This commit updates multiple roles to ensure compatibility with Ansible 2.20.
Several include paths and task-loading mechanisms required adjustments,
as Ansible 2.20 applies stricter evaluation rules for complex Jinja expressions
and no longer resolves certain relative include paths the way Ansible 2.18 did.
Key changes:
- Replaced legacy once_finalize.yml and once_flag.yml with the new structure
under tasks/utils/once/finalize.yml and tasks/utils/once/flag.yml.
- Updated all include_tasks statements to use 'path_join' with playbook_dir,
ensuring deterministic and absolute file resolution across roles.
- Fixed all network helper includes by converting direct relative paths such as
'roles/docker-compose/tasks/utils/network.yml' to proper Jinja-evaluated paths.
- Normalized MATOMO_* variable names for consistency with the updated variable
scope behavior in Ansible 2.20.
- Removed deprecated patterns that were implicitly supported in Ansible 2.18
but break under the more strict variable and path resolution model in 2.20.
These changes are part of the full migration step required to ensure the
infinito-nexus roles remain stable, deterministic, and forward-compatible with
Ansible 2.20.
Details of the discussion and reasoning can be found in this conversation:
https://chatgpt.com/share/69300a8d-24d4-800f-bec0-e895a695618a
- added new file roles/web-app-shopware/files/framework.yaml defining trusted_proxies and trusted_headers for Symfony
- mounted framework.yaml into /var/www/html/config/packages/ in docker-compose
- exposed new role vars SHOPWARE_FRAMEWORK_HOST/DOCKER for mounting path
- rendered framework.yaml via Ansible copy task with proper permissions
- adjusted env.j2 to set TRUSTED_PROXIES and TRUSTED_HOSTS dynamically from domains and networks
- added SHOPWARE_DOMAIN var to vars/main.yml
- removed inline framework.yaml creation from Dockerfile (now managed via mount)
- updated proxy template (html.conf.j2) to include X-Forwarded-Ssl header
- improved init.sh permission handling for shared volumes
See ChatGPT conversation for implementation details and rationale:
https://chatgpt.com/share/690d4fe7-2830-800f-8b6d-b868e7fe0e97
- Dynamically calculate PEERTUBE_MAX_OLD_SPACE_SIZE (~35% of container RAM, clamped between 768–3072 MB)
- Dynamically calculate PEERTUBE_TRANSCODING_CONCURRENCY (~½ vCPUs, min 1, max 8)
- Added default resource limits for Redis and Peertube containers
- Updated test suite to include human_to_bytes filter in built-in filter list
https://chatgpt.com/share/690914d2-6100-800f-a850-94e6d226e7c9
- Added 'font-src data:' to CSP whitelist to allow inline fonts in Admin UI
- Refactored init.sh to run as root only for volume permission setup, then drop privileges to www-data
- Unified all bash invocations to sh for POSIX compliance
- Added missing 'bundles' named volume and mount to Docker Compose
- Set init container to run as root (0:0) for permission setup
- Added admin user rename step via Ansible task
See discussion: https://chatgpt.com/share/69087361-859c-800f-862c-7413350cca3e
- moved init.sh from template to files/ for direct copying and bind mounting
- removed hardcoded user creation from init process
- added database emptiness check before running system:install
- added new task 03_admin.yml to ensure admin user exists and update password/email via Ansible
- switched docker exec shell from bash to sh for Alpine compatibility
- updated Dockerfile and docker-compose.yml accordingly for mount-based init script
- Added symfony/redis-messenger installation with ignored build-time PHP extension checks
- Installed php83-redis in runtime stage
- Ensured consistent network includes across all Shopware services in docker-compose template
- Improves compatibility with Redis-based Symfony Messenger transport during init phase
https://chatgpt.com/share/6908068e-0bb8-800f-8855-7b3913c57158
Changes include:
- Added OpenSearch configuration and variable definitions (image, version, heap, memory limits)
- Replaced legacy search/elasticsearch logic with OpenSearch integration
- Updated docker-compose template for OpenSearch with proper JVM heap and ulimits
- Ensured both OPENSEARCH_URL and ELASTICSEARCH_URL are set for compatibility
Reference: https://chatgpt.com/share/6907b0d4-ab14-800f-b576-62c0d26c8ad1
