confidential-containers/quickstart.md

Getting Started

This document contains an overview of Confidential Containers use cases and support as well as a guide for installing Confidential Containers, deploying workloads, and troubleshooting if things go wrong.

Use cases

Confidential Containers (CoCo) supports the following use cases:

• Running unencrypted containers without Confidential Computing (CC) hardware
• Running encrypted containers without CC HW (sample container images provided)
• Running unencrypted container images or sample encrypted images with CC HW
• Running your own encrypted container images with CC HW

The first two cases are mainly for testing and development or new users who want to explore the project. This guide explains all four cases below.

Hardware Support

Confidential Containers is tested with attestation on the following platforms:

• Intel TDX
• AMD SEV

Confidential Containers can also be run on Linux without any Confidential Computing hardware for testing or development.

The following platforms are untested or partially supported:

• AMD SEV-ES
• IBM Z SE

The following platforms are in development:

• Intel SGX
• AMD SEV-SNP

Limitations

Confidential Containers is still maturing. See release notes for currrent limitations.

Installing

You can enable Confidential Containers in an existing Kubernetes cluster using the Confidential Containers Operator.

• TBD: we will move the below sections to the operator documentation and only refer to that link Installing the operator *

Follow the steps described in https://github.com/confidential-containers/operator/blob/main/docs/INSTALL.md

Assuming the operator was installed successfully you can move on to creating a workload (the following section is optional).

Details on the CC operator installation

A few points to mention if your interested in the details:

Deploy the the operator:

kubectl apply -f https://raw.githubusercontent.com/confidential-containers/operator/main/deploy/deploy.yaml

You may get the following error when deploying the operator:

Error from server (Timeout): error when creating "https://raw.githubusercontent.com/confidential-containers/operator/main/deploy/deploy.yaml": Timeout: request did not complete within requested timeout - context deadline exceeded

This is a timeout on the kubectl side and simply run the command again which will solve the problem.

After you deployed the operator and before you create the custom resource run the following command and observer the expected output (STATUS is ready):

kubectl get pods -n confidential-containers-system

Output:

NAME                                              READY   STATUS    RESTARTS   AGE
cc-operator-controller-manager-5df7584679-kffzf   2/2     Running   0          4m35s

Deploying the operator vs a custom resource

The operator is responsible for creating the custom resource definition (CRD) which we can then use for creating a custom resource (CR).

In our case the operator has created the ccruntime CRD as can be observed in the following command:

kubectl get crd | grep ccruntime

Output:

ccruntimes.confidentialcontainers.org   2022-09-08T06:10:37Z

The following command provides the details on the CcRuntime CRD:

kubectl explain ccruntimes.confidentialcontainers.org

Output:

KIND:     CcRuntime
VERSION:  confidentialcontainers.org/v1beta1

DESCRIPTION:
     CcRuntime is the Schema for the ccruntimes API

FIELDS:
   apiVersion	<string>
     APIVersion defines the versioned schema of this representation of an
     object. Servers should convert recognized schemas to the latest internal
     value, and may reject unrecognized values. More info:
     https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources

   kind	<string>
     Kind is a string value representing the REST resource this object
     represents. Servers may infer this from the endpoint the client submits
     requests to. Cannot be updated. In CamelCase. More info:
     https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds

   metadata	<Object>
     Standard object's metadata. More info:
     https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata

   spec	<Object>
     CcRuntimeSpec defines the desired state of CcRuntime

   status	<Object>
     CcRuntimeStatus defines the observed state of CcRuntime

The complete CRD can be seen by running the following command:

kubectl explain --recursive=true ccruntimes.confidentialcontainers.org

You can also see the details of the CcRuntime CRD in the following .go file: https://github.com/confidential-containers/operator/blob/main/api/v1beta1/ccruntime_types.go#L90

Create the custom resource:

kubectl apply  -f https://raw.githubusercontent.com/confidential-containers/operator/main/config/samples/ccruntime.yaml

Check that the ccruntime was created successfully:

kubectl get ccruntimes

Output:

NAME               AGE
ccruntime-sample   5s

Use the following command to observe the details of the CR yaml::

kubectl get ccruntimes ccruntime-sample -o yaml | less

Note that we are using runtimeName: kataataame: kata

If we were use enclave-cc for example we would observe that runtimeName: enclave-cc

Once we also create the custom resource the validation will show us 2 additional pods created:

kubectl get pods -n confidential-containers-system

Output:

NAME                                              READY   STATUS    RESTARTS   AGE
cc-operator-controller-manager-5df7584679-kffzf   2/2     Running   0          21m
cc-operator-daemon-install-xz697                  1/1     Running   0          6m45s
cc-operator-pre-install-daemon-rtdls              1/1     Running   0          7m2s

Once the CR was created you will notice we have multiple runtime classes:

kubectl get runtimeclass

Output:

NAME            HANDLER         AGE
kata            kata            9m55s
kata-clh        kata-clh        9m55s
kata-clh-tdx    kata-clh-tdx    9m55s
kata-qemu       kata-qemu       9m55s
kata-qemu-tdx   kata-qemu-tdx   9m55s
kata-qemu-sev   kata-qemu-sev   9m55s

Details on each of the runtime classes:

-- kata - standard kata runtime using the QEMU hypervisor including all CoCo building blocks for a non CC HW -- kata-clh - standard kata runtime using the cloud hypervisor including all CoCo building blocks for a non CC HW -- kata-clh-tdx - using the Cloud Hypervisor, with TD-Shim, and support for Intel TDX CC HW -- kata-qemu - same as kata -- kata-qemu-tdx - using QEMU, with TDVF, and support for Intel TDX CC HW -- * TBD: we need to add the SEV runtimes as well *

Running a workload

Creating a sample CoCo workload

Once you've used the operator to install Confidential Containers, you can run a pod with CoCo by simply adding a runtime class. First, we will use the kata runtime class which uses CoCo wihout hardware support. Initially we will try this with an unencrypted container image.

In our example we will be using the bitnami/nginx image as described in the following yaml:

apiVersion: v1
kind: Pod
metadata:
  labels:
    run: nginx
  name: nginx
spec:
  containers:
  - image: bitnami/nginx:1.22.0
    name: nginx
  dnsPolicy: ClusterFirst
  runtimeClassName: kata

With Confidential Containers, the workload container images are never downloaded on the host. For verifying that the container image doesn’t exist on the host you should log into the k8s node and ensure the following command returns an empty result:

root@cluster01-master-0:/home/ubuntu# crictl  -r  unix:///run/containerd/containerd.sock image ls | grep bitnami/nginx

You will run this command again after the container has started.

Create a pod YAML file as previously described (we named it nginx.yaml) .

Create the workload:

kubectl apply -f nginx.yaml

Output:

pod/nginx created

Ensure the pod was created successfully (in running state):

kubectl get pods

Output:

NAME    READY   STATUS    RESTARTS   AGE
nginx   1/1     Running   0          3m50s

Now go back to the k8s node and ensure that you still don’t have any bitnami/nginx images on it:

root@cluster01-master-0:/home/ubuntu# crictl  -r  unix:///run/containerd/containerd.sock image ls | grep bitnami/nginx

Creating a CoCo workload using a pre-existing encrypted image

We will now proceed to download and run an encrypted container image using the CoCo building blocks.

• TBD: based on https://github.com/confidential-containers/operator/issues/77 *

Change the yaml to point to the sample container image The keys for this image should be in the rootfs

Creating a CoCo workload using a pre-existing encrypted image on CC HW

For running one of the sample workloads provided in the previous step, but now taking advantage of a specific TEE vendor, the user will have to set the runtime class of the workload accordingly in the workload yaml file.

TDX

In case the user wants to run the workload on a TDX capable hardware, using QEMU (which uses TDVF as its firmware) the kata-qemu-tdx runtime class must be specified. In case the user prefers using Cloud Hypervisor (which uses TD-Shim as its firmware) then the kata-clh-tdx runtime class must be specified.

SEV

kata-qemu-sev Export cert chain Start KBS (even for unencrypted)

Use the sample image or an unencrypted image

Building a new encrypted container image and deploying it as a CoCo workload

• TBD: instructions to build encrypted container image and other requirements (attestation, key etc) *

Use EAA KBC and Verdictd (TDX)

EAA is used to perform attestation at runtime and provide guest with confidential resources such as keys. It is based on rats-tls.

Verdictd is the Key Broker Service and Attestation Service of EAA. The EAA KBC is an optional module in the attestation-agent at compile time, which can be used to communicate with Verdictd. The communication is established on the encrypted channel provided by rats-tls.

EAA can now be used on intel TDX and intel SGX platforms.

Create encrypted image

Before build encrypted image, you need to make sure Skopeo and Verdictd(EAA KBS) have been installed:

• Skopeo: the command line utility to perform encryption operations.
• Verdictd: EAA Key Broker Service and Attestation Service.

1. Pull unencrypted image.

Here use alpine:latest for example:

${SKOPEO_HOME}/bin/skopeo copy --insecure-policy　docker://docker.io/library/alpine:latest oci:busybox

2. Follow the Verdictd README #Generate encrypted container image to encrypt the image.

3. Publish the encrypted image to your registry.

Deploy encrypted image

1. Build rootfs with EAA component:

Specify AA_KBC=eaa_kbc parameters when using kata-containers rootfs.sh scripts to create rootfs.

