res.body);
    })
    .catch((err) => {
        console.log(err);
    });

has a call to kc.loadFromCluster() and loadFromCluster is not a method. What should we be using?

When we use service account (e.g. login in cluster mode) and wanna run exec command, it will override websocket headers with Authorization: ...

        this.config.applytoHTTPSOptions(opts);
        const client = new ws.client({ tlsOptions: opts });

How it used:
https://github.com/theturtle32/WebSocket-Node/blob/7a30e3949a2e5b4f7ea701b7959cc8bd030f4c8f/lib/WebSocketClient.js#L236

       for (var key in self.config.tlsOptions) {
           if (self.config.tlsOptions.hasOwnProperty(key)) {
               requestOptions[key] = self.config.tlsOptions[key];
           }
       }

As a result, request options will be:

{ agent: false,
  hostname: '11.22.33.44',
  port: '6443',
  method: 'GET',
  path: '/api/v1/namespaces/monitoring/pods/xxx/exec?stdout=true&stderr=true&stdin=true&tty=true&command=sh&command=xxx&container=yyy',
  headers: 
   [ Authorization: 'Bearer aabbccddee...' ],
  ca: <Buffer ... >,
  cert: null,
  key: null }

And response will be:

Error: Server responded with a non-101 status: 400 Bad Request: missing required Host header

How to reproduce:

Run exec command from in-cluster or use service account in kube config

With 1.10, this exec-based client auth credential provider alpha feature was introduced kubernetes/features#541 for client-go. Exec-based auth plugin should be supported by this Javascript client in addition to client-go. Umbrella issue: kubernetes#62185

/kind feature
/sig api-machinery

For eg : /apis/apps/v1beta1/deployments has moved into core now and the new url is /apis/apps/deployments

Cluster version info :

root@kubemaster:/home/test# kubectl version
Client Version: version.Info{Major:"1", Minor:"9", GitVersion:"v1.9.3", GitCommit:"d2835416544f298c919e2ead3be3d0864b52323b", GitTreeState:"clean", BuildDate:"2018-02-07T12:22:21Z", GoVersion:"go1.9.2", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"9", GitVersion:"v1.9.3", GitCommit:"d2835416544f298c919e2ead3be3d0864b52323b", GitTreeState:"clean", BuildDate:"2018-02-07T11:55:20Z", GoVersion:"go1.9.2", Compiler:"gc", Platform:"linux/amd64"}

Please update the apis

Managed to download file from pod by creating a writestream and using it as stdout

const downloadCommand = ["tar", "czpf", "-", `${remotePath}`];
const localFile = fs.createWriteStream(localPath);
Exec.exec(nameSpace, podName, containerName, downloadCommand, localFile, null, null, false);

But I wouldn't know how to upload a file/folder to the container.

Anyone can help me out?

I'd like to use latest recent changes (Add a fromCluster option to KubeConfig.)

How to install this package from git?

@brendandburns is there some k8s machinery that we can use to automate running the generator when upstream cuts a new version?

I think this library could include kube resources definitions. I have one for my own, look something like this:

/**
 * https://kubernetes.io/docs/api-reference/v1.6/#deploymentspec-v1beta1-apps
 */
export interface DeploymentSpec {
  /**
   * Minimum number of seconds for which a newly created pod should be ready without any of its container crashing, for it to be considered available. Defaults to 0 (pod will be considered available as soon as it is ready)
   */
  minReadySeconds?: number,
  /**
   * Number of desired pods. This is a pointer to distinguish between explicit zero and not specified. Defaults to 1.
   */
  replicas?: number,
  /**
   * The number of old ReplicaSets to retain to allow rollback. This is a pointer to distinguish between explicit zero and not specified. Defaults to 2.
   */
  revisionHistoryLimit?: number,
  /**
   * The config this deployment is rolling back to. Will be cleared after rollback is done.
   */
  rollbackTo?: RollbackConfig,
  /**
   * Label selector for pods. Existing ReplicaSets whose pods are selected by this will be the ones affected by this deployment.
   */
  selector?: LabelSelector,
  /**
   * The deployment strategy to use to replace existing pods with new ones.
   */
  strategy?: DeploymentStrategy
  /**
   * Template describes the pods that will be created.
   */
  template: PodTemplateSpec
}

export interface Deployment extends Resource {
  apiVersion: 'extensions/v1beta1'
  kind: 'Deployment',
  spec: DeploymentSpec
}

Though they are manually created, not generated as I assume can be.

I use those interfaces to created helm charts or kube configs (JS results are converted to yalm configs and applied to server). I find using Typescript for making configs much more convenient, flexible and safe than directly writing configs in native YAML format.

i am trying to access the pod with the below command, do i have to add something more for access the pod

const k8s = require('@kubernetes/client-node');

var k8sApi = k8s.Config.defaultClient();
k8sApi.listNamespacedPod('default')
    .then((res) => {
        console.log(res.body);
    });

i am new to this, can u give me simple example

Publishing .ts files will cause builds to fail on all projects that are using TypeScript's strict null checking, even in the with --skipLibCheck, --skipDefaultLibCheck, and the @kubernetes/client-node file placed in the excludes list.

Instead, you probably want to:

• Apply the following diff to the package.json, so that the dist/index.js is properly targeted and the typings file dist/index.d.ts is included:

  @@ -10,7 +10,8 @@
       "*.ts",
       "*.js"
     ],
  -  "main": "index.js",
  +  "main": "dist/index.js",
  +  "types": "dist/index.d.ts",
     "scripts": {
       "clean": "rm -Rf node_modules/ dist/",
       "build": "tsc",

• Increment the version to 0.1.4.
• Run npm publish using npm client > 5. (4 will probably not work; I was able to publish a mirror here using 5.7.2.

There have been a handful of updates to the client since the last publish from 6c4fc27. Is it possible to publish a new version? I would be happy to tag and publish as well.

Thanks!

This is an umbrella issue for remaining tasks to promote javascript client to beta.

@brendandburns would you add remaining tasks here and let me know how can I help?