[ci] Divide subsystem-regression-tests into 2 jobs (#4076)

Currently `subsystem-regression-tests` job fails if the first benchmarks
fail and there is no result for the second benchmark. Also dividing the
job makes the pipeline faster (currently it's a longest job)

cc https://github.com/paritytech/ci_cd/issues/969
cc @AndreiEres

---------

Co-authored-by: Andrei Eres <eresav@me.com>
This commit is contained in:
Alexander Samusev
2024-04-11 17:29:00 +02:00
committed by GitHub
parent f13408d57f
commit 6ebf491b50
2 changed files with 25 additions and 4 deletions
+6 -2
View File
@@ -70,7 +70,9 @@ publish-subsystem-benchmarks:
- .kubernetes-env
- .publish-gh-pages-refs
needs:
- job: subsystem-regression-tests
- job: subsystem-benchmark-availability-recovery
artifacts: true
- job: subsystem-benchmark-availability-distribution
artifacts: true
- job: publish-rustdoc
artifacts: false
@@ -109,7 +111,9 @@ trigger_workflow:
needs:
- job: publish-subsystem-benchmarks
artifacts: false
- job: subsystem-regression-tests
- job: subsystem-benchmark-availability-recovery
artifacts: true
- job: subsystem-benchmark-availability-distribution
artifacts: true
script:
- echo "Triggering workflow"