2. Launch Verdictd

Verdictd performs remote attestation at runtime and provides the key needed to decrypt the image. It is actually both Key Broker Service and Attestation Service of EAA. So when deploy the encrypted image, Verdictd is needed to be launched:

verdictd --listen <$ip>:<$port> --mutual

Note: The communication between Verdictd and EAA KBC is based on rats-tls, so you need to confirm that rats-tls has been correctly installed in your running environment.

3. Agent Configuration

Add configuration aa_kbc_params= 'eaa_kbc::<$IP>:<$PORT>' to agent config file, the IP and PORT should be consistent with verdictd.

Use offline SEV KBC and simple-kbs (SEV)

TODO: add instructions for simple-kbs with encrypted images

Trusted Ephemeral Storage for container images

With CoCo, container images are pulled inside the guest VM. By default container images are saved in guest memory which is protected by CC hardware. Since memory is an expensive resource, CoCo implemented trusted ephemeral storage for container image and RW layer.

This solution is verified with Kubernetes CSI driver open-local. Please follow this user guide to install open-local.

We can use following example trusted_store_cc.yaml to have a try:

apiVersion: v1
kind: Pod
metadata:
  name: trusted-lvm-block
spec:
  runtimeClassName: kata-qemu-tdx
  containers:
   - name: sidecar-trusted-store
     image: pause
     volumeDevices:
     - devicePath: "/dev/trusted_store"
       name: trusted-store
   - name: application
     image: busybox
     command:
     - sh
     - "-c"
     - |
         sleep 10000
  volumes:
   - name: trusted-store
     persistentVolumeClaim:
       claimName: trusted-store-block-pvc
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: trusted-store-block-pvc
spec:
  volumeMode: Block
  accessModes:
  - ReadWriteOnce
  resources:
    requests:
      storage: 1Gi
  storageClassName: open-local-lvm

Before deploy the workload, we can follow this documentation and use ccv0.sh to enable CoCo console debug(optional, check whether working as expected).

Create the workload:

kubectl apply -f trusted_store_cc.yaml

Ensure the pod was created successfully (in running state):

kubectl get pods

Output:

NAME                READY   STATUS    RESTARTS   AGE
trusted-lvm-block   2/2     Running   0          31s

After we enable the debug option, we can login into the VM with ccv0.sh script:

./ccv0.sh -d open_kata_shell

Check container image is saved in encrypted storage with following commands:

root@localhost:/# lsblk --fs
NAME                             FSTYPE LABEL UUID FSAVAIL FSUSE% MOUNTPOINT
sda
└─ephemeral_image_encrypted_disk                      906M     0% /run/image

root@localhost:/# cryptsetup status ephemeral_image_encrypted_disk
/dev/mapper/ephemeral_image_encrypted_disk is active and is in use.
  type:    LUKS2
  cipher:  aes-xts-plain64
  keysize: 512 bits
  key location: dm-crypt
  device:  /dev/sda
  sector size:  4096
  offset:  32768 sectors
  size:    2064384 sectors
  mode:    read/write

root@localhost:/# mount|grep image
/dev/mapper/ephemeral_image_encrypted_disk on /run/image type ext4 (rw,relatime)

root@localhost:/# ls /run/image/
layers  lost+found  overlay

Troubleshooting

Confidential Containers integrates several components. If you run into problems, it can sometimes be difficult to figure out what is going on or how to move forward. Here are some tips.

Kubernetes

To figure out which basic area you problem is in, first make sure that your Kubernetes cluster can schedule non-confidential workloads on your worker node. Remove the kata-* runtime class from your pod yaml and try to run a pod. If your pod still doesn't run, please refer to a more general Kubernetes troubleshooting guide.

If your cluster is healthy but you cannot start confidential containers, you might be able get some helpful information from Kubernetes. Try kubectl describe pod <your-pod> Sometimes this will give you a useful message pointing to a failed attestation or some sort of missing environment setup. Most of the time you will see a generic message such as the following:

Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create containerd task: failed to create shim: Failed to Check if grpc server is working: rpc error: code = DeadlineExceeded desc = timed out connecting to vsock 637456061:1024: unknown

Unfortunately this is a generic message. You'll need to go deeper to figure out what is going on.

CoCo Debugging

A good next step is to figure out if things are breaking before or after the VM boots. You can see if there is a hypervisor process running with something like this.

ps -ef | grep qemu

If you are using a different hypervisor, adjust command accordingly. If there are no hypervisor processes running on the worker node, the VM has either failed to start or was shutdown. If there is a hypervisor process, the problem is probably inside the guest.

