Merge changes
At the end of our development flow, there's just one thing left to do: merge our changes to the main
branch. With this step, we make the pipeline generally available and indicate that it is production-ready.
This section will explain GitOps and virtual data builds before merging our changes to the main
branch.
GitOps and Virtual Data Builds
Remember when we created a working branch back in section 3? At that point, we split off from the main
branch to start developing. In the GitOps paradigm, this is how we iterate on our pipelines.
Working in a feature branch means we leave the production version untouched. We only apply the changes to production once we are sure they do what we want. And if there's an unforeseen mistake, we can revert to a working state with a single command.
What's special about Y42 is that it doesn't just apply this paradigm to code but also to data builds. This means that when we merge a branch into main
, its data builds will also immediately be available on main
. And if we overlooked a mistake, we can instantaneously roll back to another data build.
We call this approach Virtual Data Builds. A build is virtual until you decide to merge it in, at which point it becomes the production build.
Merge to main
At this point, we can be confident that our pipeline works as we expect it to. So let's merge our changes to main
.
Create a new merge
- At the top center of the screen, click on the name of your branch (e.g.,
getting-started-guide
) - Select
Merge branch
from the dropdown menu - Verify that the Merge points from
getting-started-guide
tomain
- Press
Merge
- Wait for the merge to complete
Verify the merge
- In
main
, you should now see the entire pipeline configuration as we created it in the previous sections - Verify from the data preview that all assets contain the right data
- In your data warehouse, look up the tables generated by Y42 and check if they contain the same data
- At the top center of the screen, press
⋯
and selectOpen Git history
to see all commits on the branch - If something went wrong, you can right-click the commit and select
Revert commit
to roll back the changes
Congratulations 🎉
Outstanding! You have taken your first steps on Y42 and set up ingestion, a pipeline, a scheduler, and alerting. And with the merge to main
, you have brought everything live to production.
Further reading
If you would like to dive further into the Y42's paradigms, take a look at the following blog posts:
- Virtual Data Builds - A data warehouse environment for every Git commit (opens in a new tab)
- Stateful data assets - A declarative, stateful way to building data pipelines (opens in a new tab)
- GitOps for Data - Version control code and data together - Part 1 (opens in a new tab)
- GitOps for Data - Enabling the Write-Audit-Publish pattern by default - Part 2 (opens in a new tab)