Troubleshooting

Applications Domain Not Working?

You see the deployment succeeded, and logs are running, but the domain isn't working? Here's what to check:

1. Correct Port Mapping: Ensure the domain is using the correct port for your application. For example, if you're using Next.js, the port should be 3000, or for Laravel, it should be 8000. If you change the app port, update the domain to reflect that.

2. Avoid Using Ports in Advanced Settings: Generally, there's no need to use the Ports feature unless you want to access your app via IP:port. Leaving this feature enabled may interfere with your domain.

3. Let's Encrypt Certificates: It's crucial to point the domain to your server's IP before adding it in Dokploy. If the domain is added first, the certificate won't be generated, and you may need to recreate the domain or restart Traefik.

4. Listen on 0.0.0.0, Not 127.0.0.1: If your app is bound to 127.0.0.1 (which is common in Vite apps), switch it to 0.0.0.0 to allow external access.

Logs and Monitoring Not Working After Changing Application Placement?

This is expected behavior. If the application is running on a different node (worker), the UI won't have access to logs or monitoring, as they're not on the same node.

Mounts Are Causing My Application Not to Run?

Docker Swarm won't run your application if there are invalid mounts, even if the deployment shows as successful. Double-check your mounts to ensure they are valid or check the General Swarm Section and find your application, and you will see the real error.

Volumes in Docker Compose Not Working?

For Docker Compose, all file mounts defined in the volumes section will be stored in the files folder. This is the default directory structure:

I added a volume to my docker compose, but is not finding the volume?

For docker compose all the file mounts you've created in the volumes section will be stored to files folder, this is the default structure of the docker compose.

/application-name
    /code
    /files

So instead of using this invalid way to mount a volume:

volumes:
    - "/folder:/path/in/container" ❌

You should use this format:

volumes:
    - "../files/my-database:/var/lib/mysql" ✅
    - "../files/my-configs:/etc/my-app/config" ✅

Using Files from Your Repository

Why this happens:

• On the first deployment, the files exist and are mounted correctly
• On subsequent deployments, Dokploy cleans the directory and performs a fresh git clone
• Docker loses the reference to the files that were in the filesystem, and the new files have a new reference
• This causes mounted directories and files to be empty or missing inside the container

Solution:

1. Go to Advanced → Mounts in your Docker Compose application
2. Create a new File Mount for each file or directory you need from your repository
3. Copy the content from your repository files into the File Mount content field
4. Specify the file path for your configuration
5. Reference the file mount in your docker-compose.yml using the ../files/ path:

volumes:
  - "../files/my-config.json:/etc/my-app/config" ✅
  - "../files/my-directory:/path/in/container" ✅

Example: Instead of mounting directly from your repository:

volumes:
  - ./:/mnt/extra-addons/va_subscription_18 ❌
  - ./docker/config/odoo.conf:/etc/odoo/odoo.conf ❌

Use Dokploy's file mounts:

volumes:
  - ../files/va_subscription_18:/mnt/extra-addons/va_subscription_18 ✅
  - ../files/odoo.conf:/etc/odoo/odoo.conf ✅

Logs Not Loading When Deploying to a Remote Server?

There are a few potential reasons for this:

1. Slow Server:: If the server is too slow, it may struggle to handle concurrent requests, leading to SSL handshake errors.
2. Insufficient Disk Space: If the server doesn't have enough disk space, the logs may not load.

Docker Compose Domain Not Working?

When adding a domain in your Docker Compose file, it's not necessary to expose the ports directly. Simply specify the port where your app is running. Exposing the ports can lead to conflicts with other applications or ports.

Example of what not to do:

services:
  app:
    image: dokploy/dokploy:latest
    ports:
      - 3000:3000

Recommended approach:

services:
  app:
    image: dokploy/dokploy:latest
    ports:
      - 3000
      - 80

This is only valid for Docker Compose not for Docker Stack.

When using Docker Stack, the ports are exposed automatically, so you don't need to specify them explicitly.

Example of what not to do:

services:
  app:
    image: dokploy/dokploy:latest
    ports:
      - 3000

Recommended approach:

services:
  app:
    image: dokploy/dokploy:latest
    expose:
      - 3000

Then, when creating the domain in Dokploy, specify the service name and port, like this:

domain: my-app.com
serviceName: app
port: 3000

• Another reason of the domains are not working it may be because the healthchecks you've defined are not working, so this will cause the domains never work, so you have two options:

1. Remove the healthcheck from the service
2. Make sure the healthcheck is working

Templates and Compose Services Returning 404

If you're experiencing 404 errors when accessing services created from templates (like Docker Registry, Stalwart, Uptime Kuma, etc.) or Docker Compose services, this is usually related to how Traefik handles routing for different service types.

