Slavi Pantaleev d6bf789710 Remove matrix_container_global_registry_prefix variable
This is done for a few reasons:

- less globals and more indepdendence for each role is better. We rely
  on various externally-hosted roles and they don't rely on this global
  either.

- `matrix_container_global_registry_prefix` could make people think they
  could just override this variable and have all their images pull from
  elsewhere. This is rarely the case, unless you've taken special care
  to mirror all the various components (from their respective
  registries) to your own. In such a case, you probably know what you're
  mirroring and can adjust individual variables.

- nowadays, various components live on different registries.
  With Docker Inc tightening rate limits for Docker Hub, it's even more
  likely that we'll see increased diversity in where images are hosted
2025-02-23 10:15:41 +02:00

101 lines
7.3 KiB
YAML

---
- name: Fail if invalid homeserver implementation
ansible.builtin.fail:
msg: "You need to set a valid homeserver implementation in `matrix_homeserver_implementation`"
when: "matrix_homeserver_implementation not in ['synapse', 'dendrite', 'conduit', 'conduwuit']"
- name: (Deprecation) Catch and report renamed settings
ansible.builtin.fail:
msg: >-
Your configuration contains a variable, which now has a different name.
Please change your configuration to rename the variable (`{{ item.old }}` -> `{{ item.new }}`).
when: "item.old in vars"
with_items:
- {'old': 'host_specific_hostname_identity', 'new': 'matrix_domain'}
- {'old': 'hostname_identity', 'new': 'matrix_domain'}
- {'old': 'hostname_matrix', 'new': 'matrix_server_fqn_matrix'}
- {'old': 'hostname_riot', 'new': 'matrix_server_fqn_element'}
- {'old': 'matrix_server_fqn_riot', 'new': 'matrix_server_fqn_element'}
- {'old': 'matrix_local_bin_path', 'new': '<there is no global bin path anymore - each role has its own>'}
- {'old': 'matrix_client_element_e2ee_default', 'new': 'matrix_static_files_file_matrix_client_property_io_element_e2ee_default'}
- {'old': 'matrix_client_element_e2ee_secure_backup_required', 'new': 'matrix_static_files_file_matrix_client_property_io_element_e2ee_secure_backup_required'}
- {'old': 'matrix_client_element_e2ee_secure_backup_setup_methods', 'new': 'matrix_static_files_file_matrix_client_property_io_element_e2ee_secure_backup_setup_methods'}
- {'old': 'matrix_container_global_registry_prefix', 'new': '<no global variable anymore; you need to override the `_registry_prefix` variable in each component separately>'}
# We have a dedicated check for this variable, because we'd like to have a custom (friendlier) message.
- name: Fail if matrix_homeserver_generic_secret_key is undefined
ansible.builtin.fail:
msg: |
The `matrix_homeserver_generic_secret_key` variable must be defined and have a non-null and non-empty value.
If you're observing this error on a new installation, you should ensure that the `matrix_homeserver_generic_secret_key` is defined.
If you think you've defined it, but are still getting this error, then it's likely that you have a typo
in your domain name in `inventory/hosts` or in one of the directories leading up to your `inventory/host_vars/matrix.example.com/vars.yml` file.
If you're observing this error on an existing homeserver installation, you can fix it easily and in a backward-compatible way by adding
`{% raw %}matrix_homeserver_generic_secret_key: "{{ matrix_synapse_macaroon_secret_key }}"{% endraw %}`
to your `vars.yml` file. Using another secret value for the new variable is also possible and shouldn't cause any trouble.
when: "matrix_homeserver_generic_secret_key is none or matrix_homeserver_generic_secret_key == ''"
- name: Fail if required variables are undefined
ansible.builtin.fail:
msg: "The `{{ item.var }}` variable must be defined and have a non-null and non-empty value"
with_items:
- {'var': matrix_domain, 'value': "{{ matrix_domain | default('') }}"}
- {'var': matrix_server_fqn_matrix, 'value': "{{ matrix_server_fqn_matrix | default('') }}"}
