> ## Documentation Index
> Fetch the complete documentation index at: https://dpos.xdr.ooo/llms.txt
> Use this file to discover all available pages before exploring further.

# DL/DA Clusters

> Configure worker lists once, stage every node, and run final bringup from DL and DA masters.

# DL/DA Clusters

The Mirror Manager stores the cluster worker lists in Configuration and uses them when generating the final DP commands.

## Configure workers

In Menu 1:

* **DL Worker IP addresses**: DL workers only; do not include the DL master
* **DA Worker IP addresses**: DA workers only; do not include the DA master
* **Worker SSH Password (aella)**: one common password used by each master to connect to its own workers

## Common staging first

For clustered deployments, run the common Menu 7 steps on every target node:

```text theme={null}
DL master
DL worker(s)
DA master
DA worker(s)
```

Complete the common staging/prerequisite phase on **all nodes** before starting final bringup.

## Final bringup

Run the role-specific final command only from the masters:

```text theme={null}
DL master  ──> orchestrates configured DL workers
DA master  ──> orchestrates configured DA workers
```

Do not run the master-orchestration command manually on workers.

The generated command uses the worker configuration saved in Mirror Manager. Use Menu 7 output rather than rebuilding `--worker-ips` or password flags by hand.

## Recommended validation order

1. finish the DL master/worker bringup and confirm DL readiness
2. run the DA master/worker bringup
3. confirm all expected Kubernetes nodes and services are ready
4. validate application, NTP, DNS, and cluster connectivity

The exact release-specific order printed by Menu 7 remains authoritative when it is more specific than this overview.