Container instances

Run prebuilt container images as a managed service — one or more components, each its own pod, reachable from outside only through the endpoints you add.

• API reference — Every HTTP endpoint of the containerinstances module: authentication, IAM action, parameters, bodies and status codes.
• Browse files on a volume — List, preview, download, upload, rename and delete files on a container instance's persistent volumes from the Files tab.
• Manage components — Read a component's state on the Components tab, change its containers, add components, re-deploy one component, and start, stop or restart the whole instance.
• Configuration reference — Configuration keys of the containerinstances module: sections, environment variables, types and defaults.
• Create a container instance — Create a container instance in the portal — name and placement, image and registry, environment and command, compute plan, ports and public exposure — or with the CLI.
• Delete a container instance — Delete a container instance, and what is removed with it — its identity, its endpoints, the vault of a blueprint instance — and what stays, such as kept persistent volumes.
• Endpoints — Make a container instance reachable from outside the cluster — HTTPS on a hostname with its own certificate, or a load balancer address — and put the platform sign-in in front of it.
• Container instances — Run prebuilt container images as a managed service — one or more components, each its own pod, reachable from outside only through the endpoints you add.
• View logs and open a terminal — Read a component's container logs, and open a shell inside one of its pods from the Components tab — and what a terminal session can reach.
• Names and limits — Naming rules for container instances and their parts, the limits that apply to them, and the compute plans.
• Permissions — The actions that govern container instances, the roles that hold them, and what else a task needs.
• Compare and roll back revisions — See every saved change to a container instance's configuration, compare a revision with the one before it, and roll the instance back to an earlier one.
• Scale a component — Set how many replicas a component runs, let it autoscale on CPU, or give it a bigger compute plan.
• Back up and restore — Take snapshots of a container instance's persistent volumes, schedule them with a backup policy and consistency hooks, and restore one in place.
• Statuses — What the statuses of a container instance, its components, its pods and its endpoints mean, and the messages a failing component shows.
• Terminal sessions — Where the portal's terminal sessions into container instance pods run, the limits that apply to them, and what the platform records about them today.
• Add volumes and files — Give a component a persistent disk, scratch space or configuration files, and mount them into its containers, through the CLI or the API.