nicolargo/glances · warning · HTTPException

Unknown plugin {plugin} (available plugins: {self.plugins_li

Error message

Unknown plugin {plugin} (available plugins: {self.plugins_list})

What it means

Raised as HTTP 400 when the requested plugin name is not in the server's plugins_list. This is a client error: the endpoint exists but the plugin is unknown or not loaded. The message lists all available plugins to guide correction.

Source

Thrown at glances/outputs/glances_restful_api.py:1013

        """
        self._check_if_plugin_available(plugin)

        # Update the stat
        self.__update_stats(get_plugin_dependencies(plugin))

        try:
            # Get the RAW value of the stat ID
            statval = self.stats.get_plugin(plugin).get_api()
        except Exception as e:
            raise HTTPException(status.HTTP_404_NOT_FOUND, f"Cannot get plugin {plugin} ({str(e)})")

        return GlancesJSONResponse(statval)

    def _check_if_plugin_available(self, plugin: str) -> None:
        if plugin in self.plugins_list:
            return

        raise HTTPException(
            status.HTTP_400_BAD_REQUEST, f"Unknown plugin {plugin} (available plugins: {self.plugins_list})"
        )

    def _api_top(self, plugin: str, nb: int = 0):
        """Glances API RESTful implementation.

        Return the JSON representation of a given plugin limited to the top nb items.
        It is used to reduce the payload of the HTTP response (example: processlist).

        HTTP/200 if OK
        HTTP/400 if plugin is not found
        HTTP/404 if others error
        """
        self._check_if_plugin_available(plugin)

        # Update the stat
        self.__update_stats(get_plugin_dependencies(plugin))

View on GitHub (pinned to a240d8dfb3)

Solutions

1. GET /api/4/plugins first and use an exact name from the returned list
2. Check the 400 body — it enumerates available plugins
3. Re-enable the plugin by removing it from --disable-plugin or adding it to --enable-plugin
4. Verify the plugin is supported on your OS/hardware

Example fix

// before
GET /api/4/gpu
// after
GET /api/4/plugins   // confirm name, then
GET /api/4/gpu

AVAILABLE = set(httpx.get(f'{BASE}/api/4/plugins').json())
if plugin not in AVAILABLE:
    raise ValueError(f'{plugin} not in {sorted(AVAILABLE)}')

def is_known_plugin(name: str, available: list[str]) -> bool:
    return isinstance(name, str) and name in available

try:
    r = httpx.get(f'{BASE}/api/4/{plugin}')
    r.raise_for_status()
except httpx.HTTPStatusError as e:
    if e.response.status_code == 400:
        available = e.response.json()['detail'].split('available plugins: ')[-1]
        raise ValueError(f'unknown plugin, choose from {available}') from e
    raise

When it happens

Trigger: GET /api/4/<plugin> (and /top, /history, /limits, /item, /key variants) with a misspelled plugin name, a plugin disabled via --disable-plugin, or a plugin unavailable on the platform (e.g., 'gpu' with no NVIDIA hardware).

Common situations: Clients hardcoding plugin names from a different glances version, typos like 'mem' vs 'memswap', or querying platform-specific plugins on unsupported hosts.

Related errors

• Unknown configuration item {section}
• Unknown argument item {item}
• Invalid JSON body
• Missing username or password in request body
• Cannot get help view data ({str(e)})

AI-assisted analysis of nicolargo/glances@a240d8dfb3 (2026-08-27). Data as JSON: /api/errors/ca8b3df05b67fd18. Report an issue: GitHub.