> For the complete documentation index, see [llms.txt](https://v2.dataos.info/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://v2.dataos.info/references/resources/service/core-concepts.md).

# Core concepts

This section includes core concepts specific to a service resource

### Ports

`servicePorts` declares the ports exposed by the Service. Each port requires a `name` and a numeric `port`.

```yaml
spec:
  servicePorts:
    - name: postgres
      port: 5432
```

For TCP services, consumers can resolve the Service endpoint from the Service runtime metadata. For HTTP services, configure the container to listen on the declared port and add any ingress configuration supported by the runtime.

### Replicas

`replicas` controls how many identical Service instances are maintained. Use more than one replica for stateless services that can handle traffic concurrently. Stateful services, such as a single Postgres instance, commonly run with `replicas: 1`.

```yaml
spec:
  replicas: 1
```

### High availability

`highAvailabilityConfig` defines how multiple replicas of the Worker are distributed across infrastructure boundaries such as hostnames, zones, or regions. This ensures that if one replica becomes unavailable due to an issue in a specific part of the infrastructure, the other replicas continue to operate independently.

```yaml
spec:
  compute: karpenter-compute
  highAvailabilityConfig:
    mode: preferred
    # level: hostname
    level: zone
    # level: region
  resources:
    requests:
      cpu: 100m
      memory: 128Mi
    limits:
      cpu: 1000m
      memory: 1024Mi
  executionMode: runnable
```

* `level` specifies the level across which replicas should be distributed, `hostname`, `zone`, or `region` , determining the intended placement separation. The actual distribution still depends on `mode`.
* `mode` specifies how strictly that separation is applied. With `mode: preferred`, the system tries to place each replica on a different hostname, zone, or region as specified by `level`, but if not enough distinct units are available, it still proceeds and places replicas together. With `mode: required`, replicas are placed only if the separation defined by `level` can be exactly met — if not enough distinct hostnames, zones, or regions are available, the Worker fails.

### Stateful Services

Set `stateful: true` when the Service needs stable identity or attached storage.

```yaml
spec:
  stateful: true
```

Stateful Services can use either inline disk storage or an existing Volume Resource:

* Use `disk` when storage belongs to the Service lifecycle and can be deleted with the Service.
* Use `use.volumes` when the data must persist independently of the Service.

See also: [Provision inline storage](https://v2.dataos.info/references/resources/service.md) and [Persist storage across workloads](https://v2.dataos.info/references/resources/service.md).

### Stack and stackSpec

`stack` selects the runtime stack. `stackSpec` contains stack-specific configuration. For the `container` stack, provide the container image, command, and arguments.

```yaml
spec:
  stack: container
  stackSpec:
    image: docker.io/library/postgres:18-alpine
    command:
      - docker-entrypoint.sh
    arguments:
      - postgres
```

### Projections

Use `use.projection` to project Secrets, environment variables, and files into the Service runtime. The following example reads `POSTGRES_PASSWORD` from a Secret and exposes it as an environment variable.

For projection syntax, template engines, data sources, and troubleshooting, see [Projections](https://v2.dataos.info/references/resources/projections.md).

```yaml
spec:
  use:
    projection:
      templateType: liquid
      secrets:
        - id: ${{tenant-id}}:${{secret-name}}
          contextAlias: pgSecret
      projections:
        envVars:
          - key: POSTGRES_PASSWORD
            template: "{{ secrets['pgSecret'].POSTGRES_PASSWORD }}"
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://v2.dataos.info/references/resources/service/core-concepts.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
