lpereira/lwan

Lwan Web Server

Lwan is a high-performance & scalable web server.

The project web site contains more details.

Build status

Installing

You can either build Lwan yourself, use a container image, or grab a package from your favorite distribution.

Building

Before installing Lwan, ensure all dependencies are installed. All of them are common dependencies found in any GNU/Linux distribution; package names will be different, but it shouldn't be difficult to search using whatever package management tool that's used by your distribution.

Required dependencies

• CMake, at least version 2.8
• libdeflate, zlib-ng or ZLib

Optional dependencies

The build system will look for these libraries and enable/link if available.

• Lua 5.4
• Valgrind
• Brotli
  • Can be disabled by passing -DENABLE_BROTLI=NO
• ZSTD
  • