Cannot start kube-solo when working behind http proxy

[knl]

Hi all,

I heard a lot about kubernetes and I wanted to try it. Hence I discovered kube-solo and wanted to use it to learn kubernetes. However, my laptop is behind a corporate firewall, and I can't get kube-solo to work, because corectl is failing, as it can't download coreos images. After I briefly switched to a public network and downloaded things, I was able to start my local kubernetes cluster. However, after I got back to the corporate network, and re-tried the things (Kube-solo -> Up), it hanged at "Waiting for Kubernetes cluster to be ready. This can take a bit..." (I waited one hour...):

knezn@mac001 % /Applications/Kube-Solo.app/Contents/Resources/up.command; exit;

Starting VM ...

> booting k8solo-01 (1/1)
---> 'k8solo-01' started successfuly with address 192.168.64.2 and PID 4260
---> 'k8solo-01' boot logs can be found at '/Users/knezn/.coreos/running/90394897-5956-4F9D-AFC4-E274C45DAB31/log'
---> 'k8solo-01' console can be found at '/Users/knezn/.coreos/running/90394897-5956-4F9D-AFC4-E274C45DAB31/tty'

Waiting for etcd service to be ready on VM...
...

Waiting for Kubernetes cluster to be ready. This can take a bit...

As far as I can see, there are no logs that could point me to direction why kubernetes are slow to start or hanging.

I have http_proxy, https_proxy, and no_proxy set in my ~/.zshrc which get sourced anytime kube-solo starts something. However, it seems that no tool used by kube-solo honors those settings.

Since we're not the only company using proxies (but not in production), and kube-solo is aimed at developer's machines, it would be great if the whole toolset honors proxy settings.

[rimusz]

please open an issue at corectl as kube-solo is just a wrapper around it.

[verystrongjoe]

any update about this?

[rimusz]

do not think so @AntonioMeireles had a chance to get it working yet