.gitlab-ci: Remove tags no longer provided by gitlab.com

GitLab.com removed a number of tags from their hosted
runners and this meant our CI was being redirected to
our private runners at a larger cost to the Samba Team.

The new infrastructure is much larger than when we last
selected runners so we can just use the default, even for
the code coverage build.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=15638

Pair-Programmed-With: Stefan Metzmacher <[email protected]>

Signed-off-by: Andrew Bartlett <[email protected]>
Signed-off-by: Stefan Metzmacher <[email protected]>
Reviewed-by: Andreas Schneider <[email protected]>

Autobuild-User(master): Stefan Metzmacher <[email protected]>
Autobuild-Date(master): Tue May  7 13:40:55 UTC 2024 on atb-devel-224

Author: abartlet

.gitlab-ci-coverage-runners.yml

@@ -1,10 +1,4 @@
 include:
   - /.gitlab-ci-default-runners.yml
 
-.shared_runner_test:
-  # We need the more powerful n1-standard-2 runners
-  # in order to handle the lcov overhead.
-  #
-  # See .gitlab-ci-default-runners.yml for more details
-  tags:
-    - gitlab-org-docker
+# Currently we're happy with the defaults

.gitlab-ci-default-runners.yml

@@ -1,48 +1,26 @@
-# From https://docs.gitlab.com/ee/user/gitlab_com/#shared-runners:
+# From https://docs.gitlab.com/ee/ci/runners/hosted_runners/linux.html
 #
 #   ...
 #
-#   All your CI/CD jobs run on n1-standard-1 instances with 3.75GB of RAM, CoreOS
-#   and the latest Docker Engine installed. Instances provide 1 vCPU and 25GB of
-#   HDD disk space. The default region of the VMs is US East1. Each instance is
-#   used only for one job, this ensures any sensitive data left on the system can’t
-#   be accessed by other people their CI jobs.
-#
-#   The gitlab-shared-runners-manager-X.gitlab.com fleet of runners are dedicated
-#   for GitLab projects as well as community forks of them. They use a slightly
-#   larger machine type (n1-standard-2) and have a bigger SSD disk size. They don’t
-#   run untagged jobs and unlike the general fleet of shared runners, the instances
-#   are re-used up to 40 times.
-#
-#   ...
-#
-# The n1-standard-1 runners seem to be tagged with 'docker' together with 'gce'.
-#
-# The more powerful n1-standard-2 runners seem to be tagged with
-# 'gitlab-org-docker' or some with just 'gitlab-org'.
-#
+#   Runner Tag              vCPUs   Memory   Storage
+#   saas-linux-small-amd64  2       8 GB     25 GB
 #
 # Our current private runner 'docker', 'samba-ci-private', 'shared' and
 # 'ubuntu2204'. It runs with an ubuntu2204 kernel (5.15) and provides an
-# ext4 filesystem and similar RAM as the n1-standard-2 runners.
+# ext4 filesystem, 2 CPU and 4 GB (shared tag) 8G (samba-ci-private tag) RAM.
 #
 
 .shared_runner_build:
-  # We use n1-standard-1 shared runners by default.
-  #
-  # There are currently 5 shared runners with 'docker' and 'gce',
-  # while there are only 2 provising 'docker' together with 'shared'.
+  # We use saas-linux-small-amd64 shared runners by default.
+  # We avoid adding explicit tags for them in order
+  # to work with potential changes in future
   #
-  # We used to fallback to our private runner if the docker+shared runners
-  # were busy, but now that we use the 5 docker+gce runners, we try to only
-  # use shared runners without a fallback to our private runner!
-  # Lets see how that will work out.
-  tags:
-    - docker
-    - gce
+  # In order to generate valid yaml, we define a dummy variable...
+  variables:
+    SAMBA_SHARED_RUNNER_BUILD_DUMMY_VARIABLE: shared_runner_build
 
 .shared_runner_test:
-  # Currently we're fine using the n1-standard-1 runners also for testing
+  # We use saas-linux-small-amd64 shared runners by default.
   extends: .shared_runner_build
 
 .private_runner_test: