- Requirements and limitations
- How to add a merge request to a merge train
- How to remove a merge request from a merge train
- How to view a merge request’s current position on the merge train
- Start/Add to merge train when pipeline succeeds
- Immediately merge a merge request with a merge train
When pipelines for merged results are enabled, the pipeline jobs run as if the changes from your source branch have already been merged into the target branch.
However, the target branch may be changing rapidly. When you’re ready to merge, if you haven’t run the pipeline in a while, the target branch may have already changed. Merging now could introduce breaking changes.
Merge trains can prevent this from happening. A merge train is a queued list of merge requests, each waiting to be merged into the target branch.
Each merge request on the train runs the merged results pipeline immediately before its changes are merged into the target branch. If the pipeline fails, the breaking changes are not merged, and the target branch is unaffected.
Many merge requests can be added to the train. Each is trying to merge into the target branch. Each request runs its own merged results pipeline, which includes the changes from all of the other merge requests in front of it on the train. All the pipelines run in parallel, to save time.
If the pipeline for the merge request at the front of the train completes successfully, the changes are merged into the target branch, and the other pipelines will continue to run.
If one of the pipelines fails, it is removed from the train, and all pipelines behind it restart, but without the changes that were removed.
Three merge requests (
C) are added to a merge train in order, which
creates three merged results pipelines that run in parallel:
- The first pipeline runs on the changes from
Acombined with the target branch.
- The second pipeline runs on the changes from
Bcombined with the target branch.
- The third pipeline runs on the changes from
Ccombined with the target branch.
If the pipeline for
B fails, it is removed from the train. The pipeline for
C restarts with the
C changes, but without the
A then completes successfully, it merges into the target branch, and
to run. If more merge requests are added to the train, they will now include the
changes that are included in the target branch, and the
C changes that are from
the merge request already in the train.
Learn more about how merge trains keep your master green.
Merge trains have the following requirements and limitations:
- GitLab 12.0 and later requires Redis 3.2 or higher.
- Pipelines for merged results must be configured properly.
- Each merge train can run a maximum of twenty pipelines in parallel. If more than twenty merge requests are added to the merge train, the merge requests will be queued until a slot in the merge train is free. There is no limit to the number of merge requests that can be queued.
Watch this video for a demonstration on how parallel execution of Merge Trains can prevent commits from breaking the default branch.
To add a merge request to a merge train:
- Visit a merge request.
- Click the Start/Add to merge train button.
- Visit a merge request.
- Click the Remove from merge train button.
After a merge request has been added to the merge train, the merge request’s current position will be displayed under the pipeline widget:
You can add a merge request to a merge train only when the latest pipeline in the merge request is finished. While the pipeline is running or pending, you cannot add the merge request to a train because the current change of the merge request may be broken thus it could affect the following merge requests.
In this case, you can schedule to add the merge request to a merge train when the latest pipeline succeeds (This pipeline is Pipelines for merged results, not Pipelines for merge train). You can see the following button instead of the regular Start/Add to merge train button while the latest pipeline is running.
In the case where you have a high-priority merge request (for example, a critical patch) to be merged urgently, you can use Merge Immediately option for bypassing the merge train. This is the fastest option to get the change merged into the target branch.
However, every time you merge a merge request immediately, it could affect the existing merge train to be reconstructed, specifically, it regenerates expected merge commits and pipelines. This means, merging immediately essentially wastes CI resources. Because of these downsides, you will be asked to confirm before the merge is initiated:
If a merge request is not mergeable (for example, it’s WIP, there is a merge conflict, etc), your merge request will be dropped from the merge train automatically.
In these cases, the reason for dropping the merge request is in the system notes.
To check the reason:
- Open the merge request that was dropped from the merge train.
- Open the Discussion tab.
- Find a system note that includes either:
- The text … removed this merge request from the merge train because …
- … aborted this merge request from the merge train because … The reason is given in the text after the because … phrase.
Follow this issue to track progress on this issue.
A Merge Train pipeline cannot be retried because the merge request is dropped from the merge train upon failure. For this reason, the retry button does not appear next to the pipeline icon.
In the case of pipeline failure, you should re-enqueue the merge request to the merge train, which will then initiate a new pipeline.
First of all, please check if merge immediately is available as a workaround in your workflow. This is the most recommended workaround you’d be able to take immediately. If it’s not available or acceptable, please read through this section.
Merge train is enabled by default when you enable Pipelines for merged results,
however, you can disable this feature by setting the
:disable_merge_trains feature flag to
When you disable this feature, all existing merge trains are aborted and
the Start/Add Merge Train button no longer appears in merge requests.
To check if the feature flag is enabled on your GitLab instance, please ask an administrator to execute the following commands :
> sudo gitlab-rails console # Login to Rails console of GitLab instance. > Feature.enabled?(:disable_merge_trains) # Check if it's disabled or not. > Feature.enable(:disable_merge_trains) # Disable Merge Trains.