Understanding the Difference

Dokploy uses two different methods for configuring Traefik routing:

1. Applications (Nixpacks, Dockerfile, Buildpacks): Use the Traefik file system for routing configuration

   • Domain changes are applied automatically
   • No need to redeploy after updating domains
   • Configuration files are created in Traefik's dynamic configuration directory

2. Templates and Compose Services: Use Traefik labels for routing configuration

   • Require redeployment after any domain changes
   • Labels are read from Docker container metadata
   • Learn more about Traefik labels in the official documentation

Solution

When working with Templates or Compose services:

1. Configure your domain in the Domains section of your service
2. Redeploy the service - This is crucial! The domain changes won't take effect until you redeploy
3. Wait for the deployment to complete
4. Your service should now be accessible via the configured domain

Getting "Bad Gateway Error" When Accessing Your Application Domain

If you're encountering a Bad Gateway Error when accessing your application through its domain, this typically indicates one of several common configuration issues:

Common Causes

1. Port Mismatch: The configured port might be incorrect
2. Listen Address Configuration: The service might be listening only on 127.0.0.1 instead of 0.0.0.0

Common Solution for Modern JavaScript Frameworks

This issue frequently occurs with modern JavaScript frameworks like Vite, Astro, or Vue.js applications. By default, these frameworks often listen only on localhost (127.0.0.1).

To resolve this, you need to configure your application to listen on all available network interfaces (0.0.0.0).

Example Configuration for Vite

Here's how to properly configure a Vite application:

import { defineConfig } from "vite";
import react from "@vitejs/plugin-react";
 
export default defineConfig({
	plugins: [react()],
	preview: {
		port: 3000,
		host: true,    // This enables listening on all network interfaces
	},
	server: {        // Also add this for development server
		host: true,    // This enables listening on all network interfaces
		port: 3000
	}
});

Framework-Specific Notes

• Vite Apps: Use the configuration above
• Astro: Similar configuration in astro.config.mjs
• Vue.js: Configure in vite.config.js if using Vite
• Other Frameworks: Check your framework's documentation for network interface configuration

Remember to deploy again your application after making these changes for them to take effect.

Docker Compose Volume Mounts

When using Docker Compose, you can configure volume mounts in your docker-compose.yml file:

volumes:
  - my-database:/var/lib/mysql

Use of closed network when restarting Traefik

If you see this error in the logs of Traefik, it means that the network is being closed, this is the normal behavior when restarting Traefik.

05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="accept tcp [::]:443: use of closed network connection" entryPointName=websecure
05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="accept tcp [::]:9000: use of closed network connection" entryPointName=traefik
05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="accept tcp [::]:80: use of closed network connection" entryPointName=web
05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="close tcp [::]:9000: use of closed network connection" entryPointName=traefik
05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="close tcp [::]:443: use of closed network connection" entryPointName=websecure
05/23/25, 12:21:12 PM info 2025-05-23T09:21:12Z ERR: error="close tcp [::]:80: use of closed network connection" entryPointName=web

Creating Configuration Files

If you need to create configuration files before deploying your compose setup:

1. Go to Advanced -> Mounts
2. Create a new File Mount
3. Add your configuration content in the content field
4. Specify the file path for your configuration

Note: All File Mounts are automatically created in the /files directory. For example, if you create a file named my-config.json, it will be available at /files/my-config.json.

You can then reference this configuration file in your docker-compose.yml:

volumes:
  - ../files/my-config.json:/etc/my-app/config

Failed to initialize Docker Swarm

Error response from daemon: must specify a listening address because the address to advertise is not recognized as a system address, and a system's IP address to use could not be uniquely identified

This error occurs when the Docker Swarm is not properly initialized.

To fix this, you need to assign a the public IP address to the Docker Swarm, ideally you can use a private IP address from your network, but if you require features from docker swarm, you will need to use a public IP address.

curl -sSL https://dokploy.com/install.sh | ADVERTISE_ADDR=your-ip sh

My Dokploy UI Instance is Not Accessible

If you can't access your Dokploy UI instance, there could be several causes. While this issue won't occur with Dokploy Cloud (where our team manages the infrastructure), self-hosted instances might encounter configuration problems.

