Self-hosting (Docker)

AskTable can run fully self-hosted: every component runs as a Docker container in your own environment, and data never leaves your network. This page covers the all-in-one Docker Compose deployment.

Prerequisites

Hardware (minimum): 4-core CPU, 16 GB RAM, 100 GB disk, x86. Software: a Linux server + Docker + Docker Compose v2.23.1+ (deployment uses the configs.content inline feature). Network: the host needs outbound HTTPS (443) to:

Address	Purpose
`registry.cn-shanghai.aliyuncs.com`	Pull images
`dockerauth.cn-hangzhou.aliyuncs.com`	Image authentication
`aiproxy.asktable.com`	AI model service (when using AskTable’s hosted models)
`lf.asktable.com`	AI call tracing

Fully offline: preload the images into your network and choose Custom in Configuring LLMs to use an in-network model — then no internet access is needed.

docker-compose.yaml

Create a directory and save the following as docker-compose.yaml. Each service’s role is in the inline comments; backing services talk to each other over the Compose internal network and only asktable exposes a host port. For servers outside mainland China, switch the images to the Docker Hub versions in the comments (datamini/...).

docker-compose.yaml

services:
  # Main service: backend API + frontend + built-in MCP (the only service exposed externally)
  asktable:
    image: registry.cn-shanghai.aliyuncs.com/datamini/asktable-all-in-one:latest  # mainland China; elsewhere use datamini/asktable-all-in-one:latest
    container_name: asktable
    depends_on:
      - asktable-pg
      - asktable-redis
      - asktable-atbox
    ports:
      - "8000:80"
    privileged: true
    environment:
      TZ: Asia/Shanghai
      AT_REGION: global                   # cn or global; keep static after deployment
      DATABASE_HOST: asktable-pg
      DATABASE_PORT: 5432
      DATABASE_USER: asktable
      DATABASE_PASSWORD: asktable          # [change me]
      DATABASE_DB: asktable
      REDIS_URL: redis://asktable-redis:6379
      ATBOX_URL: http://asktable-atbox:5300
      # Workbook's own database: same instance/role, separate database (PG has no cross-db query -> hard isolation)
      WORKBOOK_PG_DSN: postgresql://asktable:asktable@asktable-pg:5432/asktable_workbook
    volumes:
      - ./asktable_data:/asktable

  # PostgreSQL 17 (pgvector): primary database (metadata, config, field vectors)
  asktable-pg:
    image: pgvector/pgvector:pg17
    container_name: asktable-pg
    restart: unless-stopped
    command: ["postgres", "-c", "max_connections=100"]
    environment:
      POSTGRES_USER: asktable
      POSTGRES_PASSWORD: asktable          # [change me] keep consistent with DATABASE_PASSWORD
      POSTGRES_DB: asktable
    volumes:
      - ./postgres_data:/var/lib/postgresql/data
    configs:
      # Runs once on a fresh deployment with an empty data volume; creates the workbook database
      - source: workbook-initdb
        target: /docker-entrypoint-initdb.d/01-create-workbook-db.sql

  # Redis 7.4: task queue
  asktable-redis:
    image: registry.cn-shanghai.aliyuncs.com/dminfra/redis:7.4.4
    container_name: asktable-redis
    restart: unless-stopped
    command: redis-server --appendonly yes
    volumes:
      - ./redis_data:/data

