-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
vSphere infrastructure provider template #25
Conversation
nodes are tainted with node.cloudprovider.kubernetes.io/uninitialized because we set cloudProvider: external. this prevents some crucial pods from being scheduled (for example we need the kube-vip pod to start on the first node so it can announce the control plane IP)
026df59
to
fc98d1b
Compare
set insecure-flag="true" to get around a possible thumbprint verification bug in vsphere-csi driver. Also update csi driver versions to v3.3.0 which support k8s v1.28-v1.30.
create manifest at /capi/manifests so it's applied after bootstrap
The provider acts weird when it comes to checking the thumbprint. The provider suggests to use sha256 for the thumbprint, but kubectl explain VCluster says that thumbprint expects a sha1 hash. So it's clear there's some conflicting information in the documentation. Some experimenting needs to be done in regards to removing the insecure-flag and using sha256 or sha1 for the thumbprint. The error can be seen in vsphere-cpi pods, the logs complain about "thumbprint does not match". Right now there is no error because of a combination of insecure-flag: true and using sha1. Not sure which one of these solves the issue, but it's solved for now.
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.
Immaculate work getting it to work! Let's clean up the rough edges!
This reverts commit b4e0279.
This reverts commit b3ce6ea281cc13a65dca4431a6e5a97a2355278b.
use kube-vip rbac manifests instead of mounting the admin.conf file
the secret can be enabled/disabled by changing VSPHERE_PROXY_DISABLE in clusterctl.yaml
58b1ee7
to
2dd8c9d
Compare
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.
Looks great, did another pass
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.
LGTM, waiting for @neoaggelos final approval
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.
LGTM
Gaps:
insecure-flag: true
issue and sha1/sha256 thumbprint needs to be investigated. TLDR, there's conflicting documentation where the provider says the thumbprint needs to be in sha256 format, but providing the thumbprint in sha256 causes a "thumbprint mismatch" error. I have resolved this by using a sha1 thumbprint and settinginsecure-flag: true
. Need to try only settinginsecure-flag: true
with sha256, or removinginsecure-flag: true
and use sha1.NOTE Do not merge before testing with multi-node cluster.
NOTE Verify if L297-301, L319-323, L293-295 are needed for multi-node cluster before merging.