-
Notifications
You must be signed in to change notification settings - Fork 262
Libvirt support for bootstrap + etcd phases #3213
Conversation
|
Can one of the admins verify this patch? |
b117489 to
344f930
Compare
|
ok to test |
344f930 to
22541be
Compare
modules/libvirt/volume/main.tf
Outdated
| @@ -0,0 +1,9 @@ | |||
| variable "coreos_qow_path" { | |||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: variables are usually in variables.tf
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes. Yes they are.
|
I tried running this, but it ultimately failed while trying to create AWS resources (while realizing |
|
@crawford I'm not sure where that's happening. Right now, we only support the bootstrap phase, so you have to manually run I tried this without a valid AWS configuration and it worked. |
22541be to
ed52540
Compare
|
retest this please |
ed52540 to
1c6068d
Compare
|
all green! don't merge until the operator changes are merged. |
|
retest this please |
|
Failing on container linux version - should I skip this test? |
|
@enxebre I thought you had disabled the check CL version test, no? |
|
@squat @squeed the test disabled is for the cl channel, the cl version should succeed, this probably needs to be updated as this PR is changing the paths https://github.com/coreos/tectonic-installer/blob/master/tests/rspec/lib/aws_cluster.rb#L269 |
1c6068d to
8002aab
Compare
|
@enxebre ah, good find! let's see if this does it. |
|
rebased. |
|
The tests are failing, and I don't know how they ever worked in UT2: it's trying to get the worker instance ids from the bootstrap.tfstate file. Now that we have split tfstates... |
|
@squeed tests are failing because they can't determine the bootkube service has run successfully because they can't ssh into the machines. Have you deployed this locally successfully? I'll try to give it a go |
… libvirt This does the following: * Move most of asset generation to assets/base * Consumes this "base" module in assets/aws * Moves aws-specific variables to variables-aws.tf * Adds the beginning of libvirt support
Just adds some libvirt plumbing
|
It's red because I typo'd the tnc image name. Let's try this again. |
|
All green! ready to merge for real, this time. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@squeed the PR looks great to me. The only thing I'm wondering is, since this is a 3rd party provider, should we document where/how to get the libvirt provider?
We'd been defaulting it to ClusterName in InstallConfig.Generate, and I see no reason for the user to want to create a separate name for the network alone. The variable dates back to 4a08942 (steps: bootstrap / etcd / topology support for libvirtm 2018-04-24, coreos/tectonic-installer#3213), where it is not explicitly motivated.
This introduces basic libvirt support for the following steps: