"Upgrade request required" when posting port forward

[yanivoliver]

When attempting to post a port forward to a pod, for example:

v1 = client.CoreV1Api()
v1.connect_post_namespaced_pod_portforward("pod-name", "namespace-name")

An error is thrown:

*** ApiException: (400)
Reason: Bad Request
HTTP response headers: HTTPHeaderDict({'Date': 'Thu, 26 Jul 2018 14:11:26 GMT', 'Content-Length': '139', 'Content-Type': 'application/json'})
HTTP response body: {"kind":"Status","apiVersion":"v1","metadata":{},"status":"Failure","message":"Upgrade request required","reason":"BadRequest","code":400}

This happens regardless of whether the additional param of 'ports' is passed to the API call.

Environment:

Python 2.7.10 (default, Jul 15 2017, 17:16:57)
[GCC 4.2.1 Compatible Apple LLVM 9.0.0 (clang-900.0.31)] on darwin
Package version: kubernetes (6.0.0)

[slindner05]

Anyone ever make progress on this or get port-forward to work from within the python client?

[ssharma555]

I am facing this issue. Any steps to get this working?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[yanivoliver]

/remove-lifecycle stale

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[gpkc]

/remove-lifecycle stale

[micw523]

This should be tracked to #166
/close

[k8s-ci-robot]

@micw523: Closing this issue.

In response to this:

This should be tracked to #166
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[atmask]

This still seems to be a relevant issue even after the release of 12.0.0. Was any progress made here?

[trobert2]

I am also facing this issue. Could it be possible to get some guidance?

[eguven]

I don't think that this

client.CoreV1Api().connect_post_namespaced_pod_portforward('pod-name', 'namespace-name')

is intended to be used directly. In examples, you wrap it to get a raw socket. Or you can use the create_connection monkey patching option.

[dminca]

I managed to get this working by waiting for the Pod until it reports Ready state

from kubernetes import client, utils, watch
from kubernetes.client.rest import logger
from kubernetes.stream import portforward
import time

k8s_objects = utils.create_from_yaml(
    k8s_client=self._kubernetes_client,
    yaml_file=filename,
    verbose=True
)
w = watch.Watch()
core_v1 = self._kubernetes_core_v1_client
start_time = time.time()
for event in w.stream(
        func=core_v1.list_namespaced_pod,
        namespace=k8s_objects[0][0].metadata.namespace,
        label_selector="env=integration,tier=backend",
        timeout_seconds=60):
    if event["object"].status.phase == "Running":
        w.stop()
        end_time = time.time()
        logger.info("%s started in %0.2f sec", k8s_objects[0][0].metadata.name, end_time - start_time)
        pf = portforward(
            self._kubernetes_core_v1_client.connect_get_namespaced_pod_portforward,
            name="myapp",
            namespace="default",
            ports=8080
        )
        self._tcp = pf.socket(8080)
        self._tcp.setblocking(True)
        logger.info("Port %s exposed. You can reach it on localhost.", pf.local_ports[8080].port_number)
        return
    if event["type"] == "DELETED":
        logger.debug("%s deleted before it started", k8s_objects[0][0].metadata.name)
        w.stop()
        return

You'll need watch permission on the pods resource

apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: watch-permission
rules:
- apiGroups:
  - ""
  resources:
  - pods
  - pods/portforward
  verbs:
  - watch