kevinveenbirkenbachandClaude Opus 5
b09f2cbd82
Outside compose TOR_CONTAINER_DNS_HOST was empty, so the drop-in rendered no
listen-address and lookup('container_dns') handed containers the clearnet
resolver alone. A swarm stack host ran the egress router without a listener its
containers could reach, and .onion stayed unresolvable for every one of them
while compose resolved it fine. Ownership already governs the listener, the
upstream policy and the resolv.conf pin; the deploy mode only ever stood in for
it. With the mode clause gone both modes render the drop-in compose has been
running in production, and compose keeps its exact behaviour: where matched == 0
the old and the new expression are the same boolean.
In CI run 31927586060 the swarm node dnsmasq logged "ignoring query from
non-local network 172.18.0.16" seconds after a veth appeared and vanished on
docker0, and every container lookup timed out from then on. The health service
retried for 77 minutes against a resolver that no longer answered.
Measured on dnsmasq 2.93 before landing this: a listen-address disables
--local-service even when that flag comes from the init script; without a bind
directive dnsmasq takes the 0.0.0.0 and [::] wildcard pair, which would collide
with a systemd-resolved stub and occupy port 53 on every address; bind-dynamic
and bind-interfaces both hold their per-address listeners across 2000 veth
cycles; and the stock dnsmasq.conf on arch, fedora and centos carries no
bind-interfaces, so the drop-in's bind-dynamic cannot abort startup there.
Two expectations in tests/unit/plugins/lookup/test_container_dns.py still encode
the compose-only asymmetry and fail with this change.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Ansible Plugins (plugins/) 🧩
This document is the central reference for plugin types and plugin placement in Infinito.Nexus.
Not to be confused with application addons. This
plugins/directory holds Ansible plugins (action / filter / lookup) that extend the automation engine itself. Some applications also call their own extensions "plugins" (e.g. Nextcloud, WordPress). Those application-level extensions, whatever the app calls them natively, are declared through the unifiedmeta/addons/contract, documented in Unified Addons. An app "plugin" is an addon; an Ansible plugin lives here.
Current Project Layout 🗂️
plugins/
├── action/ # custom action plugins
├── filter/ # custom Jinja2 filters
└── lookup/ # custom lookup plugins
Configured plugin paths in this repository are defined in ansible.cfg:
action_plugins = ./plugins/actionfilter_plugins = ./plugins/filterlookup_plugins = ./plugins/lookup
All Common Ansible Plugin Types 📚
The table below summarizes common Ansible plugin types and their conventional directory names.
| Emoji | Plugin Type | Conventional Directory | Purpose |
|---|---|---|---|
| ⚙️ | action |
action_plugins/ or plugins/action/ |
Wrap/extend module execution behavior at task runtime. |
| 🔐 | become |
become_plugins/ |
Implement privilege escalation methods. |
| 💾 | cache |
cache_plugins/ |
Persist gathered facts and cached data between runs. |
| 📣 | callback |
callback_plugins/ |
React to play/task events (logging, reporting, notifications). |
| 🖧 | cliconf |
cliconf_plugins/ |
Define CLI command transport behavior (mainly network platforms). |
| 🔌 | connection |
connection_plugins/ |
Control how Ansible connects to targets (ssh, local, etc.). |
| 🔍 | filter |
filter_plugins/ or plugins/filter/ |
Add Jinja2 filters for data transformation. |
| 🌐 | httpapi |
httpapi_plugins/ |
API transport layer over HTTP(S), mainly for network devices. |
| 🧭 | inventory |
inventory_plugins/ |
Build dynamic inventories from APIs, services, or files. |
| 👉 | lookup |
lookup_plugins/ or plugins/lookup/ |
Resolve dynamic values during playbook processing. |
| 🛰️ | netconf |
netconf_plugins/ |
NETCONF transport integration for network automation. |
| 🐚 | shell |
shell_plugins/ |
Adjust shell command wrapping/execution behavior. |
| 🧠 | strategy |
strategy_plugins/ |
Customize task scheduling/execution strategy. |
| ✅ | test |
test_plugins/ |
Add Jinja2 tests used in conditions and templates. |
| 📦 | vars |
vars_plugins/ |
Load variables from custom sources. |
Which Plugin Type to Use 🧭
action ⚙️
Use when task execution behavior itself must be changed or wrapped.
Examples:
- centralized retry logic (
uri_retry.py), - argument normalization before module dispatch,
- pre/post task orchestration.
filter 🔍
Use for pure data transformation in Jinja2 expressions.
Examples:
- mapping/normalizing dictionaries,
- deriving computed fields for templates,
- deterministic formatting helpers.
lookup 👉
Use when a value must be dynamically resolved at runtime.
Examples:
- computing IDs from role metadata,
- resolving paths/version data,
- reading structured project data for templates/tasks.
callback 📣
Use to hook into execution events and shape output/reporting.
inventory 🧭
Use when host/group data should come from a dynamic source.
strategy 🧠
Use to alter how tasks are scheduled/executed across hosts.
vars 📦
Use to load variables from custom backends.
connection, become, shell 🔌🔐🐚
Use to customize low-level execution and privilege mechanics.
cache 💾
Use to persist facts/results across runs for performance and consistency.
cliconf, httpapi, netconf 🖧🌐🛰️
Use in network automation contexts requiring protocol-specific transport behavior.
test ✅
Use to add custom Jinja2 test operators for conditional logic.
Best Practices 🧪
- Keep plugin interfaces explicit and stable.
- Keep plugin logic focused; move shared logic to
utils/. - Add unit tests for plugin behavior and integration tests for end-to-end flows.
- Prefer deterministic outputs and avoid hidden global state.
- Document any plugin-specific task arguments in the plugin docstring and README.
References 🔗
- Ansible plugin development: docs.ansible.com
- Ansible plugin index (all types): docs.ansible.com
- Ansible
ansible-docplugin types:ansible-doc -t ...