Let's go through the possible cases where your Dokploy UI instance might be inaccessible:

1. Insufficient Storage Space

If you've made many deployments and don't have available space on your server, the Dokploy database might enter recovery mode, preventing access to the user interface. Here's a quick solution to clear cache and free up server space:

docker system prune -a
docker builder prune -a
docker image prune -a

2. Container Race Condition During Restart

During a restart, a race condition might occur where Dokploy's dependent containers don't start in the correct order. To troubleshoot this:

First, verify the running containers:

docker ps

You should see all four of these containers running:

2a5b955c32b6   dokploy/dokploy:latest      "docker-entrypoint.s…"   4 days ago     Up 4 days     0.0.0.0:3000->3000/tcp, :::3000->3000/tcp                                                                             dokploy.1.4bkuszk98muz372kw5mvwkw0h
5a989bf52bc6   postgres:16                 "docker-entrypoint.s…"   4 days ago     Up 4 days     5432/tcp                                                                                                              dokploy-postgres.1.9hvjaxrmby7ex2denjtwo0csf
a29d56342175   redis:7                     "docker-entrypoint.s…"   4 days ago     Up 4 days     6379/tcp                                                                                                              dokploy-redis.1.epl51a9bt8yr7ur0f1akeeyuk
05be01c5612f   traefik:v2.5                "/entrypoint.sh trae…"   4 days ago     Up 4 days     0.0.0.0:80->80/tcp, :::80->80/tcp, 0.0.0.0:443->443/tcp, :::443->443/tcp, 0.0.0.0:8080->8080/tcp, :::8080->8080/tcp   dokploy-traefik.1.2oktabjmfu558x2d2dy6czt8m

If all four containers are running but you still can't access the interface, it's time to debug:

Debugging Process

1. Check Container Logs

Start by examining the logs of each container:

docker service logs dokploy # Dokploy UI
docker service logs dokploy-postgres # Postgres
docker service logs dokploy-redis # Redis
docker logs dokploy-traefik # Traefik

2. Common Database Connection Issue

A common case is when the Postgres container starts after the Dokploy container, preventing Dokploy from connecting to the database. You might see logs like this when running docker service logs dokploy:

> dokploy@v0.22.3 start /app
> node -r dotenv/config dist/server.mjs
 
Default middlewares already exists
Network is already initilized
Main config already exists
Default traefik config already exists
Migration failed [Error: getaddrinfo ENOTFOUND dokploy-postgres] {
  errno: -3008,
  code: 'ENOTFOUND',
  syscall: 'getaddrinfo',
  hostname: 'dokploy-postgres'
}
Setting up cron jobs....
Main Server Error [Error: getaddrinfo ENOTFOUND dokploy-postgres] {
  errno: -3008,
  code: 'ENOTFOUND',
  syscall: 'getaddrinfo',
  hostname: 'dokploy-postgres'
}

To fix this, restart the Dokploy service:

docker service scale dokploy=0
# Then
docker service scale dokploy=1

3. Traefik Configuration Issues

If all containers are running but you still can't access the UI, and Dokploy logs show no errors, the Traefik container might have configuration issues.

When running docker logs dokploy-traefik, you might see errors like:

2025-04-07T15:20:18Z ERR Error occurred during watcher callback error="/etc/dokploy/traefik/dynamic/dokploy.yml: field not found, node: passHostHeader" providerName=file

First, try restarting Traefik:

docker restart dokploy-traefik

If you still can't access it and the same error persists in the Traefik logs, you'll need to check the Traefik configuration. In this case, the error indicates that the passHostHeader field is missing in the configuration.

If you've modified any Traefik configuration for an application and added invalid configuration, the logs will point to the error. For example, the error above mentions field not found, node: passHostHeader, which means we need to manually modify the configuration files in /etc/dokploy/traefik.

Here's an example of an invalid configuration:

http:
  routers:
    dokploy-router-app:
      rule: Host(`my-domain.com`)
      service: dokploy-service-app
      entryPoints:
        - web
      middlewares:
        - redirect-to-https
    dokploy-router-app-secure:
      rule: Host(`my-domain.com`)
      service: dokploy-service-app
      entryPoints:
        - websecure
      tls:
        certResolver: letsencrypt
  services:
    dokploy-service-app:
      loadBalancer:
        servers:
          - url: http://dokploy:3000
          - passHostHeader: true

The correct configuration should be:

http:
  routers:
    dokploy-router-app:
      rule: Host(`my-domain.com`)
      service: dokploy-service-app
      entryPoints:
        - web
      middlewares:
        - redirect-to-https
    dokploy-router-app-secure:
      rule: Host(`my-domain.com`)
      service: dokploy-service-app
      entryPoints:
        - websecure
      tls:
        certResolver: letsencrypt
  services:
    dokploy-service-app:
      loadBalancer:
        servers:
          - url: http://dokploy:3000
        passHostHeader: true

After fixing the configuration, restart Traefik:

docker restart dokploy-traefik

You should now be able to access the user interface.

