-
Notifications
You must be signed in to change notification settings - Fork 348
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
Support e2e runs despite custom taints on nodes #599
Comments
For clarity; putting this on the milestone for v0.14.2 as it is being worked on. The feature won't be supported (even if it merges) until the next k8s release (I don't think this will get cherry picked; I could be wrong). So it will be a v0.15.0 feature eventually. |
As I was working on this I actually ran into another dev that had a WIP PR to solve the same issue: kubernetes/kubernetes#76654 His approach allows the user to provide label selectors for nodes for which being non-schedulable is not an error condition in the tests. |
Moved to next milestone for tracking. |
Slid in priority and missed the v1.15 k8s code freeze; will be shooting for v1.16 and more recent PR variation is here: kubernetes/kubernetes#78500 |
Won't this be included as part of the |
@hectorj2f this is an upstream k8s issue, not just a Sonobuoy issue so we are limited by their release/freeze schedule. kubernetes/kubernetes#81043 just merged, fixing the issue (allowing you to work around it I should say) but will be in v1.17 of k8s. So until then you just have to either untaint the nodes or set the flag for allowing more not-ready nodes. (I think you can get context on those two approaches by looking at the original k8s ticket for this). |
Alright @johnSchnake. V1.17 sounds like too far in time. |
Upstream tracking issue for the ability to run e2e tests despite having custom taints on nodes.
kubernetes/kubernetes#74282
Would resolve things like #571 and #507
The text was updated successfully, but these errors were encountered: