Changed
- Use non-exp apiVersion for azure machine pool types in
template nodepool
.
template nodepool
.spec.config.configMap
in app/<cluster-name>-default-apps
for CAPZ
clusters.--proxy
and --proxy-port
flags to login cmd
to enable proxy-url: socks5://localhost:9000
in the cluster section of the configuration added to kubeconfigclientcert
Workload Clustersv1alpha3
version when using CAPI CRDs.kubectl gs template cluster
now by default creates a node pool with the name nodepool0
, instead of machine-pool0
, as the latter is no longer valid according to the cluster-aws schema as of v0.24.0.cluster template
supports a generic way to template CAPI
based clusters where all the input parameters are given as values.yaml
from the corresponding cluster
and default-apps
chart.CAPZ
implementation and switched to the generic templating implementation.kubectl gs template cluster
for Cluster API provider AWS has been adapted to work with the values schema of cluster-aws v0.28.0.CAPZ
based clusters--aws-prefix-list-id
and --aws-transit-gateway-id
for private clusters--role
to --aws-cluster-role-identity-name
and adapted manifest output to the new name awsClusterRoleIdentityName
(see cluster-aws change)As part of our automatic upgrades journey, we have learnt that cluster chart should structure in such a way that allows overwriting all sections in different levels
Warning: This results in a breaking change in the template output of a capa
clusters machine pools.
capa
machine pools to template usings an object instead of arrays as of cluster-aws v0.24.0
capvcd
installations.gitops add base
from bases/clusters/<PROVIDER>
to bases/clusters/<PROVIDER>/template
gitops add base
generated templates to capa
and gcp
(CAPG) providers with value: machine-pool0