opentelemetry.sdk._logs package

Submodules

• opentelemetry.sdk._logs.export
  • BatchLogRecordProcessor
    • BatchLogRecordProcessor.on_emit()
    • BatchLogRecordProcessor.shutdown()
    • BatchLogRecordProcessor.force_flush()
  • ConsoleLogExporter
  • ConsoleLogRecordExporter
    • ConsoleLogRecordExporter.export()
    • ConsoleLogRecordExporter.shutdown()
    • ConsoleLogRecordExporter.force_flush()
  • LogExporter
  • LogRecordExporter
    • LogRecordExporter.export()
    • LogRecordExporter.shutdown()
    • LogRecordExporter.force_flush()
  • LogExportResult
    • LogExportResult.SUCCESS
    • LogExportResult.FAILURE
  • LogRecordExportResult
    • LogRecordExportResult.SUCCESS
    • LogRecordExportResult.FAILURE
  • SimpleLogRecordProcessor
    • SimpleLogRecordProcessor.on_emit()
    • SimpleLogRecordProcessor.shutdown()
    • SimpleLogRecordProcessor.force_flush()
  • InMemoryLogExporter
  • InMemoryLogRecordExporter
    • InMemoryLogRecordExporter.clear()
    • InMemoryLogRecordExporter.get_finished_logs()
    • InMemoryLogRecordExporter.export()
    • InMemoryLogRecordExporter.shutdown()
    • InMemoryLogRecordExporter.force_flush()

Module contents

class opentelemetry.sdk._logs.ConcurrentMultiLogRecordProcessor(max_workers=2)

Bases: LogRecordProcessor

Implementation of LogRecordProcessor that forwards all received events to a list of log processors in parallel.

Calls to the underlying log processors are forwarded in parallel by submitting them to a thread pool executor and waiting until each log processor finished its work.

Parameters:

max_workers (int) – The number of threads managed by the thread pool executor and thus defining how many log processors can work in parallel.

add_log_record_processor(log_record_processor)

on_emit(log_record)

Emits the ReadWriteLogRecord.

Implementers should handle any exceptions raised during log processing to prevent application crashes. See the class docstring for details on error handling expectations.

Return type:

None

shutdown()

Called when a opentelemetry.sdk._logs.Logger is shutdown

Return type:

None

force_flush(timeout_millis=30000)

Force flush the log processors in parallel.

Parameters:

timeout_millis (int) – The maximum amount of time to wait for logs to be exported.

Return type:

bool

Returns:

True if all the log processors flushes the logs within timeout, False otherwise.

class opentelemetry.sdk._logs.Logger(resource, multi_log_record_processor, instrumentation_scope, *, logger_metrics, _logger_config)

Bases: Logger

property instrumentation_scope

property resource

emit(record=None, *, timestamp=None, observed_timestamp=None, context=None, severity_number=None, severity_text=None, body=None, attributes=None, event_name=None, exception=None)

Emits the ReadWriteLogRecord by setting instrumentation scope and forwarding to the processor.

Return type:

None

class opentelemetry.sdk._logs.LoggerProvider(resource=None, shutdown_on_exit=True, multi_log_record_processor=None, *, meter_provider=None, _logger_configurator=None)

Bases: LoggerProvider

property resource

get_logger(name, version=None, schema_url=None, attributes=None)

Returns a Logger for use by the given instrumentation library.

For any two calls with identical parameters, it is undefined whether the same or different Logger instances are returned.

This function may return different Logger types (e.g. a no-op logger vs. a functional logger).

Parameters:

• name (str) –

  The name of the instrumenting module, package or class. This should not be the name of the module, package or class that is instrumented but the name of the code doing the instrumentation. E.g., instead of "requests", use "opentelemetry.instrumentation.requests".

  For log sources which define a logger name (e.g. logging.Logger.name) the Logger Name should be recorded as the instrumentation scope name.

• version (Optional[str]) – Optional. The version string of the instrumenting library. Usually this should be the same as importlib.metadata.version(instrumenting_library_name).

• schema_url (Optional[str]) – Optional. Specifies the Schema URL of the emitted telemetry.

• attributes (Optional[Mapping[str, AnyValue]]) – Optional. Specifies the instrumentation scope attributes to associate with emitted telemetry.

Return type:

Logger

add_log_record_processor(log_record_processor)

Registers a new LogRecordProcessor for this LoggerProvider instance.

The log processors are invoked in the same order they are registered.

shutdown()

Shuts down the log processors.

Return type:

None

force_flush(timeout_millis=30000)

Force flush the log processors.

Parameters:

timeout_millis (int) – The maximum amount of time to wait for logs to be exported.

Return type:

bool

Returns:

True if all the log processors flushes the logs within timeout, False otherwise.

class opentelemetry.sdk._logs.LoggingHandler(level=0, logger_provider=None)

Bases: Handler

A handler class which writes logging records, in OTLP format, to a network destination or file. Supports signals from the logging module. https://docs.python.org/3/library/logging.html

emit(record)

Emit a record. Skip emitting if logger is NoOp.

The record is translated to OTel format, and then sent across the pipeline.

Return type:

None

flush()

Flushes the logging output. Skip flushing if logging_provider has no force_flush method.

Return type:

None

class opentelemetry.sdk._logs.LogLimits(max_attributes=None, max_attribute_length=None, max_log_record_attributes=None, max_log_record_attribute_length=None)

Bases: LogRecordLimits

class opentelemetry.sdk._logs.LogRecordLimits(max_attributes=None, max_attribute_length=None, max_log_record_attributes=None, max_log_record_attribute_length=None)

Bases: object

This class is based on a SpanLimits class in the Tracing module.

