diff --git a/.github/workflows/molecule.yml b/.github/workflows/molecule.yml index c21c6d87..71049694 100644 --- a/.github/workflows/molecule.yml +++ b/.github/workflows/molecule.yml @@ -17,7 +17,7 @@ on: description: JSON array of Elastic major versions timeout: type: number - default: 45 + default: 60 max-parallel: type: number default: 10 diff --git a/molecule/shared/create.yml b/molecule/shared/create.yml index 28b4900b..cf9fbc90 100644 --- a/molecule/shared/create.yml +++ b/molecule/shared/create.yml @@ -130,14 +130,16 @@ REMOTE_SCRIPT changed_when: true register: _launch_result - # 30 retries × 30s = 15 min. Deliberately shorter than the workflow - # timeout (45 min for the standard molecule.yml call, 120 min for + # 90 retries × 30s = 45 min. Deliberately shorter than the workflow + # timeout (60 min for the standard molecule.yml call, 180 min for # full_stack) so the task fails loudly with the last "No capacity: # …" stdout instead of being silently cancelled by the workflow # timeout — that made the queue-starvation cases very hard to # diagnose (see PR investigating the elasticsearch_roles_calculation - # / elasticstack_default post-outage hangs). - retries: 30 + # / elasticstack_default post-outage hangs). Bumped from 30 after a + # sustained multi-PR CI storm evicted every heavy scenario on the + # first attempt. + retries: 90 delay: 30 until: _launch_result.rc == 0