influxdata/influxdb · error · ProcessingEngineError::PluginError

Plugin not found: {}

Error message

Plugin not found: {}

What it means

update_plugin() locates the plugin to update by iterating every database schema's processing_engine_triggers and matching trigger_name against the requested plugin_name; it rewrites the found plugin's __init__.py. If no trigger matches, it returns ProcessingEngineError::PluginError('Plugin not found: {plugin_name}'). Matching is exact (case-sensitive) on the trigger name, not the file name.

Source

Thrown at influxdb3_processing_engine/src/lib.rs:1120

                    async_fs::write(&plugin_path, content).await.map_err(|e| {
                        ProcessingEngineError::PluginError(plugins::PluginError::ReadPluginError(e))
                    })?;

                    return Ok(db_schema.name.to_string());
                }

                // For multi-file plugins (directories), update __init__.py by default
                let init_file = plugin_path.join(INIT_PY);
                async_fs::write(&init_file, content).await.map_err(|e| {
                    ProcessingEngineError::PluginError(plugins::PluginError::ReadPluginError(e))
                })?;

                return Ok(db_schema.name.to_string());
            }
        }

        Err(ProcessingEngineError::PluginError(
            plugins::PluginError::AnyhowError(anyhow::anyhow!("Plugin not found: {}", plugin_name)),
        ))
    }

    /// Replace an entire plugin directory atomically with new files.
    pub async fn replace_plugin_directory(
        self: &Arc<Self>,
        plugin_name: &str,
        files: Vec<(String, String)>, // Vec of (relative_path, content)
    ) -> Result<String, ProcessingEngineError> {
        if self.environment_manager.plugin_dir_only {
            return Err(ProcessingEngineError::PluginError(
                PluginError::PluginInstallationDisabled,
            ));
        }

        // Find the trigger to get the plugin filename
        let (db_name, plugin_filename) = {
            let mut result = None;

View on GitHub (pinned to d28e26e048)

Solutions

1. List the installed plugin files via the HTTP endpoints (/api/v3/plugins/files, /api/v3/plugins/directory) and check the trigger names in your create-trigger command history
2. Use the exact, case-sensitive trigger name the trigger was created with
3. If the plugin genuinely does not exist, create it first, then update

Example fix

# before: trigger was created as 'my-scheduler'
PUT /api/v3/plugins/files?name=MyScheduler   # -> Plugin not found: MyScheduler

# after
PUT /api/v3/plugins/files?name=my-scheduler

# confirm the plugin file exists before updating it
curl -sf "${INFLUXDB3_HOST_URL}/api/v3/plugins/files" \
  | grep -q "\"${PLUGIN_NAME}\"" \
  || { echo "plugin '${PLUGIN_NAME}' not installed" >&2; exit 1; }

When it happens

Trigger: Calling the plugin update endpoint/method with a name that does not exactly equal an installed trigger's name — a typo, different casing, or using the plugin filename (e.g. 'my_plugin.py') where the trigger name (e.g. 'my-plugin') is required. Also updating a plugin that was never created.

Common situations: Renaming plugins after creation; automating update flows where the name comes from a config that drifted from the installed trigger name; environments with several databases where the operator assumed a global name.

Related errors

• No plugin directory configured
• Failed to rename temp directory to target
• request trigger execution cancelled
• invalid database name: {db_name}
• unsupported Arrow type for Python conversion: {:?}. Supporte

AI-assisted analysis of influxdata/influxdb@d28e26e048 (2026-08-16). Data as JSON: /api/errors/07eeb215d75bfce1. Report an issue: GitHub.