fix: ensure docker in docker can run properly via new env variable

[davemooreuws]

Fixes #769

Adds new env variable called NITRIC_DOCKER_HOST , this can be set to an IP like 172.17.0.4 on runners like GitLab and other Docker in Docker environments.

I have tested on:

• GitLab runners with docker 27

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 2 lines in your changes missing coverage. Please review.

Project coverage is 1.08%. Comparing base (1723d34) to head (1b4ad81).
Report is 1 commits behind head on main.

Files	Patch %	Lines
pkg/project/service.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #770   +/-   ##
=====================================
  Coverage   1.08%   1.08%           
=====================================
  Files         80      79    -1     
  Lines       8886    8853   -33     
=====================================
  Hits          96      96           
+ Misses      8780    8747   -33     
  Partials      10      10           

Flag	Coverage Δ
unittests	1.08% <0.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Resolved

[tjholm]

It looks like gitlab exposes a docker hostname locally.

DOCKER_HOST as you've mentioned appears to be a convention for configuring this in CI/CD.

When you tested setting DOCKER_HOST did you use the env variable in the CLI to set host.docker.internal to the same value?

Rather than searching the eth0 which may not apply in all cases we could look at allowing it to be configured? e.g. use DOCKER_HOST or something else like NITRIC_HOST?

Main concern is that a search for eth0 will not cover all cases or may change/break in ways we may not be able to control or predict.

[davemooreuws]

It looks like gitlab exposes a docker hostname locally.

DOCKER_HOST as you've mentioned appears to be a convention for configuring this in CI/CD.

When you tested setting DOCKER_HOST did you use the env variable in the CLI to set host.docker.internal to the same value?

Rather than searching the eth0 which may not apply in all cases we could look at allowing it to be configured? e.g. use DOCKER_HOST or something else like NITRIC_HOST?

Main concern is that a search for eth0 will not cover all cases or may change/break in ways we may not be able to control or predict.

Agreed, we want to avoid it being brittle. Will need to double check setting the same value as DOCKER_HOST. I think a nitric based env var will work more broadly (have a feeling the azure and GCP might need configuring as well), but let's try it tomorrow. We can keep the WSL fix in.

[nitric-bot]

🎉 This PR is included in version 1.50.4 🎉

The release is available on GitHub release

Your semantic-release bot 📦🚀