Committing code
When working on a development project, it is important to stick to these simple commit rules:
Work in feature branches where possible (see Branching policy)
Commit early, commit often.
Have the Jira story ID at the beginning of your commit messages. (You can also use Gitlab and JIRA integration defined in Working with Jira)
Git logs shall be human readable in sequence, describing the development activity.
Use imperative forms in the commit message:
ST-320 make fluentd,kibana,elasticsearch optional
* add a pipeline example for disabling ELK
* add enabled checks
* filebeat,journalbeat support added to clusters
# Please enter the commit message for your changes. Lines starting
# with '#' will be ignored, and an empty message aborts the commit.
#
# Date: Tue May 12 11:24:17 2020 +1200
#
# On branch st-320-swap-out-fluentd
# Your branch is up-to-date with 'origin/st-320-swap-out-fluentd'.
You can find additional information on how to write a good commit message here.
Squashing commits
If you want to replace a series of small commits with a single commit or if you want to make their order more logical you can use an interactive rebase (git rebase -i) to squash multiple commits into one or reorder them. When squashing commits it is important to consider the following:
You should never rebase commits you have pushed to a remote server.
You should also never rebase commits authored by other people.
In general the squashing of commits is discouraged for SKA repositories.