---
title: 'Installation'
description: 'Simple installation - takes about three minutes!'
position: 1
category: 'Getting started'
menuTitle: 'Installation'
link: https://codesandbox.io/embed/vigorous-firefly-80kq5?hidenavigation=1& theme=dark
---
Simple installation - takes about three minutes!
## Prerequisites
- __Must haves__
* [node.js >= 12 ](https://nodejs.org/en/download ) / [Docker ](https://www.docker.com/get-started )
* [MySql ](https://dev.mysql.com/downloads/mysql/ ) / [Postgres ](https://www.postgresql.org/download/ ) / [SQLserver ](https://www.microsoft.com/en-gb/sql-server/sql-server-downloads ) / SQLite Database
- Nice to haves
- Existing schemas can help to create APIs quickly.
- An example database schema can be found < a class = "grey--text" href = "https://github.com/lerocha/chinook-database/tree/master/ChinookDatabase/DataSources" > < u > here</ u ></ a > .
## Quick try
### 1-Click Deploy to Heroku
< a href = "https://heroku.com/deploy?template=https://github.com/npgia/nocodb-seed-heroku" >
< img
src="https://www.herokucdn.com/deploy/button.svg"
width="300px"
alt="Deploy NocoDB to Heroku with 1-Click"
/>
< / a >
### Node app / Docker
< code-group >
< code-block label = "NPX" active >
```bash
npx create-nocodb-app
```
< / code-block >
< code-block label = "Docker" >
```bash
docker run -d --name nocodb -p 8080:8080 nocodb/nocodb:latest
```
< / code-block >
< code-block label = "Using Git" >
```bash
git clone https://github.com/nocodb/nocodb-seed
cd nocodb-seed
npm install
npm start
```
< / code-block >
< / code-group >
> To persist data in docker you can mount volume at `/usr/app/data/` since 0.10.6. In older version mount at `/usr/src/app`.
## Development Setup
If you want to modify the source code,
- Start the backend locally
```bash
cd packages/nocodb
npm install
npm run watch:run
```
- Start the frontend locally
```bash
cd packages/nc-gui
npm install
npm run dev
```
- Open ``localhost:3000/dashboard`` in browser
< alert >
nocodb/packages/nocodb includes nc-lib-gui which is the built version of nc-gui hosted in npm registry. < br >
You can visit localhost:8000/dashboard in browser after starting the backend locally if you just want to modify the backend only.
< / alert >
## Production Setup
NocoDB requires a database to store metadata of spreadsheets views and external databases.
And connection params for this database can be specified in `NC_DB` environment variable.
### Docker
< code-group >
< code-block label = "MySQL" active >
```bash
docker run -d -p 8080:8080 \
-e NC_DB="mysql2://host.docker.internal:3306?u=root& p=password& d=d1" \
-e NC_AUTH_JWT_SECRET="569a1821-0a93-45e8-87ab-eb857f20a010" \
nocodb/nocodb:latest
```
< / code-block >
< code-block label = "Postgres" >
```bash
docker run -d -p 8080:8080 \
-e NC_DB="pg://host:port?u=user& p=password& d=database" \
-e NC_AUTH_JWT_SECRET="569a1821-0a93-45e8-87ab-eb857f20a010" \
nocodb/nocodb:latest
```
< / code-block >
< code-block label = "SQL Server" >
```bash
docker run -d -p 8080:8080 \
-e NC_DB="mssql://host:port?u=user& p=password& d=database" \
-e NC_AUTH_JWT_SECRET="569a1821-0a93-45e8-87ab-eb857f20a010" \
nocodb/nocodb:latest
```
< / code-block >
< / code-group >
### Environment variables
| Variable | Mandatory | Comments | If absent |
|-------------------------|-----------|----------------------------------------------------------------------------------|--------------------------------------------|
| NC_DB | Yes | See our database URLs | A local SQLite will be created in root folder |
| NC_DB_JSON | Yes | Can be used instead of `NC_DB` and value should be valid knex connection JSON | |
| NC_DB_JSON_FILE | Yes | Can be used instead of `NC_DB` and value should be a valid path to knex connection JSON | |
| DATABASE_URL | No | JDBC URL Format. Can be used instead of NC_DB. Used in 1-Click Heroku deployment| |
| DATABASE_URL_FILE | No | path to file containing JDBC URL Format. Can be used instead of NC_DB. Used in 1-Click Heroku deployment| |
| NC_DASHBOARD_URL | No | Custom dashboard url path | `/dashboard` |
| NC_TOOL_DIR | No | App directory to keep metadata and app related files | Defaults to current working directory. In docker maps to `/usr/app/data/` for mounting volume. |
| NC_PUBLIC_URL | Yes | Used for sending Email invitations | Best guess from http request params |
| NC_AUTH_JWT_SECRET | Yes | JWT secret used for auth and storing other secrets | A Random secret will be generated |
| NC_JWT_EXPIRES_IN | No | JWT token expiry time | `10h` |
| NC_CONNECT_TO_EXTERNAL_DB_DISABLED | No | Disable Project creation with external database | |
| NC_INVITE_ONLY_SIGNUP | No | Allow users to signup only via invite url, value should be any non-empty string. | |
| NC_BACKEND_URL | No | Custom Backend URL | ``http://localhost:8080`` will be used |
| NC_REQUEST_BODY_SIZE | No | Request body size [limit ](https://expressjs.com/en/resources/middleware/body-parser.html#limit ) | `1048576` |
| NC_EXPORT_MAX_TIMEOUT | No | After NC_EXPORT_MAX_TIMEOUT csv gets downloaded in batches | Default value 5000(in millisecond) will be used |
| DB_QUERY_LIMIT_DEFAULT | No | Default pagination limit | 25 |
| DB_QUERY_LIMIT_MAX | No | Maximum allowed pagination limit | 100 |
| DB_QUERY_LIMIT_MIN | No | Minimum allowed pagination limit | 1 |
| NC_DISABLE_TELE | No | Disable telemetry | |
| NC_GOOGLE_CLIENT_ID | No | Google client id to enable google authentication | |
| NC_GOOGLE_CLIENT_SECRET | No | Google client secret to enable google authentication | |
| NC_MIGRATIONS_DISABLED | No | Disable NocoDB migration | |
| NC_ONE_CLICK | No | Used for Heroku one-click deployment | |
| PORT | No | For setting app running port | `8080` |
| NC_SENTRY_DSN | No | For Sentry monitoring | |
| NC_DISABLE_ERR_REPORT | No | Disable error reporting | |
| AWS_ACCESS_KEY_ID | No | For Litestream - S3 access key id | If Litestream is configured and NC_DB is not present. SQLite gets backed up to S3 |
| AWS_SECRET_ACCESS_KEY | No | For Litestream - S3 secret access key | If Litestream is configured and NC_DB is not present. SQLite gets backed up to S3 |
| AWS_BUCKET | No | For Litestream - S3 bucket | If Litestream is configured and NC_DB is not present. SQLite gets backed up to S3 |
| AWS_BUCKET_PATH | No | For Litestream - S3 bucket path (like folder within S3 bucket) | If Litestream is configured and NC_DB is not present. SQLite gets backed up to S3 |
### Docker Compose
< code-group >
< code-block label = "MySQL" active >
```bash
git clone https://github.com/nocodb/nocodb
cd nocodb
cd docker-compose
cd mysql
docker-compose up
```
< / code-block >
< code-block label = "Postgres" >
```bash
git clone https://github.com/nocodb/nocodb
cd nocodb
cd docker-compose
cd pg
docker-compose up
```
< / code-block >
< code-block label = "SQL Server" >
```bash
git clone https://github.com/nocodb/nocodb
cd nocodb
cd docker-compose
cd mssql
docker-compose up
```
< / code-block >
< / code-group >
### AWS ECS (Fargate)
#### Create ECS Cluster
```
aws ecs create-cluster \
--cluster-name < YOUR_ECS_CLUSTER >
```
#### Create Log group
```
aws logs create-log-group \
--log-group-name /ecs/< YOUR_APP_NAME > /< YOUR_CONTAINER_NAME >
```
#### Create ECS Task Definiton
Every time you create it, it will add a new version. If it is not existing, the version will be 1.
```bash
aws ecs register-task-definition \
--cli-input-json "file://./< YOUR_TASK_DEF_NAME > .json"
```
< alert >
This json file defines the container specification. You can define secrets such as NC_DB and environment variables here.
< / alert >
Here's the sample Task Definition
```json
{
"family": "nocodb-sample-task-def",
"networkMode": "awsvpc",
"containerDefinitions": [{
"name": "< YOUR_CONTAINER_NAME > ",
"image": "nocodb/nocodb:latest",
"essential": true,
"logConfiguration": {
"logDriver": "awslogs",
"options": {
"awslogs-group": "/ecs/< YOUR_APP_NAME > /< YOUR_CONTAINER_NAME > ",
"awslogs-region": "< YOUR_AWS_REGION > ",
"awslogs-stream-prefix": "ecs"
}
},
"secrets": [{
"name": "< YOUR_SECRETS_NAME > ",
"valueFrom": "< YOUR_SECRET_ARN > "
}],
"environment": [{
"name": "< YOUR_ENV_VARIABLE_NAME > ",
"value": "< YOUR_ENV_VARIABLE_VALUE > "
}],
"portMappings": [{
"containerPort": 8080,
"hostPort": 8080,
"protocol": "tcp"
}]
}],
"requiresCompatibilities": [
"FARGATE"
],
"cpu": "256",
"memory": "512",
"executionRoleArn": "< YOUR_ECS_EXECUTION_ROLE_ARN > ",
"taskRoleArn": "< YOUR_ECS_TASK_ROLE_ARN > "
}
```
#### Create ECS Service
```bash
aws ecs create-service \
--cluster < YOUR_ECS_CLUSTER > \
--service-name < YOUR_SERVICE_NAME > \
--task-definition < YOUR_TASK_DEF > :< YOUR_TASK_DEF_VERSION > \
--desired-count < DESIRED_COUNT > \
--launch-type "FARGATE" \
--platform-version < VERSION > \
--health-check-grace-period-seconds < GRACE_PERIOD_IN_SECOND > \
--network-configuration "awsvpcConfiguration={subnets=["< YOUR_SUBSETS > "], securityGroups=["< YOUR_SECURITY_GROUPS > "], assignPublicIp=ENABLED}" \
--load-balancer targetGroupArn=< TARGET_GROUP_ARN > ,containerName=< CONTAINER_NAME > ,containerPort=< YOUR_CONTAINER_PORT >
```
< alert >
If your service fails to start, you may check the logs in ECS console or in Cloudwatch. Generally it fails due to the connection between ECS container and NC_DB. Make sure the security groups have the correct inbound and outbound rules.
< / alert >
## Sample Demos
### Code Sandbox
< code-sandbox :src = "link" > < / code-sandbox >
### Docker deploying with one command
< youtube id = "K-UEecQyiOk" > < / youtube >
### Using NPX
< youtube id = "v6Nn75P1p7I" > < / youtube >
### Heroku Deployment
< youtube id = "WB7yYXfhocY" > < / youtube >