GitLab CI DinD "Cannot connect to the Docker daemon at tcp://docker:2375" in CI

By Daniel Zoghalchali·Latchkey

Your job runs Docker commands but the client cannot reach a daemon at tcp://docker:2375. Usually the docker:dind service is not declared, has not finished starting, or the TLS/port configuration does not match the daemon.

What this error means

A docker build/docker info step fails with "Cannot connect to the Docker daemon at tcp://docker:2375. Is the docker daemon running?".

Cannot connect to the Docker daemon at tcp://docker:2375. Is the docker daemon running?

Common causes

No docker:dind service attached

Without a docker:dind service the docker hostname does not resolve to a running daemon.

TLS/port mismatch

Newer dind defaults to TLS on 2376; pointing the client at plaintext 2375 without disabling TLS fails the connection.

How to fix it

Declare the dind service and DOCKER_HOST

1. Add the docker:dind service to the job.
2. Set DOCKER_HOST to the dind service and match the TLS setting.
3. Re-run the job.

build-image:
  image: docker:27
  services:
    - docker:27-dind
  variables:
    DOCKER_HOST: tcp://docker:2375
    DOCKER_TLS_CERTDIR: ""
  script:
    - docker info

Use TLS on 2376 instead

Keep TLS enabled and point the client at the TLS port with the generated certs.

variables:
  DOCKER_HOST: tcp://docker:2376
  DOCKER_TLS_CERTDIR: "/certs"
  DOCKER_CERT_PATH: "/certs/client"
  DOCKER_TLS_VERIFY: 1

How to prevent it

• Always pair Docker jobs with a matching docker:dind service.
• Keep DOCKER_HOST and the TLS/port settings consistent.
• Pin matching docker and dind image versions.

Frequently asked questions

Related guides

References

• Docker documentation
• Docker build cache
• GitHub Actions documentation