How to set up asynchronous logging in spdlog?

Initialize a thread pool with spdlog::init_thread_pool, then create async loggers using spdlog::basic_logger_mt or similar methods. The README provides examples with configurable queue sizes and overflow policies like block or overrun.

Is spdlog faster than glog?

spdlog is optimized for speed with benchmarks showing higher throughput, often millions of logs per second. However, glog might offer better integration with Google ecosystems, so choose based on performance needs versus existing tooling.

Can spdlog log to rotating files?

Yes, use spdlog::rotating_logger_mt with parameters for max file size and count, as shown in the rotating_example where logs rotate after reaching 5 MB with up to 3 backup files.

How to change log levels at runtime in spdlog?

Use spdlog::set_level for global changes or per-logger methods, and load levels from environment variables with spdlog::cfg::load_env_levels, allowing dynamic adjustments without recompilation.

Does spdlog support custom log formats?

Yes, set custom patterns via spdlog::set_pattern and add user-defined flags with spdlog::custom_flag_formatter, as demonstrated in the custom_flags_example for tailored output.

How to handle errors in spdlog?

Implement custom error handlers with spdlog::set_error_handler to manage failures like file write errors, with examples in the err_handler_example showing callback-based notifications.

Open-Awesome

spdlog

NOASSERTIONC++v1.17.0

A fast, header-only C++ logging library with rich formatting and asynchronous support.

GitHub

29.2k stars5.3k forks0 contributors

What is spdlog?

spdlog is a fast, feature-rich logging library for C++ applications. It provides both synchronous and asynchronous logging with support for multiple output targets, custom formatting, and runtime log level control. It solves the need for a high-performance, easy-to-integrate logging solution in modern C++ projects.

Target Audience

C++ developers building applications that require efficient, flexible logging, including system software, embedded systems, and high-performance services.

Value Proposition

Developers choose spdlog for its exceptional speed, header-only simplicity, extensive feature set, and seamless integration with the fmt library for advanced formatting.

Overview

Fast C++ logging library.

Use Cases

Best For

High-performance C++ applications requiring minimal logging overhead
Systems needing asynchronous logging to avoid blocking main threads
Projects requiring custom log sinks (e.g., Qt widgets, syslog, rotating files)
Debugging complex applications with backtrace and runtime log level control
Cross-platform C++ development with consistent logging across OSes
Embedded or resource-constrained environments where header-only usage is preferred

Not Ideal For

Projects requiring zero external dependencies and using only the C++ standard library
Legacy codebases with C++ compilers older than C++11
Microcontroller environments where every kilobyte of flash memory is critical
Applications needing seamless multi-language logging without custom integration layers

Pros & Cons

Pros

Extreme Performance

Benchmarks in the README show millions of logs per second in both synchronous and asynchronous modes, making it ideal for high-throughput applications like system software.

Rich Formatting with fmt

Integrates with the fmt library for type-safe, powerful formatting, as demonstrated in usage samples with custom specifiers for integers, floats, and binary data.

Flexible Sink System

Supports console, files, syslog, Qt widgets, and custom sinks with easy extensibility, shown in multi-sink examples that allow different log levels and formats per target.

Asynchronous Logging

Optional asynchronous mode with configurable thread pools and overflow policies reduces blocking in main threads, though with trade-offs like MDC incompatibility.

Cons

External Dependency on fmt

Relies on the fmt library for formatting, which adds an extra dependency that must be managed and can complicate builds in environments with strict dependency controls.

Asynchronous Mode Limitations

Mapped Diagnostic Context (MDC) is not supported in asynchronous mode due to thread-local storage constraints, limiting advanced debugging features in high-performance scenarios.

Compilation Overhead in Header-Only Mode

Using the header-only version can significantly increase build times, as noted in the README's recommendation to use the compiled version for faster compiles.

Frequently Asked Questions

Related Projects

imgui

Dear ImGui: Bloat-free Graphical User interface for C++ with minimal dependencies

Stars74,986

Forks11,931

Last commit13 hours ago

C/C++

A curated list of awesome C++ (or C) frameworks, libraries, resources, and shiny things. Inspired by awesome-... stuff.

Stars72,398

Forks8,320

Last commit5 days ago

ProtoBuf

Protocol Buffers - Google's data interchange format

Stars71,621

Forks16,200

Last commit1 hour ago

JSON for Modern C++

Stars50,198

Forks7,437

Last commit10 hours ago

Community-curated · Updated weekly · 100% open source

Found a gem we're missing?

Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.

Submit a project Star on GitHub