Batch insert CI profiling spec data
Instead of inserting a row after each example to an external database,
we save the CI profiling reports into the rspec_profiling
directory
and insert the data in the update-tests-metadata
CI stage.
This should make each spec run faster and also reduce the number of
PostgreSQL connections needed by concurrent CI builds.
scripts/insert-rspec-profiling-data
also inserts one file at a time
via the PostgreSQL COPY command for faster inserts. created_at
and updated_at
are stored in the CSV when the example runs.
Note that CI statistics will only be inserted if the build is successful.
update-tests-metadata
doesn't run otherwise. We might miss some slow test
runs when the build is red.
Sample run: https://gitlab.com/gitlab-org/gitlab-ee/pipelines/51379201