- RECOVERY
- CLEANUP
- LOGS
- HUMAN
Short answer
How do I restart a hung service using my AI?
Your AI triggers a pod restart via the Porter API to clear memory leaks or frozen processes. This moves you from identifying a hang to fixing it in a single chat command. You'll get a fresh instance of your service immediately.
Recovery outcomes.
Where every issue lands.
What happens after your AI acts.
RECOVERY
Service is back online
The AI executes a restart, killing the hung process and spinning up a clean pod. Your application returns to a healthy state.
CLEANUP
Leaks are cleared
By cycling the pod, the AI wipes away accumulated connection leaks or memory bloat. This prevents the service from hanging again shortly after.
LOGS
Error context captured
The AI can help you identify why the hang happened before it triggers the restart. You get the context needed to fix the root cause later.
HUMAN
Manual intervention needed
If the restart fails to fix the issue, the AI will tell you. You'll need to step in to check your code or infrastructure settings.
The workflow
What your AI does when a service hangs.
The AI handles the heavy lifting of interacting with the Porter API to restore your uptime.
Detect the hang
Your AI analyzes the service status or error logs you provide to confirm the application is unresponsive.
restart_appIdentify the target
The agent locates the specific application in your Porter account that needs attention.
restart_appTrigger the restart
The AI sends the command to the Porter API to cycle the pods for that specific app.
restart_appVerify recovery
The AI checks if the new pod is running and accepting connections to ensure the fix worked.
restart_app
Try it
Copy these to start.
Use these prompts to kick off a recovery task.
Starting points
These are basic starting lines. Swap out the app names for your actual Porter service names.
Porter Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_97HRl70OC90xV4WamIvMCoFpPgzxIQZ45L5KjFmm/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Porter capabilities are ready to use.
{
"mcpServers": {
"porter-paas-mcp": {
"url": "https://edge.vinkius.com/vk_preview_97HRl70OC90xV4WamIvMCoFpPgzxIQZ45L5KjFmm/mcp"
}
}
}Restart my 'api-gateway' service in Porter.
My 'web-frontend' app is hanging, can you restart it?
Check if the 'worker-node' is unresponsive and restart it if it is.
I'm seeing connection leaks in 'auth-service', please trigger a restart.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Porter PaaS once, then ask.
Just link your Porter account. Your credentials stay encrypted and secure, and you can immediately start managing your apps through your AI client.
Connect Porter PaaS to your AIFAQ
How the task behaves.
- 01
Can the AI delete my entire application?
No. The AI only has the capability to restart existing applications.
- 02
Will this restart cause downtime?
Yes, a pod restart involves a brief period where the old pod is killed before the new one is ready. This is standard for a restart command.
- 03
Can the AI change my environment variables?
No. The AI can only trigger the restart_app tool; it cannot modify your app configuration or settings.
- 04
Does the AI need my Porter password?
No. You connect via a secure link, and your login details are never exposed to the AI.
- 05
What if the restart doesn't work?
If the restart fails to resolve the hang, the AI will report the error. At that point, you'll need to investigate your code or logs manually.
More questions about Porter PaaS? The Connector page answers them. See everything the Porter PaaS Connector can do
Connect Porter to Claude, Cursor, ChatGPT & more