- {'var': matrix_server_fqn_element, 'value': "{{ matrix_server_fqn_element | default('') }}"}
- {'var': matrix_homeserver_container_client_api_endpoint, 'value': "{{ matrix_homeserver_container_client_api_endpoint | default('') }}"}
- {'var': matrix_homeserver_container_federation_api_endpoint, 'value': "{{ matrix_homeserver_container_federation_api_endpoint | default('') }}"}
- {'var': matrix_architecture, 'value': "{{ matrix_architecture | default('') }}"}
when: "item.value is none or item.value == ''"
- name: Fail if matrix_architecture is set incorrectly
ansible.builtin.fail:
msg: "Detected that variable matrix_architecture {{ matrix_architecture }} appears to be set incorrectly. See docs/alternative-architectures.md. Server appears to be {{ ansible_architecture }}."
when: matrix_architecture not in ['amd64', 'arm32', 'arm64']
- name: Fail if matrix_playbook_reverse_proxy_type is set incorrectly
ansible.builtin.fail:
msg: |
You have not explicitly specified the type of reverse-proxy you'd like to use (with a `matrix_playbook_reverse_proxy_type` variable in your configuration).
The playbook used to default to nginx (matrix-nginx-proxy), but things are changing with Traefik being the new default reverse-proxy.
Learn more here: https://github.com/spantaleev/matrix-docker-ansible-deploy/blob/master/CHANGELOG.md#traefik-is-the-default-reverse-proxy-now
when: matrix_playbook_reverse_proxy_type == ''
- name: Fail if matrix_playbook_reverse_proxy_type is set incorrectly
ansible.builtin.fail:
msg: "Detected that variable matrix_playbook_reverse_proxy_type (current value: `{{ matrix_playbook_reverse_proxy_type }}`) appears to be set incorrectly. See roles/custom/matrix-base/defaults/main.yml for valid choices."
when: matrix_playbook_reverse_proxy_type not in ['playbook-managed-traefik', 'other-traefik-container', 'none']
- name: Fail if uppercase domain used
ansible.builtin.fail:
msg: "Detected that you're using an uppercase domain name - `{{ item }}`. This will cause trouble. Please use all-lowercase!"
with_items:
- "{{ matrix_domain }}"
- "{{ matrix_server_fqn_matrix }}"
- "{{ matrix_server_fqn_element }}"
when: "item != item | lower"
- name: Fail if encountering usage of removed role (mx-puppet-skype)
ansible.builtin.fail:
msg: >-
Your configuration seems to include a reference to `matrix_mx_puppet_skype_enabled`. Are you trying to install the mx-puppet-skype bridge?
The playbook no longer includes a role for installing mx-puppet-skype, because the mx-puppet-bridge is unmaintained and has been reported as broken for a long time.
To get rid of this error, remove all `matrix_mx_puppet_*` references from your configuration.
To clean up your server from mx-puppet-skype's presence, see this changelog entry: https://github.com/spantaleev/matrix-docker-ansible-deploy/blob/master/CHANGELOG.md#mx-puppet-skype-removal.
If you still need bridging to Skype, consider switching to the go-skype bridge instead. See `docs/configuring-playbook-bridge-go-skype-bridge.md`.
when: "'matrix_mx_puppet_skype_enabled' in vars"
- name: Fail if mautrix-instagram and mautrix-meta-instagram are in conflict
ansible.builtin.fail:
msg: >-
Your configuration enables both the old mautrix-instagram bridge and the new mautrix-meta-instagram bridge.
By default, both bridges are configured to use the same bridge bot username (`@{{ matrix_mautrix_meta_instagram_appservice_username }}:{{ matrix_domain }}`) which is a conflict.
We recommend that you disable at least one of the bridges (preferrably the old mautrix-instagram bridge), or to resolve the conflict in another way.
To resolve the conflict without disabling a bridge, consider adjusting one of `matrix_mautrix_instagram_appservice_bot_username` or `matrix_mautrix_meta_instagram_appservice_username` - they both have a value of {{ matrix_mautrix_meta_instagram_appservice_username }} right now.
when:
- matrix_mautrix_instagram_enabled | bool
- matrix_mautrix_meta_instagram_enabled | bool
- matrix_mautrix_instagram_appservice_bot_username == matrix_mautrix_meta_instagram_appservice_username