Clouddriver 6.9.2
- provider/cf: fix stop server group state checking (#4661) (4727a643)
Note: This release requires Halyard version 1.32.0 or later.
This release includes fixes, features, and performance improvements across a wide feature set in Spinnaker. Here we share a summary of notable improvements, followed by the comprehensive changelog.
Spinnaker 1.20 no longer automatically adds a unique suffix to the name of jobs. Prior to this release, the Kubernetes V2 Run Job stage added a unique suffix to the name of the deployed job, with no ability to control or configure this behavior.
To continue having a random suffix added to the job name, set the metadata.generateName
field instead of metadata.name
, which causes the Kubernetes API to append a random suffix to the name.
If a job sets metadata.name
directly, that name will be used without modification for each execution of the job. As jobs are immutable, each execution of the stage will delete any existing job with the supplied name before creating the job.
This is the first release to enable the artifacts UI by default. Previously, the legacy and standard (“rewrite”) artifacts UIs were hidden behind the artifacts
and artifactsRewrite
feature flags. These flags are no longer supported in 1.20, and the standard artifacts UI is enabled by default.
The most noticeable difference for users previously using the legacy artifacts UI will be the absence of a separate Expected Artifacts section when configuring a pipeline. Expected artifacts are now defined inline against Triggers in the Artifacts Constraints section or inline in stages. For the 1.20 release only, a temporary legacyArtifactsEnabled
flag is available to. Add this flag to your settings-local.js
to revert to the legacy UI. See this RFC for more details.
Starting in Spinnaker 1.20, Kustomize support in Spinnaker is GA, and the Kustomize option in the Bake (Manifest) stage is enabled by default. Previously, the Kustomize option in the Bake (Manifest) stage was hidden behind a feature flag.
This release will be the last to include support for the legacy Kubernetes Provider (V1). Please migrate all Kubernetes accounts to the standard (V2) provider. Check out this RFC and blog post for more details.
In this release, all stages that are not provider-specific will be exposed by default. Operators can now choose to hide any stages from end-users using Deck’s hiddenStages
setting. The flags to enable the Gremlin, Travis, Wercker, and Create Load Balancers stages are no longer supported in Spinnaker 1.20 or Halyard 1.35. If using a version of Spinnaker prior to 1.20 with Halyard 1.35, you may still enable the travis
, wercker
, and infrastructureStages
flags by setting them in your settings-local.js
.
Spinnaker’s Docker containers are now hosted on us-docker.pkg.dev/spinnaker-community
. Previously, they were hosted on gcr.io/spinnaker-marketplace
.
In this release, users can now view the load balancers, listeners and target groups associated with their ECS services within the Load Balancers Infrastructure tab.
Also in this release, pipeline deployments now take into account container health checks and the overall task health status before determining a service to be up. If using a prior version, pipeline deployments would consider the service healthy once the tasks were started, but potentially prior to the container health checks running. This could lead to prematurely marking a deployment as successful.
localRepositoryDirectory
to Halyard default (35b8d966)iamRole
with BaseIAMRole
(35b8d966)artifactsRewrite
feature flag is enabled (51dba018)ExtensionPoint
that a plugin to expose endpoints under a common path (69fd0318)Execution.spelEvaluator
in Redis (6ad38fbb)