From 4294957e68ab9c81fdb1216b9b32e273cb60d3c0 Mon Sep 17 00:00:00 2001 From: Guillaume Abrioux Date: Fri, 11 Sep 2026 08:28:38 +0200 Subject: [PATCH 1/2] mergify: require Jenkins CI before merging The default queue only requires a github review approval. That let Mergify merge while Jenkins jobs were still running if the PR is already up to date. Let's wait for the PR pipeline statuses and ignore PRs labeled DNM. Signed-off-by: Guillaume Abrioux --- .mergify.yml | 13 +++++++++++++ CONTRIBUTING.md | 4 +++- 2 files changed, 16 insertions(+), 1 deletion(-) diff --git a/.mergify.yml b/.mergify.yml index b2a5cfd3..7226227e 100644 --- a/.mergify.yml +++ b/.mergify.yml @@ -1,3 +1,16 @@ +queue_rules: + - name: default + # GitHub branch protection currently injects only github-review-approved. + # Require the Jenkins statuses reported by cephadm-ansible-prs so a PR + # that is already up-to-date cannot skip CI and merge while jobs are pending. + merge_conditions: + - 'check-success="Testing: flake8"' + - 'check-success="Testing: mypy"' + - 'check-success="Testing: unittests"' + - 'check-success="Testing: el9-functional"' + - 'check-success="Testing: el10-functional"' + - label != DNM + pull_request_rules: - name: backport tentacle conditions: diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 45a71a2f..1b528bde 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -38,7 +38,9 @@ Merging PR is controlled by [mergify](https://mergify.io/) by the following rules: - at least one approval from a maintainer -- a SUCCESS from the CI pipeline "cephadm-ansible PR Pipeline" +- SUCCESS from every Jenkins status reported by the cephadm-ansible PR + pipeline (`Testing: flake8`, `Testing: mypy`, `Testing: unittests`, + `Testing: el9-functional`, `Testing: el10-functional`) If your work is not ready for review and merging, request the `DNM` (**D**o **N**ot **M**erge) label via a comment or the title of your PR. This will From 661904c126cd4bbfd1c0874c3090fcde7a6389a7 Mon Sep 17 00:00:00 2001 From: Guillaume Abrioux Date: Fri, 11 Sep 2026 08:51:57 +0200 Subject: [PATCH 2/2] tests: give functional VMs 2G of RAM 1G is not enough for the functional libvirt nodes. Bump MEMORY to 2048 so the VMs stop getting killed mid run. Signed-off-by: Guillaume Abrioux --- tests/functional/Vagrantfile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/functional/Vagrantfile b/tests/functional/Vagrantfile index 5874903c..54199be4 100644 --- a/tests/functional/Vagrantfile +++ b/tests/functional/Vagrantfile @@ -6,7 +6,7 @@ NNODE = 6 NCLIENTS = 2 LABEL_PREFIX = 'ceph-' -MEMORY = 1024 +MEMORY = 2048 PUBLIC_SUBNET = '192.168.9' CLUSTER_SUBNET = '192.168.10' BOX = ENV['CEPH_ANSIBLE_VAGRANT_BOX']