Fleet Management
Fixed
- Use capz
v0.4.x on Azure CRDs so that we have the experimental CRDs on their old api group.
Changed
- CRDs ownership for Phoenix
Fixed
- Fix a problem with fetching Catalog CRs in
validate apps. - Fixing a problem where the function to fetch the SSH secret to generate the templates was using
inCluster config ignoring the kubeconfig.
Added
- Add metadata about additional vSphere CRDs for documentation.
Fixed
- Adjust name of Cluster API certificate and service for
v1alpha4. The annotation cert-manager.io/inject-ca-from value changes from giantswarm/cluster-api-core-webhook to giantswarm/cluster-api-core-cert and the service name from cluster-api-core-webhook to cluster-api-core.
Changed
KVMClusterConfig: make worker node labels optional.- Rename provider
VMWare to vSphere. - Split Cluster API core CRDs by provider, to enable independent versioning (e.g.
v1alpha2 for AWS and v1alpha4 for vSphere). - Configure webhook patch for the
ClusterClass CRD. - Update Cluster API core CRDs to
v0.4.4 for improved defaulting and printer columns. - In the
AppCatalogEntry CRD, rename the printer column APP VERSION to UPSTREAM VERSION and switch the order of VERSION and UPSTREAM VERSION. This affects the output of kubectl get appcatalogentries. - Update repository to use go v1.17.
- Remove reference to deprecated
AppCatalog CRD from the AppCatalogEntry CRD.
Added
- Add templating for EKS clusters using the management cluster API
- Add templating for EKS node pools using the management cluster API
- Add templating for CAPA node pools using the management cluster API
Changed
- In the
get catalogs command output, rename the colum APP VERSION to UPSTREAM VERSION and change the column order.
Fixed
- Fix a problem where the template subcommands would be slower than expected because of obsolete API requests.
Added
- Add CRs to create a bastion host in CAPZ cluster template.
- Enable termination events for CAPZ node pools.
Changed
- Use org-namespace for AWS Clusters by default
Added
- Nodepool nodes are labeled with nodepool id on AWS using
giantswarm.io/machine-pool. MachinePool and AzureMachinePool are labeled with the giantswarm.io/machine-pool label.get releases command to return details of available releases.
Added
- Nodepool nodes are labeled with nodepool id on Azure using
giantswarm.io/machine-pool.
Changed
- Update the
template cluster command to add CAPI defaults and validation using the management cluster API.