  # Python sandbox: isolated execution of untrusted, agent-generated Python.
  # Options marked "security" are the sandbox itself — do not remove; "tunable" resource limits can be adjusted.
  asktable-atbox:
    image: registry.cn-shanghai.aliyuncs.com/datamini/asktable-atbox:latest  # mainland China; elsewhere use datamini/asktable-atbox:latest
    container_name: asktable-atbox
    restart: unless-stopped
    environment:
      TZ: Asia/Shanghai
      ATBOX_MAX_WORKERS: 8          # tunable: concurrent workers
      ATBOX_DEFAULT_TIMEOUT: 30s    # tunable: default execution timeout
      ATBOX_MAX_TIMEOUT: 120s       # tunable: max execution timeout
      ATBOX_ENABLE_NETWORK: "0"     # security: no network inside the sandbox
    # security: read-only rootfs + writable scratch (/var/atbox/root/tmp is the chroot workdir); paired, do not remove
    read_only: true
    tmpfs:
      - /var/atbox/root/tmp:rw,noexec,nosuid,nodev,mode=1777,size=512m
      - /tmp:rw,noexec,nosuid,nodev,mode=1777,size=64m
    # security: drop all capabilities, add back only the minimal set chroot + UID-drop need; do not change
    cap_drop:
      - ALL
    cap_add:
      - CHOWN
      - DAC_OVERRIDE
      - SETUID
      - SETGID
      - SYS_CHROOT
    security_opt:
      - no-new-privileges:true      # security: block privilege escalation
    pids_limit: 128                 # tunable: max processes (anti fork-bomb)
    mem_limit: 1g                   # tunable: memory cap
    cpus: 1.0                       # tunable: CPU cap

configs:
  # Inline create-database SQL; requires Docker Compose v2.23.1+ (configs.content support)
  workbook-initdb:
    content: |
      CREATE DATABASE asktable_workbook;

Before a production deployment, change the values marked [change me]: DATABASE_PASSWORD (keep it consistent with POSTGRES_PASSWORD).

Start and initialize

Start

From the directory containing docker-compose.yaml:

docker compose up -d

The first start initializes the databases automatically (including the separate Workbook database).

Open

Open http://<your-server>:8000/ in a browser.

Initialize

Follow the wizard to name the organization, create an admin account, and configure an LLM (model group). AskTable creates the single default organization and makes the first administrator its Owner.

Access URLs

Entry	URL
Console UI	`http://<server>:8000/`
API reference	`http://<server>:8000/api/v1/redoc`
MCP server	`http://<server>:8000/mcp/`

Configure an LLM

AskTable is not tied to a specific model — you configure one before you can ask questions. The first-run wizard walks you through it. See Configuring LLMs.

Operations and connection capacity

The built-in PostgreSQL connection limit is 100, and the default pools fit typical single-node deployments. Read Operations and capacity planning only when adding Web or Worker processes or when DATABASE_BUSY occurs repeatedly.

Enable Workbook

Workbook is the built-in writable dataset, stored in a separate asktable_workbook database that is hard-isolated from the main one. The compose above already sets WORKBOOK_PG_DSN and inlines the create-database SQL:

Fresh deployment: the database is created automatically on first start — nothing to do.
Upgrading an existing deployment: the data volume is already initialized and the init script no longer runs, so create the database once manually:
docker compose exec asktable-pg psql -U asktable -d asktable -c "CREATE DATABASE asktable_workbook;" docker compose up -d

Upgrading

docker compose pull
docker compose down
docker compose up -d

Upgrading from an older version (remove Qdrant, switch to the pgvector image, handle the collation-version warning, rebuild embeddings) — see Migration.

Uninstalling

docker compose down
rm -rf asktable_data postgres_data redis_data

Removing the data volumes erases all data permanently. Back up first.

Next steps

Configuring LLMs — connect your own or a hosted LLM
Operations and capacity planning — database pools, process concurrency, and troubleshooting
AI Search (Value Index) — match abbreviations to full stored values (optional, needs aisearch)
Advanced configuration — subpath deployment, white-labeling
Commercial licensing — trial limits and the enterprise license
Connecting Data — bring your data in

​Prerequisites

​docker-compose.yaml

​Start and initialize

​Access URLs

​Configure an LLM

​Operations and connection capacity

​Enable Workbook

​Upgrading

​Uninstalling

​Next steps

Prerequisites

docker-compose.yaml

Start and initialize

Access URLs

Configure an LLM

Operations and connection capacity

Enable Workbook

Upgrading

Uninstalling

Next steps