> ## Documentation Index
> Fetch the complete documentation index at: https://datum-4926dda5-docs-compute-and-vpc-guides.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Operations

> Scale, restart, roll out, and inspect individual Instances of a running workload with datumctl compute.

After you deploy a workload, use these commands for day-to-day operation: scaling, restarting Instances, watching a rollout, and inspecting a single Instance when something isn't right.

## Scaling

`datumctl compute scale` sets the minimum number of Instances per location for every placement of a workload:

```bash theme={null}
# Run three Instances in each location of the workload
datumctl compute scale api --min=3
```

The count is per location, from 1 to 1,000. Workloads don't scale automatically. For different counts per placement, edit `scaleSettings.minReplicas` in the manifest and apply it with `datumctl compute deploy -f`. For more information, see [Placement and scaling](/compute/placement-and-scaling).

## Restarting

`datumctl compute restart` starts a rolling restart. Datum replaces Instances without changing the image or configuration. A restart is useful to pick up an external change, such as a rotated Secret:

```bash theme={null}
datumctl compute restart api
```

`restart` replaces every Instance of the workload in every location.

## Watching a rollout

Any change that replaces Instances, such as a `deploy` with a new image or a `restart`, starts a rollout. `datumctl compute rollout` attaches to the rollout and prints progress for each placement and location:

```bash theme={null}
datumctl compute rollout api
```

Each row moves through phases — `Pending`, `Updating`, `Done` — until every placement finishes. Datum replaces Instances one at a time in each location, and waits for each replacement to be ready before it starts the next one. If a placement stalls for more than 30 seconds without progress, its phase becomes `Blocked` and the command prints the reason (a quota limit, a scheduling issue) reported by the platform.

<Note>
  Press Control+C to detach from the watch. The rollout continues in the background. To reattach and see current progress, run `datumctl compute rollout api`.
</Note>

## Inspecting Instances

`datumctl compute instances` lists every Instance in the project. Each row is one copy of a workload's container in one location, whether it's running or trying to start:

```bash theme={null}
# List every Instance in the project
datumctl compute instances

# List the Instances of one workload
datumctl compute instances --workload=api

# List the Instances in one location
datumctl compute instances --location=us-central-1
```

The status reflects availability, not only whether a process is running. For example, an Instance blocked on quota shows `Pending (quota exceeded)` instead of a generic "not ready" status. Add `-o wide` for the underlying instance type alongside the default columns, or `-o json`/`-o yaml` for the full resource.

To see the runtime configuration, environment, network addresses, and a plain-language explanation of any failure for one Instance, use `describe`:

```bash theme={null}
datumctl compute instances describe api-default-us-central-1-0
```

When an Instance can't start because of quota, `describe` also prints the next commands to run, such as `datumctl compute quota`.

## Next steps

* `datumctl compute scale --help`, `restart --help`, `rollout --help`, and `instances --help` for the full flag reference.
* [Deploying workloads](/datumctl/compute/deploying-workloads) — the `deploy` and `workloads` commands that create what you're operating on here.
* [Destroying workloads](/datumctl/compute/destroying-workloads) — remove a workload and every Instance it created.