Skip to main content
Deployment can be started for a service either manually from LocalOps console or by pushing new code to the configured service repo and branch name.

​
Automatic deployments

How a deployment gets triggered depends on what the service was created from - a source repo, a Docker image, or a Helm chart.

​
GitHub/GitLab repos

Whenever your team pushes new code to Git repository, LocalOps can automatically pull the hcanges, build them as Docker images very quickly and deploy seamlessly on the corresponding environment. If you have a local development environment for the git repository configured for the given service, you can simply follow these steps to start deployment.
If your GitHub organization doesn’t allow direct git push on the branch, you can ask your team to raise pull requests (PRs) targeting the configured branch name of the service. If you merge the PR, a new deployment will start in LocalOps with the latest commit on that branch. Or if you directly edit files in Github website and make new commits on the configured branch name of the service, a new deployment will start. A new build & deployment process will start in either of these cases. Navigate to “Deployments” tab within the same service section to see the logs.
For commits to get automatically deployed, the github user who made the commits, must be added to your LocalOps account as a user. And they must have linked their Github profile with LocalOps from inside Profile settings: https://console.localops.co/profile

​
Docker images

A service created from a Docker image has no repo to watch, so there is no push event to react to. Trigger the deployment from wherever you publish the image - typically the last step of the pipeline that pushed the new tag - by calling the Deploy Service endpoint with the tag you just published:
From CI, run that call as the step right after your image push:
Use our deploy action rather than calling the endpoint by hand:
Store the token as an encrypted secret. Pass exactly one of commit_id, docker_image_tag or helm_chart_version - whichever matches how the service was created.
Tag images with something immutable - the commit SHA works well - so each deployment points at exactly one build.

​
Helm charts

A service created from a Helm chart works the same way, with the chart version in place of the image tag:
Call it from the job that publishes the chart, so a newly pushed chart version rolls out without anyone opening the console.
All three shapes are the same endpoint - it takes commit_id, docker_image_tag or helm_chart_version depending on how the service was created. See the Deploy Service reference for the full request and response schema.

​
Manual deployment

  1. Click on “environments” on the left
  2. Navigate into the relevant environment
  3. Navigate to Services tab
  4. Click on the service where you want to trigger new deployment
  5. On the top right corner, click on “Deploy”.
  6. Verify the details on the confirmation window - the commit for a repo-based service, the image tag for a Docker image service, or the chart version for a Helm chart service
  7. Click on “Confirm” to proceed.
A new deployment will start. Navigate to “Deployments” tab within the same services section to see the logs.