[FINE] kubernetes_connect, openshift_connect: add timeout settings

[cben]

Backport of ManageIQ/manageiq-providers-kubernetes#10
plus ManageIQ/manageiq-providers-openshift#8 (unnecessary on master but required in backports)

https://bugzilla.redhat.com/show_bug.cgi?id=1440950 (master)

1. I suppose bugzilla flags have to be confirmed & cloned first...
2. bump kubeclient ~> 2.4.0 manageiq-gems-pending#156 must be fine/backported before this (this requires kubeclient 2.4.0)
3. Then this could pass tests and could land on fine.
4. Then I'll repeat for EUWE...

cc @simon3z @simaishi.

[miq-bot]

Checked commit cben@d972272 with ruby 2.2.6, rubocop 0.47.1, and haml-lint 0.20.0
3 files checked, 0 offenses detected
Everything looks fine. 🍪

[simon3z]

@cben how come ManageIQ/manageiq-providers-openshift#8 is closed?

[cben]

Because given ManageIQ/manageiq-providers-openshift#7 it's both unnecessary and impossible to merge ManageIQ/manageiq-providers-openshift#8. That duplicated code doesn't exist any more on master.
And you said ManageIQ/manageiq-providers-openshift#7 should not be backported being a refactoring (makes sense).

If you have a less messy suggestion, I'm all ears :)

[simon3z]

LGTM 👍
cc @simaishi

[simon3z]

Because given ManageIQ/manageiq-providers-openshift#7 it's both unnecessary and impossible to merge ManageIQ/manageiq-providers-openshift#8. That duplicated code doesn't exist any more on master.
And you said ManageIQ/manageiq-providers-openshift#7 should not be backported being a refactoring (makes sense).

If you have a less messy suggestion, I'm all ears :)

@cben up to you in this case you can squash that tiny refactor here.

[simaishi]

@cben For Euwe, updating kubeclient to 2.4.0 (ManageIQ/manageiq-gems-pending#156) will need to happen in gems/pending/Gemfile. Please include that change in your Euwe PR.

[simon3z]

@cben please make Travis happy.

[cben]

See PR description, ManageIQ/manageiq-gems-pending#156 must be backported first.
(on fine manageiq-gems-pending was already split so I expect a clean cherry-pick, @simaishi let me know if that too need a [FINE] pr)

[simaishi]

@cben that looks like it's a clean cherry-pick. I can't backport non-blocker PRs to fine right now. I'll restart Travis after I backport ManageIQ/manageiq-gems-pending#156 in a few weeks.