85b44f7f57
* ci: make waiting for nodes more robust After initializing the cluster, a lot of things happen in parallel and are potentially getting in each others' way: nodes are joining, daemonsets are proliferating, the network is being set up. During this period, it's not unusual that the Kubernetes API server is unavailable for a short time, e.g. due to etcd loosing quorum or load balancing changes. This period of instability has the potential to affect all kubectl commands negatively, leading to problems especially for tests, where command failures often lead to test failures. On the other hand, we'd expect everything to be quite stable after the initial dust settles. Therefore, this commit changes how we wait after initializing a cluster. Until we have a reasonable expectation of readiness, we ignore command failures and wait for things to stabilize. The cluster is considered stable once all configured nodes and all API servers report ready. |
||
|---|---|---|
| .github | ||
| 3rdparty | ||
| bazel | ||
| bootstrapper | ||
| cli | ||
| csi | ||
| debugd | ||
| dev-docs | ||
| disk-mapper | ||
| docs | ||
| e2e | ||
| hack | ||
| image | ||
| internal | ||
| joinservice | ||
| keyservice | ||
| measurement-reader | ||
| nix | ||
| operators/constellation-node-operator | ||
| rfc | ||
| s3proxy | ||
| terraform | ||
| terraform-provider-constellation | ||
| tools | ||
| upgrade-agent | ||
| verify | ||
| .bazelignore | ||
| .bazelrc | ||
| .bazelversion | ||
| .dockerignore | ||
| .envrc | ||
| .gitignore | ||
| .golangci.yml | ||
| .grype.yaml | ||
| .lychee.toml | ||
| .shellcheckrc | ||
| .vale.ini | ||
| BUILD.bazel | ||
| CODE_OF_CONDUCT.md | ||
| CODEOWNERS | ||
| CONTRIBUTING.md | ||
| flake.lock | ||
| flake.nix | ||
| go.mod | ||
| go.sum | ||
| go.work | ||
| LICENSE | ||
| netlify.toml | ||
| README.md | ||
| renovate.json5 | ||
| SECURITY.md | ||
| version.txt | ||
| WORKSPACE.bazel | ||
Always Encrypted Kubernetes
Constellation is a Kubernetes engine that aims to provide the best possible data security. It wraps your K8s cluster into a single confidential context that is shielded from the underlying cloud infrastructure. Everything inside is always encrypted, including at runtime in memory. For this, Constellation leverages confidential computing (see the whitepaper) and more specifically Confidential VMs.
Goals
From a security perspective, Constellation is designed to keep all data always encrypted and to prevent access from the infrastructure layer (i.e., remove the infrastructure from the TCB). This includes access from datacenter employees, privileged cloud admins, and attackers coming through the infrastructure (e.g., malicious co-tenants escalating their privileges).
From a DevOps perspective, Constellation is designed to work just like what you would expect from a modern K8s engine.
Use cases
Encrypting your K8s is good for:
- Increasing the overall security of your clusters
- Increasing the trustworthiness of your SaaS offerings
- Moving sensitive workloads from on-prem to the cloud
- Meeting regulatory requirements
Features
🔒 Everything always encrypted
- Runtime encryption: All nodes run inside AMD SEV-based Confidential VMs (CVMs). Support for Intel TDX will be added in the future.
- Transparent encryption of network: All pod-to-pod traffic is automatically encrypted
- Transparent encryption of storage: All writes to persistent storage are automatically encrypted. This includes nodes' state disks, persistent volumes via CSI, and S3 object storage.
- Transparent key management: All cryptographic keys are managed within the confidential context
🔍 Everything verifiable
- "Whole cluster" attestation based on the remote-attestation feature of CVMs
- Confidential computing-optimized node images; fully measured and integrity-protected
- Supply chain protection with sigstore and SLSA Level 3.
🚀 Performance and scale
- High availability with multi-master architecture and stacked etcd topology
- Dynamic cluster autoscaling with verification and secure bootstrapping of new nodes
- Competitive performance (see K-Bench comparison with AKS and GKE)
🧩 Easy to use and integrate
- Constellation is a CNCF-certified Kubernetes. It's aligned to Kubernetes' version support policy and will likely work with your existing workloads and tools.
- Support for Azure, GCP, and AWS.
- Support for local installations with MiniConstellation.
- Support for Terraform
Getting started
If you're already familiar with Kubernetes, it's easy to get started with Constellation:
- 📦 Install the CLI or use the Terraform provider
- ⌨️ Create a Constellation cluster in the cloud or locally
- 🏎️ Run your app
Learn more: "Getting started with Constellation" videos series.
Live demos
We're running public instances of popular software on Constellation:
- Rocket.Chat: https://rocket.edgeless.systems/ (blog post)
- GitLab: https://gitlab.edgeless.systems/ (blog post)
These instances run on CVMs in Azure and Constellation keeps them end-to-end confidential.
Documentation
To learn more, see the documentation. You may want to start with one of the following sections.
- Confidential Kubernetes (Constellation vs. AKS/GKE + CVMs)
- Security benefits
- Architecture
Support
- If something doesn't work, make sure to use the latest release and check out the known issues.
- Please file an issue to get help or report a bug.
- Join the Discord to have a chat on confidential computing and Constellation.
- Visit our blog for technical deep-dives and tutorials and follow us on LinkedIn for news.
- Edgeless Systems also offers Enterprise Support.
Contributing
Refer to CONTRIBUTING.md on how to contribute. The most important points:
- Pull requests are welcome! You need to agree to our Contributor License Agreement.
- Please follow the Code of Conduct.
Warning
Please report any security issue via a private GitHub vulnerability report or write to security@edgeless.systems.
License
The Constellation source code is licensed under the GNU Affero General Public License v3.0. Edgeless Systems provides pre-built and signed binaries and images for Constellation. You may use these free of charge to create and run services for internal consumption, evaluation purposes, or non-commercial use. You can find more information in the license section of the docs.