Log a warning when the BatchSpanProcessor queue is full.

CHANGELOG.md

@@ -21,6 +21,9 @@ Increment the:
  cmake when using C++20 [#1852](https:/open-telemetry/opentelemetry-cpp/pull/1852)
 * [SEMANTIC CONVENTIONS] Upgrade to version 1.16.0
  [#1854](https:/open-telemetry/opentelemetry-cpp/pull/1854)
+* [SDK] BatchSpanProcessor now logs a warning when dropping a span because the
+ queue is full
+ [1871](https:/open-telemetry/opentelemetry-cpp/pull/1871)
 
 ## [1.8.1] 2022-12-04
 

sdk/src/trace/batch_span_processor.cc

@@ -3,6 +3,7 @@
 
 #include "opentelemetry/sdk/trace/batch_span_processor.h"
 #include "opentelemetry/common/spin_lock_mutex.h"
+#include "opentelemetry/sdk/common/global_log_handler.h"
 
 #include <vector>
 using opentelemetry::sdk::common::AtomicUniquePtr;
@@ -51,6 +52,7 @@ void BatchSpanProcessor::OnEnd(std::unique_ptr<Recordable> &&span) noexcept
 
  if (buffer_.Add(span) == false)
  {
+ OTEL_INTERNAL_LOG_WARN("BatchSpanProcessor queue is full - dropping span.");
  return;
  }
 

sdk/test/trace/batch_span_processor_test.cc

@@ -2,10 +2,12 @@
 // SPDX-License-Identifier: Apache-2.0
 
 #include "opentelemetry/sdk/trace/batch_span_processor.h"
+#include "opentelemetry/sdk/common/global_log_handler.h"
 #include "opentelemetry/sdk/trace/span_data.h"
 #include "opentelemetry/sdk/trace/tracer.h"
 
 #include <gtest/gtest.h>
+#include <algorithm>
 #include <chrono>
 #include <list>
 #include <memory>
@@ -182,10 +184,31 @@ TEST_F(BatchSpanProcessorTestPeer, TestForceFlush)
  }
 }
 
+// A mock log handler to check whether log messages with a specific level were emitted.
+struct MockLogHandler : public sdk::common::internal_log::LogHandler
+{
+ using Message = std::pair<sdk::common::internal_log::LogLevel, std::string>;
+
+ void Handle(sdk::common::internal_log::LogLevel level,
+ const char * /*file*/,
+ int /*line*/,
+ const char *msg,
+ const sdk::common::AttributeMap & /*attributes*/) noexcept override
+ {
+ messages.emplace_back(level, msg);
+ }
+
+ std::vector<Message> messages;
+};
+
 TEST_F(BatchSpanProcessorTestPeer, TestManySpansLoss)
 {
  /* Test that when exporting more than max_queue_size spans, some are most likely lost*/
 
+ // Set up a log handler to verify a warning is generated.
+ auto log_handler = nostd::shared_ptr<sdk::common::internal_log::LogHandler>(new MockLogHandler());
+ sdk::common::internal_log::GlobalLogHandler::SetLogHandler(log_handler);
+
  std::shared_ptr<std::atomic<bool>> is_shutdown(new std::atomic<bool>(false));
  std::shared_ptr<std::vector<std::unique_ptr<sdk::trace::SpanData>>> spans_received(
  new std::vector<std::unique_ptr<sdk::trace::SpanData>>);
@@ -211,6 +234,25 @@ TEST_F(BatchSpanProcessorTestPeer, TestManySpansLoss)
 
  // Span should be exported by now
  EXPECT_GE(max_queue_size, spans_received->size());
+
+ // If we haven't received all spans, some must have dropped, verify a warning was logged.
+ // Only do this when the log level is warning or above.
+#if OTEL_INTERNAL_LOG_LEVEL >= OTEL_INTERNAL_LOG_LEVEL_WARN
+ if (max_queue_size > spans_received->size())
+ {
+ auto &messages = static_cast<MockLogHandler *>(log_handler.get())->messages;
+ EXPECT_TRUE(
+ std::find(messages.begin(), messages.end(),
+ MockLogHandler::Message(sdk::common::internal_log::LogLevel::Warning,
+ "BatchSpanProcessor queue is full - dropping span.")) !=
+ messages.end());
+ }
+#endif
+
+ // Reinstate the default log handler.
+ sdk::common::internal_log::GlobalLogHandler::SetLogHandler(
+ nostd::shared_ptr<sdk::common::internal_log::LogHandler>(
+ new sdk::common::internal_log::DefaultLogHandler()));
 }
 
 TEST_F(BatchSpanProcessorTestPeer, TestManySpansLossLess)