Server settings in log4net.config

QPR ProcessAnalyzer uses the log4net component (https://logging.apache.org/log4net/) for logging, allowing to flexibly define what is logged and where the logs are stored.

Write logs to file

For example, the following settings can be set in the log4net.config file (located in the bin folder in the server application files):

• log4net > appender > maximumFileSize: Maximum size of the log file. When the maximum size is reached, the log writing continues in a new file, and the old file is archived. The default setting is 100MB. (https://logging.apache.org/log4net/log4net-1.2.13/release/sdk/log4net.Appender.RollingFileAppender.MaximumFileSize.html)
• log4net > appender > maxSizeRollBackups: Maximum number of archived log files that are stored. When the maximum file size is reached and new file is created, the oldest archived log file is removed if the maximum number of archived files is reached. The default setting is 1. (http://logging.apache.org/log4net/log4net-1.2.15/release/sdk/html/P_log4net_Appender_RollingFileAppender_MaxSizeRollBackups.htm)
• log4net > appender > lockingModel: By default, the server log works in the ExclusiveLock mode, i.e. the log file is reserved by the QPR ProcessAnalyzer process for the entire time QPR ProcessAnalyzer is running. Thus it's not possible to edit or delete log files during that time.

Send logs to OpenTelemetry

Traces

QPR ProcessAnalyzer can write traces and uses the OpenTelemetry (https://opentelemetry.io/docs) exporter to write them into the provided endpoint.

There are the following standard traces written:

• _Microsoft.EntityFrameworkCore_ to trace EntityFramework queries.
• _Microsoft.AspNetCore_ to trace QPR ProcessAnalyzer API calls.

In addition, there are the following custom traces:

• _QPR.ProcessAnalyzer.ExpressionQuery_ to trace expression query evaluation.
• _QPR.ProcessAnalyzer.DatabaseOperation_ to trace queries to external data source.
• _QPR.ProcessAnalyzer.ExpressionLanguage_ to trace calls to expression language function and property calls.

If an endpoint is not configured, traces are not collected and exported.

Metrics

QPR ProcessAnalyzer can write write metrics and uses the OpenTelemetry (https://opentelemetry.io/docs) exporter to write them into the provided endpoint.

There are the following standard metrics are written:

• _Microsoft.AspNetCore_ to track QPR ProcessAnalyzer API calls.
• _Microsoft.EntityFrameworkCore_ to track EntityFramework status.
• _Microsoft.Data.SqlClient.EventSource_ to track SqlClient status.

If an endpoint is not configured, metrics are not collected and exported.

Logs

QPR ProcessAnalyzer can write write logs and uses the OpenTelemetry (https://opentelemetry.io/docs) exporter to write them into the provided endpoint. If an endpoint is not configured, logs are not exported.

The following environment variables are used to control collecting and exporting telemetry data using OpenTelemetry Protocol Exporter (https://opentelemetry.io/docs/specs/otel/protocol/exporter):

• OTEL_EXPORTER_OTLP_ENDPOINT sends all signals to the same collector.
• OTEL_EXPORTER_OTLP_TRACES_ENDPOINT collector endpoint for traces.
• OTEL_EXPORTER_OTLP_METRICS_ENDPOINT collector endopint for metrics.
• OTEL_EXPORTER_OTLP_LOGS_ENDPOINT collector endpoint for logs.

QPR ProcessAnalyzer writes logs with level WARN by default. Logs with level WARN, ERROR, and FATAL additionally written into Snowflake events table utilizing event's severity. Logs are written into container's stdout stream using JSON format for structured logging (https://learn.microsoft.com/en-us/dotnet/core/extensions/console-log-formatter#json).

OpenTelemetry in Snowflake Native App

Writing logs is controlled by the LOG_LEVEL option for METADATA schema. If it is not set, than WARN level used. See [Snowflake documentation](https://docs.snowflake.com/en/sql-reference/parameters#label-log-level) for details. To take new `LOG_LEVEL` into use service must be restarted.

Application writes traces into [endpoint, provided by Snowflake](https://docs.snowflake.com/en/developer-guide/snowpark-container-services/monitoring-services#publishing-otlp-application-metrics-and-traces). Writing traces controlled by `TRACE_LEVEL` option for `METADATA` schema. If it is not set, than `ALWAYS` level used. See [Snowflake documentation](https://docs.snowflake.com/en/sql-reference/parameters#label-trace-level) for details.

To take new `TRACE_LEVEL` into use service must be restarted. Application writes metrics into [endpoint, provided by Snowflake](https://docs.snowflake.com/en/developer-guide/snowpark-container-services/monitoring-services#publishing-otlp-application-metrics-and-traces).

[Snowflake metrics](https://docs.snowflake.com/en/developer-guide/snowpark-container-services/monitoring-services#label-spcs-available-platform-metrics) that should be written:

• _system_ to track CPU and memory usage.
• _system_limits_ to check CPU and memory limits for container.
• _storage_ to track storage volume used by SQL Server.

Writing metrics controlled by `METRIC_LEVEL` option for `METADATA` schema. If it is not set, than `ALL` level used. See [Snowflake documentation](https://docs.snowflake.com/en/sql-reference/parameters#label-metric-level) for details. To take new `METRIC_LEVEL` into use service must be restarted