Add support for image streaming/GCFS

[sasha-s]

Fixes #1149

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

For more information, open the CLA check for this pull request.

[comment-bot-dev]

Thanks for the PR! 🚀
✅ Lint checks have passed.

[bharathkkb]

Thanks for the PR @sasha-s
Can we add this to the nodepool variable docs https://github.com/terraform-google-modules/terraform-google-kubernetes-engine#node_pools-variable

[sasha-s]

@bharathkkb Anything I should add to the PR?

[lauraseidler]

thanks for taking care of this (I opened the original issue) - if you have the time, maybe we can also add it to the safer cluster variants? or I can also do that in a follow up

[bharathkkb]

@lauraseidler I will leave it upto @sasha-s, otherwise happy to accept a follow up too.

[sasha-s]

I do not have access to the CI build, what is going on there? @bharathkkb could you please have a look?
Also, it would be great if someone (@lauraseidler ?) with better knowledge of terraform than me could add streaming to safer cluster variants.

BTW, should it be

min_cpu_platform = lookup(each.value, "min_cpu_platform", "")

here
https://sourcegraph.com/github.com/terraform-google-modules/terraform-google-kubernetes-engine@bbd9b770987b6e8e502ce3747ccc6ce0a96e79c1/-/blob/cluster.tf?L268:5&subtree=true ?

(instead of var.node_pools[0])

[sasha-s]

Ping.

[bharathkkb]

@sasha-s CI error looks like a flake- I have re triggered it.

BTW, should it be min_cpu_platform = lookup(each.value, "min_cpu_platform", "")
here https://sourcegraph.com/github.com/terraform-google-modules/terraform-google-kubernetes-engine@bbd9b770987b6e8e502ce3747ccc6ce0a96e79c1/-/blob/cluster.tf?L268:5&subtree=true ?
(instead of var.node_pools[0])

Yeah we should switch that to each.value. Otherwise it will set the value for the first nodepool for all nodepools.

[sasha-s]

Fixed the min_cpu_platform lookup in #1207

[lauraseidler]

I just realised this is configuration on node pool level - so it will automatically also work for the safer clusters :) thanks for implementing!