Now is a good time to enable debug output for Kata and containerd. To do this, first look at the containerd config file located at /etc/containerd/config.toml. At the bottom of the file there should be a section for each runtime class. For example:

[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.kata-qemu-sev]
  cri_handler = "cc"
  runtime_type = "io.containerd.kata-qemu-sev.v2"
  privileged_without_host_devices = true
  pod_annotations = ["io.katacontainers.*"]
  [plugins."io.containerd.grpc.v1.cri".containerd.runtimes.kata-qemu-sev.options]
    ConfigPath = "/opt/confidential-containers/share/defaults/kata-containers/configuration-qemu-sev.toml"

The ConfigPath entry on the final line shows the path to the Kata configuration file that will be used for that runtime class.

While you are looking at the containerd config, find the [debug] section near the top and set level to debug. Make sure to restart containerd after editing the containerd config file. You can do this with sudo systemctl restart containerd.

Now go to the Kata config file that matches your runtime class and enable every debug option available. You do not need to restart any daemons when changing the Kata config file; just run another pod or hope that Kubernetes restarts your existing pod. Note that enabling debug options in the Kata config file can change the attestation evidence of a confidential guest.

Now you should be able to view logs from containerd with the following:

sudo journalctl -xeu containerd

Kata writes many messages to this log. It's good to know what you're looking for. There are many generic messages that are not significant, often arising from a VM not shutting down cleanly after an unrelated issue.

VM Doesn't Start

If the VM has failed to start, you might have a problem with confidential computing support on your worker node. Make sure that you can start confidential VMs without confidential containers.

Check the containerd log for any obvious errors regarding VM boot. Try searching the log for the string error or for the name of your hypervisor i.e. qemu or qemu-system-x86_64.

If there are no obvious errors, try finding the hypervisor commandline. This should be in the containerd log if you have enabled debug messages correctly.

It might be tempting to try running the hypervisor command directly from the command line, but this usually isn't productive. Instead, try starting a standalone VM using the same kernel, initrd/disk, command line, firmware, and hypervisor that Kata uses. This might uncover some kind of system misconfiguration. You can also find these values in the Kata config file, but looking in the log is more direct.

Another way to print the hypervisor command is to create a bash script that prints any arguments it is called with to a file. Then modify the Kata config file so that the hypervisor path points to this scipt rather than to the hypervisor. This method can also be used to add additional parameters to the command line. Just have the bash script call the hypervisor with whatever arguments it received plus any that you want to add. This could be useful for enabling debugging or tracing flags in your hypervisor. For instance, if you are using QEMU and SEV you might want to add the argument --trace 'kvm_sev_*'. Make sure that QEMU was built with an appropriate tracing backend.

VM Does Start

If the VM does start, search the containerd log for the string vmconsole. This will show you any guest serial output. You might see some errors coming from the kernel as the guest tries to boot. You might also see the Kata agent starting. If the Kata agent has started, you can match the output to the source to get some clues about what is happening. You might also see something more obvious, like a panic coming from the Kata agent.

Debug Console

One very useful deugging tool is the Kata guest debug console. You can enable this by editing the Kata agent configuration file and adding the lines

debug_console = true
debug_console_vport = 1026

Unfortunately, the agent config file is inside the guest rootfs. If you are using an initrd, you can update the config file by unpacking the initrd, changing the file, and then compressing it again. To rebuild the initrd, use Kata's initrd builder script:

kata-containers/tools/osbuilder/initrd-builder/initrd-builder.sh

Once you've started a pod with the new initrd, get the id of the pod you want to access. Do this via ps -ef | grep qemu or equivalent. The id is the long id that shows up in many different arguments. It should look like 1a9ab65be63b8b03dfd0c75036d27f0ed09eab38abb45337fea83acd3cd7bacd. Once you have the id, you can use it to access the debug console.

sudo /opt/confidential-containers/bin/kata-runtime exec <id>

You might need to symlink the appropriate Kata configuration file for your runtime class if the kata-runtime tries to look at the wrong one.

The debug console gives you access to the guest VM. This is a great way to investigate missing dependencies or incorrect configurations.

Guest Firmware Logs

If the VM is running but there is no guest output in the log, the guest might have stalled in the firmware. Firmware output will depend on your firmware and hypervisor. If you are using QEMU and OVMF, you can see the OVMF output by adding -global isa-debugcon.iobase=0x402 and -debugcon file:/tmp/ovmf.log to the QEMU command line using the redirect script described above.