A pod (elastic container instance) passes through several statuses during its lifecycle. Understanding these statuses helps you design business logic such as polling pod state, handling failures, and calculating billing periods.
Pod statuses
The following table lists all pod statuses in ECI, their Kubernetes equivalents, and whether billing applies.
| ECI status | Description | Kubernetes status | Billed |
|---|---|---|---|
| Scheduling | ECI is allocating compute resources for the pod. | Pending | No |
| Pending | Resources are allocated. One or more containers are starting; no containers are running yet. | Pending | Yes |
| Running | All containers are created, and at least one container is running or restarting. | Running | Yes |
| Restarting | The pod is restarting. | Pending | Yes |
| Updating | The pod is being updated. | Pending | Yes |
| Terminating | The pod is being terminated. When a running pod configured with preStop is being deleted, the pod enters the Terminating state. After the preStop hook runs, the pod is deleted. | Running | Yes |
| Succeeded | All containers exited successfully and will not restart. | Succeeded | No |
| Failed | All containers exited. At least one container terminated due to a runtime error — either by exiting with a non-zero code or being terminated by the system. | Failed | No |
| Expired | The pod was a preemptible instance and was reclaimed due to resource expiration. | Failed | No |
| ScheduleFailed | The pod failed to be created. ECI automatically deletes the pod after 24 hours, and you are not charged for the pod. | Failed | No |
The restart policy applies only to containers within a pod. The pod (elastic container instance) itself does not automatically restart.
The following diagram shows how a pod transitions between statuses during its lifecycle.
When a pod is terminated, its underlying compute resources are recycled. By default, associated resources — such as elastic IP addresses (EIPs) — are released at the same time.
Metadata for pods in a terminal status (Succeeded, Failed, or Expired) is retained as follows:
All metadata is retained for the first hour after the pod reaches a terminal status.
After one hour, only the latest 100 metadata entries per region are kept.
Container statuses
Each container within a pod has its own status, independent of the pod status.
| Status | Description |
|---|---|
| Waiting | The container is waiting to start. Application containers remain in Waiting while an InitContainer is running, until the InitContainer exits. |
| Running | The container is running. |
| Terminated | The container has exited, either successfully or due to a failure. |