This class represents the limits that should be enforced on recorded data such as events, links, attributes etc.

This class does not enforce any limits itself. It only provides a way to read limits from env, default values and from user provided arguments.

All limit arguments must be either a non-negative integer or None.

• All limit arguments are optional.

• If a limit argument is not set, the class will try to read its value from the corresponding environment variable.

• If the environment variable is not set, the default value, if any, will be used.

Limit precedence:

• If a model specific limit is set, it will be used.

• Else if the corresponding global limit is set, it will be used.

• Else if the model specific limit has a default value, the default value will be used.

• Else if the global limit has a default value, the default value will be used.

Parameters:

• max_attributes (Optional[int]) – Maximum number of attributes that can be added to a log record (global fallback). Environment variable: OTEL_ATTRIBUTE_COUNT_LIMIT Default: {_DEFAULT_OTEL_ATTRIBUTE_COUNT_LIMIT}

• max_attribute_length (Optional[int]) – Maximum length an attribute value can have (global fallback). Values longer than the specified length will be truncated.

• max_log_record_attributes (Optional[int]) – Maximum number of attributes that can be added to a log record. Environment variable: OTEL_LOGRECORD_ATTRIBUTE_COUNT_LIMIT Falls back to max_attributes when unset.

• max_log_record_attribute_length (Optional[int]) – Maximum length a log record attribute value can have. Environment variable: OTEL_LOGRECORD_ATTRIBUTE_VALUE_LENGTH_LIMIT Falls back to max_attribute_length when unset.

class opentelemetry.sdk._logs.LogRecordProcessor

Bases: ABC

Interface to hook the log record emitting action.

Log processors can be registered directly using LoggerProvider.add_log_record_processor() and they are invoked in the same order as they were registered.

Implementers of custom log processors should be aware of the following:

Error Handling

According to the OpenTelemetry error handling principles, the SDK should not throw unhandled exceptions at runtime. When implementing a custom LogRecordProcessor, it is the processor’s responsibility to handle any exceptions that may be raised by the exporter’s export() method.

The LogRecordExporter.export() method may raise exceptions (e.g., network errors, timeouts). If these exceptions are not caught, they will propagate up and potentially crash the application.

Custom processor implementations should wrap exporter calls in a try/except block. See SimpleLogRecordProcessor for a reference implementation:

def on_emit(self, log_record: ReadWriteLogRecord):
    try:
        self._exporter.export((log_record,))
    except Exception:  # pylint: disable=broad-exception-caught
        logger.exception("Exception while exporting logs.")

The BatchLogRecordProcessor handles this implicitly since export operations occur in a background thread where exceptions cannot bubble up to the caller.

abstract on_emit(log_record)

Emits the ReadWriteLogRecord.

Implementers should handle any exceptions raised during log processing to prevent application crashes. See the class docstring for details on error handling expectations.

Return type:

None

abstract shutdown()

Called when a opentelemetry.sdk._logs.Logger is shutdown

Return type:

None

abstract force_flush(timeout_millis=30000)

Export all the received logs to the configured Exporter that have not yet been exported.

Parameters:

timeout_millis (int) – The maximum amount of time to wait for logs to be exported.

Return type:

bool

Returns:

False if the timeout is exceeded, True otherwise.

exception opentelemetry.sdk._logs.LogDroppedAttributesWarning(**kwargs)

Bases: LogRecordDroppedAttributesWarning

exception opentelemetry.sdk._logs.LogRecordDroppedAttributesWarning

Bases: UserWarning

Custom warning to indicate dropped log attributes due to limits.

This class is used to filter and handle these specific warnings separately from other warnings, ensuring that they are only shown once without interfering with default user warnings.

class opentelemetry.sdk._logs.ReadableLogRecord(log_record, resource, instrumentation_scope=None, limits=None)

Bases: object

Readable LogRecord should be kept exactly in-sync with ReadWriteLogRecord, only difference is the frozen=True param.

log_record: LogRecord

resource: Resource

instrumentation_scope: Optional[InstrumentationScope] = None

limits: Optional[LogRecordLimits] = None

property dropped_attributes: int

to_json(indent=4)

Return type:

str

class opentelemetry.sdk._logs.ReadWriteLogRecord(log_record, resource=<opentelemetry.sdk.resources.Resource object>, instrumentation_scope=None, limits=<factory>)

Bases: object

A ReadWriteLogRecord instance represents an event being logged. ReadWriteLogRecord instances are created and emitted via Logger every time something is logged. They contain all the information pertinent to the event being logged.

log_record: LogRecord

resource: Optional[Resource] = <opentelemetry.sdk.resources.Resource object>

instrumentation_scope: Optional[InstrumentationScope] = None

limits: LogRecordLimits

property dropped_attributes: int

class opentelemetry.sdk._logs.SynchronousMultiLogRecordProcessor

Bases: LogRecordProcessor

Implementation of class:LogRecordProcessor that forwards all received events to a list of log processors sequentially.

The underlying log processors are called in sequential order as they were added.

add_log_record_processor(log_record_processor)

Adds a Logprocessor to the list of log processors handled by this instance

Return type:

None

on_emit(log_record)

Emits the ReadWriteLogRecord.

Implementers should handle any exceptions raised during log processing to prevent application crashes. See the class docstring for details on error handling expectations.

Return type:

None

shutdown()

Shutdown the log processors one by one

Return type:

None

force_flush(timeout_millis=30000)

Force flush the log processors one by one

Parameters:

timeout_millis (int) – The maximum amount of time to wait for logs to be exported. If the first n log processors exceeded the timeout then remaining log processors will not be flushed.

Return type:

bool

Returns:

True if all the log processors flushes the logs within timeout, False otherwise.