Recreate dokploy containers

In the case you want to recreate the dokploy services, you can do the following:

Remove the dokploy-redis service:

docker service rm dokploy-redis
 
# Create a new dokploy-redis service
  docker service create \
  --name dokploy-redis \
  --constraint 'node.role==manager' \
  --network dokploy-network \
  --mount type=volume,source=dokploy-redis,target=/data \
  redis:7

Remove the dokploy-postgres service:

docker service rm dokploy-postgres
 
# Create a new dokploy-postgres service
  docker service create \
  --name dokploy-postgres \
  --constraint 'node.role==manager' \
  --network dokploy-network \
  --env POSTGRES_USER=dokploy \
  --env POSTGRES_DB=dokploy \
  --env POSTGRES_PASSWORD=amukds4wi9001583845717ad2 \
  --mount type=volume,source=dokploy-postgres,target=/var/lib/postgresql/data \
  postgres:16

Remove the dokploy-traefik service:

# If you are using docker standalone traefik
docker rm -f dokploy-traefik
 
docker run -d \
    --name dokploy-traefik \
    --restart always \
    -v /etc/dokploy/traefik/traefik.yml:/etc/traefik/traefik.yml \
    -v /etc/dokploy/traefik/dynamic:/etc/dokploy/traefik/dynamic \
    -v /var/run/docker.sock:/var/run/docker.sock:ro \
    -p 80:80/tcp \
    -p 443:443/tcp \
    -p 443:443/udp \
    traefik:v3.6.1
 
docker network connect dokploy-network dokploy-traefik
 
# If you are using docker service traefik
docker service rm dokploy-traefik
 
# Create a new dokploy-traefik service
docker service create \
  --name dokploy-traefik \
  --constraint 'node.role==manager' \
  --network dokploy-network \
  --mount type=bind,source=/etc/dokploy/traefik/traefik.yml,target=/etc/traefik/traefik.yml \
  --mount type=bind,source=/etc/dokploy/traefik/dynamic,target=/etc/dokploy/traefik/dynamic \
  --mount type=bind,source=/var/run/docker.sock,target=/var/run/docker.sock \
  --publish mode=host,published=443,target=443 \
  --publish mode=host,published=80,target=80 \
  --publish mode=host,published=443,target=443,protocol=udp \
  traefik:v3.6.1

Remove the dokploy service:

docker service rm dokploy
 
# Create a new dokploy service
 
# We need the advertise address to be set which is the Private IP of your server, you can get it by running the following command:
 
# Run this command to get the private IP of your server:
 
# Copy this value and paste in the ADVERTISE_ADDR variable:
ip addr show | grep -E "inet (192\.168\.|10\.|172\.1[6-9]\.|172\.2[0-9]\.|172\.3[0-1]\.)" | head -n1 | awk '{print $2}' | cut -d/ -f1
 
# Create the dokploy service
docker service create \
  --name dokploy \
  --replicas 1 \
  --network dokploy-network \
  --mount type=bind,source=/var/run/docker.sock,target=/var/run/docker.sock \
  --mount type=bind,source=/etc/dokploy,target=/etc/dokploy \
  --mount type=volume,source=dokploy,target=/root/.docker \
  --publish published=3000,target=3000,mode=host \
  --update-parallelism 1 \
  --update-order stop-first \
  --constraint 'node.role == manager' \
  -e ADVERTISE_ADDR="Eg: 192.168.1.100" \
  dokploy/dokploy:latest

Final Notes

While the specific issues may vary, the general troubleshooting approach remains similar to what we've described above, and is the general way we always follow when correcting a problem related to the dokploy instance not starting.. If you still can't access the user interface:

1. Check that all containers are running properly
2. Review the logs of each container for specific error messages
3. Verify all configuration files
4. Make sure to read the Traefik documentation for detailed configuration options: https://doc.traefik.io/traefik/