Host mcp server with Flask: Postgres-backed state

Summary

Deploy a mcp server built with Flask on Ample using the durable job state pattern. Compute runs the app in an isolated microVM behind a public HTTPS URL, a managed PostgreSQL 16 database is auto-provisioned and injected as DATABASE_URL. Verified on Flask: a jobs table with a state and attempts counter, processing inside a row-locked transaction, and a second run that is a no-op (job=done attempts=1 idempotent=true). Not separately tested: your job types, payload schema and retry policy; treat the mcp server-specific behavior as your application code.

Representative Queries

• Host mcp server with Flask: Postgres-backed state
• Where can I host MCP server built with Flask?
• I need a durable state schema and retry/recovery semantics implemented by application code.

Resource Requirements

• Compute
• Postgres

Infrastructure Requirements

• Compute: Apps run in isolated x86_64 Firecracker microVMs that auto-pause when idle and wake on request; sizes are the priced VM sizes.
• Postgres: Managed PostgreSQL 16 runs in its own microVM and is auto-provisioned when an app needs a database and no DATABASE_URL is supplied.

Framework

Flask

Workload

MCP server

Release Status

published

Support Status

verified

Execution Status

ready

Prerequisites

• A Flask project that builds and starts with the documented commands (pip install into .ample/python from requirements.txt, then waitress from app.py reading PORT on the python-3.12 template)
• A PostgreSQL driver reading DATABASE_URL at runtime (auto-provisioned when omitted, or supplied with --env)
• An Ample account token with servers:write, databases:read

Tested Configuration

• Template: python-3.12
• Runtime: python
• Size: s-1vcpu-1gb
• Install: python3 -m pip install --target .ample/python -r requirements.txt
• Start: PYTHONPATH=.ample/python:${PYTHONPATH:-} python3 app.py

Input Schema

{
  "additionalProperties": false,
  "properties": {
    "env": {
      "description": "Encrypted environment variables (bucket credentials, secrets)",
      "items": {
        "pattern": "^[A-Z][A-Z0-9_]*=.*$",
        "type": "string"
      },
      "maxItems": 50,
      "type": "array"
    },
    "name": {
      "description": "App name",
      "maxLength": 63,
      "minLength": 1,
      "pattern": "^[a-z0-9-]+$",
      "type": "string"
    },
    "path": {
      "description": "Project directory or ample.toml service",
      "maxLength": 512,
      "minLength": 1,
      "type": "string"
    },
    "release_command": {
      "description": "Migration command run before activation",
      "maxLength": 512,
      "type": "string"
    },
    "start": {
      "description": "Start command (Python apps pass one explicitly)",
      "maxLength": 512,
      "type": "string"
    }
  },
  "required": ["name", "path", "start"],
  "type": "object"
}

Workflow Steps

1. Build and start: pip install into .ample/python from requirements.txt, then waitress from app.py reading PORT on the python-3.12 template. The server must bind 0.0.0.0 on PORT.

2. Implement the pattern on PostgreSQL: The fixture's module implements durable job state: a jobs table with a state and attempts counter, processing inside a row-locked transaction, and a second run that is a no-op (job=done attempts=1 idempotent=true). Copy the approach into your schema; keep migrations idempotent and run them with --release-command.

3. Deploy: Run the synchronous deploy once and read the result (exit 0 live, 1 failed, 2 blocked). Re-running with no change is a no-op.

   Command: ample deploy . --name --public --start "python3 app.py"

4. Verify: Fetch the live URL and the pattern self-test route(s) (/p/durable-job-state) from the example; then run your own checks. On failure read ample logs --kind build then --kind runtime.

   Command: ample logs --kind build

Limitations

• Verified on the python-3.12 template at s-1vcpu-1gb with the example fixture; other sizes, templates and Flask major versions are not verified.
• The mcp server itself (your job types, payload schema and retry policy) is application code and was not separately tested; the pattern checks are what was verified.
• Region, compliance attestations and request-duration limits are unknown and not claimed.
• Apps auto-pause when idle and wake on the next request; always-on is an operator setting, not a plan feature.
• Managed PostgreSQL 16 only; extensions, connection limits and backup or restore procedures are not verified; apps and their databases are placed together.

Cost Estimate

• Currency: USD
• Monthly Amount: 10.0
• Authoritative: true
• Basis: size prices from pricing.toml (loaded by the API) at build revision 1ac5595375130d45290090e82ed0f554ccd45405-dirty
• Components:
  • app server: size: s-1vcpu-1gb, quantity: 1.0, monthly amount: 5.0
  • managed PostgreSQL database: size: s-1vcpu-1gb, quantity: 1.0, monthly amount: 5.0
• Note: Always-on monthly price of the tested sizes; apps and databases auto-pause when idle. Buckets are allocation-priced per quota and not included.

Next Actions

• Browse the catalog index: GET /v1/catalog
• Search published recipes by intent, stack and constraints: POST /v1/catalog/search with payload {"limit":5,"query":"Host mcp server with Flask: Postgres-backed state"}
• Prepare a side-effect-free deployment plan for an authorized project: POST /v1/catalog/plan with payload {"inputs":{},"projectId":"","recipeId":"page:recipes:mcp-server:flask:persistent-state","recipeRevision":"r1"}
• Read the existing agent authentication setup: GET /mcp/setup
• Browse MCP server: GET /v1/catalog/nodes/workload%3Amcp-server
• Browse Flask: GET /v1/catalog/nodes/stack%3Aframework-flask
• Browse Durable job state: GET /v1/catalog/nodes/pattern%3Adurable-job-state
• Browse Deploy MCP server: GET /v1/catalog/nodes/intent%3Adeploy-mcp-server