nocodb/packages/noco-docs/docs/020.getting-started/050.self-hosted/010.installation/070.aws-ecs.md

title	description	tags	keywords
AWS ECS (Fargate)	AWS ECS (Fargate) Installation	[Open Source]	[NocoDB installation NocoDB AWS Fargate installation NocoDB prerequisites]

Deploying NocoDB on AWS ECS (Fargate)

This guide will walk you through deploying NocoDB on Amazon ECS using Fargate.

Prerequisites

• AWS CLI configured with appropriate permissions
• Basic understanding of AWS ECS and Fargate

Deployment Steps

1. Create ECS Cluster

     aws ecs create-cluster --cluster-name <YOUR_ECS_CLUSTER>
   

2. Create a Log Group:

    ```bash
    aws logs create-log-group --log-group-name /ecs/<YOUR_APP_NAME>/<YOUR_CONTAINER_NAME>
    ```
   

3. Create an ECS Task Definition:

   Every time you create it, it will add a new version. If it is not existing, the version will be 1.

   aws ecs register-task-definition --cli-input-json file://task-definition.json
   

   :::tip This json file defines the container specification. You can define secrets such as NC_DB and environment variables here. :::

   Example 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>"
   }
   

4. Create an 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>
    ```
   

:::tip 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. :::

Important Notes

• Ensure that your security groups have the correct inbound and outbound rules.
• The NC_DB environment variable should be properly set to connect to your database.
• Monitor the ECS console and CloudWatch logs for any deployment issues.
• You can customize the task definition and service configuration based on your requirements.