-
Notifications
You must be signed in to change notification settings - Fork 175
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
Cloud controller is unable to patch node with IP addresses from cloud provider #266
Comments
/assign @andrewsykim @dvonthenen |
@frapposelli: GitHub didn't allow me to assign the following users: vonthenend. Note that only kubernetes members, repo collaborators and people who have commented on this issue/PR can be assigned. Additionally, issues/PRs can only have 10 assignees at the same time. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
/assign |
This is the upstream fix for this bug kubernetes/kubernetes#79391, should be fixed when we pull that in. |
It looks like that was merged and possibly made it into 1.16. Any confirmation on that? |
yeah its in v1.16 |
Since I need to rebuild my dev env, will be trying out 1.16 today or tomorrow. Does the CPI need to pull in 1.16 for this to work? or just on the k8s side? |
It seems like just k8s.io/kubernetes needs to be updated |
Looks like it's only on the k8s side. I tested with the latest CPI binary without updating the k8s.io/kubernetes that the CPI is built on and it works just fine with 1.16.2. Closing this out. /close |
@dvonthenen: Closing this issue. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
@dvonthenen you used multiple vNICs in your tests? At first glance it sounds like we do need to update k8s.io/kubernetes in the controller manager since PatchNodeStatus is still called by the cloud node controllerl |
ahhhh i missed that condition... will re-test right now |
/reopen |
@andrewsykim: Reopened this issue. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
Yep, I re-tested with 3 NICs with 3 different IPs one on each NIC and I don't see that error. Works with the latest CPI binary without updating the k8s.io/kubernetes that the CPI is built on. Keep in mind, we also made the change that only the first IP is patched remember. That change isn't in the 1.0.0 release but only on master. Also keep in mind that this PR #271 is trying to re-introduce multiple IPs. EDIT: I should say selecting a single IP from multiple ones available. |
So sounds like this is already fixed in master but we're going to allow users to configure which networks the VM addresses should come from in #271? So a VM can have multiple NICs, but there can only be 1 InternalIP and 1 ExternalIP on the Node, both of which will be configurable after #271. Is that correct @dvonthenen? |
@andrewsykim that is correct |
@dvonthenen can you verify and close? |
Same as #270 |
closed via #284 /close |
@frapposelli: Closing this issue. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
hit the same issue for pure IPv6, related ticket #312 |
When using vSphere VM Network names, the error disappeared. |
FYI we also need to backport kubernetes/kubernetes#79391 |
Hit this issue with k8s 1.18 + CCM 1.1.0 |
Is this a BUG REPORT or FEATURE REQUEST?:
/kind bug
What happened:
Cloud controller is unable to patch node with IP addresses from cloud provider on a multi-NIC VM.
What you expected to happen:
Nodes are patched with IP addresses from cloud provider.
How to reproduce it (as minimally and precisely as possible):
Build a worker node with 3 NICs and attempt to deploy the cloud controller
Anything else we need to know?:
This was encountered in the openstack provider as well and was fixed there (kubernetes/cloud-provider-openstack#407).
Environment:
uname -a
): 4.14.138-rancherThe text was updated successfully, but these errors were encountered: