Struct Client 

pub struct Client { /* private fields */ }

Implementations

impl Client

pub fn new(cfg: Config<'_>) -> Result<Self, ClientError>

Creates a new Client instance that connects to the provided ClickHouse nodes.

Arguments

• cfg - Configuration with URLs, credentials, and TLS settings.

Errors

Returns ClientError if initialization fails, including problems with TLS or URL handling.

pub async fn stream_logs_by_fingerprint( &self, filter: QueryLogFilter, sender: Sender<QueryLog>, ) -> Result<(), ClientError>

Streams grouped query log data matching the specified filter, grouped by fingerprint (normalized_query_hash).

Useful for identifying query patterns and their cumulative impact across the system.

Arguments

• filter - Filter criteria (time range, user, etc.).
• sender - A Sender<QueryLog> to push results into a stream or channel.

Errors

Returns ClientError if the query fails, parsing fails, or sending on the channel fails.

ClickHouse schema dependency

This relies on the system.query_log table and expects ClickHouse to be configured to log queries.

pub async fn stream_log_by_fingerprint( &self, fingerprint: u64, filter: QueryLogFilter, sender: Sender<QueryLogExtended>, ) -> Result<(), ClientError>

Retrieves detailed query log metrics for a specific query fingerprint.

This method is intended for drill-down analysis of a known query fingerprint (normalized_query_hash). Unlike Self::stream_logs_by_fingerprint, which aggregates metrics across all fingerprints, this function focuses on a single query group, providing extended information.

Arguments

• fingerprint — The normalized_query_hash of the query group to inspect.
• filter — Optional additional filtering (e.g., time range, user).
• sender — A Sender<QueryLogExtended> to stream the result.

Returns

• Ok(()) — If the query completed successfully.
• Err(ClientError) — On ClickHouse query failure or channel send error.

ClickHouse schema dependency

Relies on the system.query_log table and assumes it includes normalized query hashes.

pub async fn stream_logs_total( &self, filter: QueryLogFilter, sender: Sender<QueryLogTotal>, ) -> Result<(), ClientError>

Streams total aggregated query log metrics matching the specified filter.

Unlike Self::stream_logs_by_fingerprint, this method does not group by query fingerprint. Instead, it aggregates metrics across all matching Select queries within the query_log for a given filter, returning a single total result.

Useful for high-level monitoring of cluster-wide query impact over a period of time.

Arguments

• filter - Filter criteria to restrict the aggregation scope (e.g., time range, user).
• sender - A Sender<QueryLogTotal> to deliver the aggregated result to consumers.

Returns

• Ok(()) - If the aggregation completed successfully.
• Err(ClientError) - On query failure, data parsing issue, or channel send error.

ClickHouse schema dependency

Relies on the system.query_log table with query profiling enabled.

pub async fn stream_error_by_code( &self, filter: ErrorFilter, sender: Sender<Error>, ) -> Result<(), ClientError>

Streams error statistics from ClickHouse’s system.errors table based on the provided filter.

Useful for monitoring the frequency and severity of runtime errors by error code.

Arguments

• filter - Filter criteria (e.g. time range, minimum count).
• sender - A Sender<Error> to stream the results.

Errors

Returns ClientError for query or channel failures.

ClickHouse schema dependency

Requires that system.errors is populated (i.e., error collection is enabled in ClickHouse).