Container Compute Service:Create an agent sandbox in an ACK cluster

In the navigation pane on the left, choose Workloads > custom resources. Select the CustomResourceDefinition tab, and then click Create from YAML.

Use the following YAML to create a SandboxSet resource. This registers an E2B template named code-interpreter with ack-sandbox-manager. The template is based on the official E2B code-interpreter image and is used to execute Python code.

apiVersion: agents.kruise.io/v1alpha1
kind: SandboxSet
metadata:
  name: code-interpreter
  namespace: default
spec:
  scaleStrategy:
    maxUnavailable: 500
  replicas: 10  # The size of the warm pool. We recommend setting this slightly larger than the anticipated burst of requests.
  template:
    # metadata:
      # annotations:
        # Optional. If you leave this empty, the security group configuration in acs-profile is used.
        # Ensure that the corresponding security group has enough available IP addresses, at least more than the number of replicas.
        # Insufficient IPs can slow down or even fail sandbox creation.
        # network.alibabacloud.com/security-group-ids: sg-8******
      # labels:
        # Optional. Used to schedule sandbox pods to ACS in an ACK cluster.
        # alibabacloud.com/acs: "true"
    spec:
      initContainers:
        # Declare agent-runtime as a native sidecar to automatically inject runtime components like envd into the sandbox container.
        - name: runtime
          image: registry-cn-hangzhou-vpc.ack.aliyuncs.com/acs/agent-runtime:v0.0.5
          command: [ "sh", "/workspace/entrypoint_inner.sh" ]
          volumeMounts:
            # Shared directory with the main container.
            - name: envd-volume
              mountPath: /mnt/envd
          env:
            - name: ENVD_DIR
              value: /mnt/envd
            # This environment variable allows the sidecar to share the resources of the main container without incurring extra costs.
            - name: __IGNORE_RESOURCE__
              value: "true"
          restartPolicy: Always
      containers:
      - name: sandbox
        # The official e2b code-interpreter image. It supports pulling from all regions and VPCs.
        image: registry-cn-hangzhou-vpc.ack.aliyuncs.com/acs/code-interpreter:v1.6
        imagePullPolicy: IfNotPresent
        # We recommend that you set resource requests. Otherwise, in an ACK environment, the sandbox may be assigned very small specifications, which can affect performance.
        resources:
          limits:
            cpu: 1
            memory: 1Gi
          requests:
            cpu: 1
            memory: 1Gi
        startupProbe:
          failureThreshold: 10
          httpGet:
            path: /health
            port: 49999
          initialDelaySeconds: 1
          periodSeconds: 2
          timeoutSeconds: 1
        env:
          # Specifies the location of the envd component injected by the runtime.
          - name: ENVD_DIR
            value: /mnt/envd
        volumeMounts:
          # Shared directory with the runtime.
          - name: envd-volume
            mountPath: /mnt/envd
        # Starts the envd service by using a postStart hook.
        lifecycle:
          postStart:
            exec:
              command: [ "/bin/bash", "-c", "/mnt/envd/envd-run.sh" ]
      # Ensures quick container termination to increase the probability of reuse.
      terminationGracePeriodSeconds: 1
      volumes:
        - name: envd-volume
          emptyDir: { }

In the navigation pane on the left, click Pods. Select the default namespace to view the created and warmed-up code-interpreter sandbox pods.

Install Git and Python in your local environment.

Install the E2B Python SDK.

pip install e2b-code-interpreter==2.4.1

Configure environment variables.

# Use the default domain from the ack-sandbox-manager component installation. Do not include "*". Modify as needed.
export E2B_DOMAIN=your.domain.com
# Use the default API Key from the ack-sandbox-manager component installation. Modify as needed.
export E2B_API_KEY=admin-987654321

Run the git clone https://github.com/openkruise/agents.git command to download the customized ACK SDK to your local machine.

Navigate to the /agents/sdk/customized_e2b directory of the customized SDK and save the following code as a main.py file.

# Import the E2B SDK
from e2b_code_interpreter import Sandbox
from kruise_agents.patch_e2b import patch_e2b

patch_e2b(https=False)  # patch sdk

sbx: Sandbox = Sandbox.create(template="code-interpreter")
print(f"sandbox id: {sbx.sandbox_id}")
result = sbx.run_code("print('hello, world')")
print(f"run code result: {result}")
text = input("enter some text to be saved to file 'text.txt' inside sandbox:  ")
sbx.files.write("text.txt", text)
print(f"read file from sandbox via files api: [{sbx.files.read('text.txt')}]")
print(f"read file from sandbox via commands api: [{sbx.commands.run('cat text.txt')}]")
input("press ENTER to kill the sandbox")
print(sbx.kill())

Run the main.py file to create and verify the sandbox environment.

After the prompt appears for the first time, enter text such as acs agent sandbox and press Enter. This action writes acs agent sandbox to the /home/user/text.txt file in the Pod named code-interpreter-29***. If you press Enter again, the current sandbox environment is deleted.

python main.py

Expected output:

sandbox id: default--code-interpreter-29***
run code result: Execution(Results: [], Logs: Logs(stdout: ['hello, world\n'], stderr: []), Error: None)
enter some text to be saved to file 'text.txt' inside sandbox:  ack agent sandbox
read file from sandbox via files api: [ack agent sandbox]
read file from sandbox via commands api: [CommandResult(stderr='', stdout='ack agent sandbox', exit_code=0, error='')]
press ENTER to kill the sandbox
True