Skip to main content

Usage and Logs

While Platform Insights gives you a real-time snapshot, the Usage and Logs section provides historical analysis tools for tracking trends, diagnosing issues, and auditing connector behavior over specific time periods.


Usage Monitor

The Usage Monitor lets you examine platform metrics across configurable date ranges. Navigate to Insights & Monitoring > Usage Monitor to access it.

Selecting a Time Period

Use the date range picker at the top of the page to define your analysis window.

PresetRange
Last 24 HoursRolling 24-hour window from the current time.
Last 7 DaysThe previous 7 calendar days.
Last 30 DaysThe previous 30 calendar days.
CustomSelect a start date and end date manually.

Available Metrics

The metric dropdown lets you select which measurement to chart. The following metrics are available.

MetricUnitDescription
Query CountCountTotal number of queries executed across all users and data sources.
Query RuntimeMinutesCumulative execution time for all queries.
Active UsersCountUnique users with at least one interaction per day.
Data TransferredMB / GBVolume of data returned from Edge nodes to the Coordinator.
Error RatePercentageProportion of queries that returned an error response.
P95 LatencyMillisecondsThe 95th-percentile query latency, indicating typical worst-case response time.

Filtering and Grouping

You can filter and group usage data to isolate specific patterns.

FilterOptions
Data SourceSelect one or more connected data sources.
UserFilter by individual user or group.
Query TypeSELECT, INSERT, UPDATE, DELETE, or DDL.
StatusSuccess, Error, or Timeout.

Group by Data Source, User, or Day to segment the chart into separate series for comparison.

tip

To identify your most resource-intensive queries, filter by Query Runtime and group by User. This reveals which users or workflows consume the most compute.


Connector Log

The Connector Log provides a detailed, searchable record of every interaction between Edge nodes and your data sources. Navigate to Insights & Monitoring > Connector Log.

Log Entry Fields

Each log entry contains the following fields.

FieldDescription
TimestampThe exact date and time of the event (in your workspace timezone).
ConnectorThe data source connector that handled the event (e.g., snowflake-prod).
Event TypeThe type of activity: query, schema_refresh, connection_test, sync.
StatusThe outcome: success, error, timeout.
DurationHow long the operation took, in milliseconds.
UserThe user who initiated the action (or system for automated operations).
Query PreviewA truncated preview of the executed SQL statement (first 200 characters).
Error DetailIf the status is error, the full error message from the data source.

Searching and Filtering

The log supports full-text search and column-level filters.

  • Search bar: Enter keywords to search across all fields (e.g., a table name, error keyword, or user email).
  • Connector dropdown: Filter to a specific data source.
  • Status dropdown: Show only successes, errors, or timeouts.
  • Date range: Narrow the log to a specific time window.

Exporting Logs

You can export the filtered log view for offline analysis or compliance auditing.

FormatDescription
CSVComma-separated values. Compatible with spreadsheets and data tools.
JSONStructured format suitable for ingestion into log aggregation systems.

Click the Export button in the top-right corner of the log panel. The export respects your current filters and date range.

note

Connector logs are retained for 90 days by default. Contact your workspace administrator to adjust the retention period if your compliance requirements differ.


Interpreting Common Patterns

PatternPossible CauseRecommended Action
Spike in error rateData source maintenance or credential expiration.Check the connector log for specific error messages and verify source availability.
Gradual increase in P95 latencyGrowing data volumes or inefficient queries.Review slow queries and consider indexing or query optimization.
Drop in active usersSeasonal variation or access issues.Cross-reference with user management to check for deactivated accounts.
High data transfer volumeLarge result sets or frequent full-table scans.Add filters or pagination to reduce result set sizes.

Next Steps