VM Requisites

Resource	Minimum
CPU	2 vCPU
RAM	4 GB
Disk	40 GB
Operating System	Ubuntu 22.04 / 24.04

Access Port

Port	Protocol	Usage
443	HTTPS	Administration dashboard and API gateway

How To Access

https://<IP-CON-GUIONES>.sslip.io/

‍

Example: if your VM has the IP address 200.25.101.125:

https://200-25-101-125.sslip.io/

‍

Log in with the Gateway Token configured in OPENCLAW_GATEWAY_TOKEN or find it in the deployment outputs in the Cuemby portal.

Verify that OpenClaw is active

# View the OpenClaw pod
`kubectl get pods -A | grep openclaw`

# View the installation log
`tail -f /var/log/cuemby/bootstrap.log`

‍

Expected output:

NAME                        READY   STATUS    RESTARTS
openclaw-xxxxxxxxx-xxxxx    1/1     Running   0        ← Running ✓

Configuration Parameters

LLM provider | Configure at least one

Parameter	Description
OPENCLAW_ANTHROPIC_API_KEY	Anthropic API key for using Claude models. Get yours at console.anthropic.com.
OPENCLAW_OPENAI_API_KEY	OpenAI API Key for using GPT models. Get yours at platform.openai.com.

‍

⚠️ You must configure at least one API key. Without it, OpenClaw will not be able to process messages.

‍

Gateway

Parameter	Default	Description
OPENCLAW_GATEWAY_TOKEN ⚠️	auto-generated	Access token for the dashboard and API. It is recommended to manually set it to retain it between redeployments.

‍

Messaging channels (optional)

They can be configured during deployment or later from the dashboard.

‍

Parameter	Description
OPENCLAW_TELEGRAM_TOKEN	Telegram bot token. Get it from @BotFather.
OPENCLAW_DISCORD_TOKEN	Discord bot token. Configure it in the Discord Developer Portal.
OPENCLAW_SLACK_BOT_TOKEN	Slack bot xoxb-... token.
OPENCLAW_SLACK_APP_TOKEN	Token xapp-... of the Slack app (for Socket Mode).

‍

First Steps (quick start)

Open the dashboard at your URL and authenticate with the Gateway Token.
Verify that the configured LLM provider appears as active.
Connect a channel from the dashboard (Telegram, Discord, or Slack) if you didn't configure it during deployment.
Send a test message from the connected channel to verify that the entire flow is working.

💡 Tip: Set the OPENCLAW_GATEWAY_TOKEN value before deploying. If you leave it on auto-generated and need to redeploy, the token will change, and you'll have to update all integrations that use it.

Quick Troubleshooting

Problem	Probable cause	Solution
Dashboard not loading	Pod still starting	Wait ~3 min and check tail -f /var/log/cuemby/bootstrap.log.
Bot is not responding on the channel	API key not configured	Verify that ANTHROPIC_API_KEY or OPENAI_API_KEY is correctly configured in the deployment.
Unauthorized access	Incorrect Gateway Token	Review the deployment outputs in the Cuemby portal.
Channel not connecting	Invalid or expired channel token	Regenerate the token from the corresponding platform (BotFather, Discord Portal, Slack).
Pod in CrashLoopBackOff	Invalid API key or insufficient resources	Check with kubectl logs -n .

Cuemby Cloud

IaaS

About Cuemby Cloud

Cuemby Cloud is enterprise-grade cloud infrastructure managed from a single console, built for stronger security, predictable operations, and easy scaling across regions.

Datacenters regions available in Colombia, Ecuador, and Chile

Backed by Tier III / Tier IV data center locations

Zero Network Ingress and Egress Fees

24/7 local expert support

Get Started for Free

Discover Cuemby Cloud

OpenClaw