baremetal: Bootstrap keepalived should pass --cluster-config #1037
Assignees
Comments
|
/assign |
bcrochet
added a commit
to bcrochet/machine-config-operator
that referenced
this issue
Aug 7, 2019
A recent installer change started using a loopback address in kubeconfig as the default on the bootstrap node. baremetal-runtimecfg will begin using cluster-config.yml if it exists first, in order to get the domain and name of the cluster. This patch adds a flag for this to the keepalived bootstrap pod in order to utilize it. This should only affect the bootstrap pods. Fixes openshift#1037 Depends on openshift/baremetal-runtimecfg#16
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
A recent change to the installer (openshift/installer@82d81d9) copies kubeconfig-loopback instead of kubeconfig to /etc/kubernetes/kubeconfig.
This causes a problem for baremetal bootstrap. It points to localhost.
Proposed fix is to read from cluster-config.yaml first, and fall back to kubeconfig if that's not available.
The coredns bootstrap pod already passes --cluster-config to baremetal-runtimecfg. Keepalived needs to as well.
Related to openshift/baremetal-runtimecfg#15
Depends on openshift/baremetal-runtimecfg#16
The text was updated successfully, but these errors were encountered: