Propagate HTTP proxy config into VM

[cvila84]

BUG REPORT

Minikube version (use minikube version): v0.30.0

• OS (e.g. from /etc/os-release): Windows
• VM Driver (e.g. cat ~/.minikube/machines/minikube/config.json | grep DriverName): Virtualbox
• ISO version (e.g. cat ~/.minikube/machines/minikube/config.json | grep -i ISO or minikube ssh cat /etc/VERSION): v0.30.0

What happened:
minikube start --kubernetes-version v1.10.8 --docker-env HTTP_PROXY=xxx --docker-env HTTPS_PROXY=xxx --docker-env NO_PROXY=xxx is failing in timeout during kubeadm command execution

sudo /usr/bin/kubeadm init --config /var/lib/kubeadm.yaml --ignore-preflight-errors=DirAvailable--etc-kubernetes-manifests --ignore-preflight-errors=DirAvailable--data-minikube --ignore-preflight-errors=Port-10250 --ignore-preflight-errors=FileAvailable--etc-kubernetes-manifests-kube-scheduler.yaml --ignore-preflight-errors=FileAvailable--etc-kubernetes-manifests-kube-apiserver.yaml --ignore-preflight-errors=FileAvailable--etc-kubernetes-manifests-kube-controller-manager.yaml --ignore-preflight-errors=FileAvailable--etc-kubernetes-manifests-etcd.yaml --ignore-preflight-errors=Swap --ignore-preflight-errors=CRI  &&
sudo /usr/bin/kubeadm alpha phase addon kube-dns

.: Process exited with status 1

What you expected to happen:
minikube starts normally

Anything else do we need to know:
The faulty command is kubeadm alpha phase addon kube-dns. I replayed it inside the minikube VM to have more logs :

$ sudo kubeadm alpha phase addon kube-dns
unable to get URL "https://dl.k8s.io/release/stable-1.10.txt": Get https://dl.k8s.io/release/stable-1.10.txt: dial tcp 23.236.58.218:443: getsockopt: connection timed out

kubeadm is trying to download a file (that actually contains the exact version of kubernetes) but since HTTP proxy settings are not set in VM (they are only set for docker engine through --docker-env parameters), this is failing. I found that if you add --kubernetes-version, no download is done and the command succeeds:

$ sudo kubeadm alpha phase addon kube-dns --kubernetes-version v1.10.8
[addons] Applied essential addon: kube-dns

So a fix could be to propagate the --kubernetes-version passed to minikube start to kubeadm alpha phase addon kube-dns command.

[llsousa]

What about to create a parameter on minikube start for proxy definition inside minikube VM? Something like:

minikube start --minikube-env http_proxy=url.to.http.proxy:http_port \
    --minikube-env https_proxy=url.to.http.proxy:http_port \
    --minikube-env no_proxy=localhost,127.0.0.1                

Will create the file /etc/profile.d/proxy.sh with the content:

export http_proxy=url.to.http.proxy:http_port
export https_proxy=url.to.http.proxy:http_port
export no_proxy=localhost,127.0.0.1

In order to use docker environment, sudo should be run with -E switch, like:

sudo -E  /usr/bin/kubeadm alpha phase addon kube-dns

All the CLI using sudo should use -E switch.

Please comment.

[ldcastell]

This issue is really annoying, will be awesome to see the fix go through :). Is there a workaround to this?

[tstromberg]

@llsousa - That sounds reasonable. As a further iteration, it would be nice if we supported detecting the proxy from the environment, and automatically configured minikube appropriately.

"minikube start" should just work in every environment possible. If failing that, it should inform the user on the appropriate action necessary to fix the issue.

[AdamKong]

See this is the same issue? this is on win10, with virtualbox Ubuntu OS.

C:\Windows\system32>minikube start
o minikube v0.34.0 on windows (amd64)

Creating virtualbox VM (CPUs=2, Memory=2048MB, Disk=20000MB) ...

• "minikube" IP address is 192.168.99.130
• Configuring Docker as the container runtime ...
• Preparing Kubernetes environment ...
  @ Downloading kubelet v1.13.3
  @ Downloading kubeadm v1.13.3
  ! Failed to update cluster: downloading binaries: downloading kubeadm: Error downloading kubeadm v1.13.3: failed to download: failed to download to temp file: failed to create validator: failed to create validator: failed to download checksum file: Get https://storage.googleapis.com/kubernetes-release/release/v1.13.3/bin/linux/amd64/kubeadm.sha1: read tcp 192.168.1.107:52462->216.58.200.48:443: wsarecv: An existing connection was forcibly closed by the remote host.

• Sorry that minikube crashed. If this was unexpected, we would love to hear from you:

• https://github.com/kubernetes/minikube/issues/new

Anyone can offer me a temp solution? Thanks much!

[tstromberg]

@AdamKong - Your issue seems different, as the error is from outside of the VM environment. I'm pretty sure you just need to tell minikube about your proxy: https://github.com/kubernetes/minikube/blob/master/docs/http_proxy.md

[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

[tstromberg]

I haven't heard any more user complaints on this, so I think we've done a good enough job on this. Closing until someone opens something more specific.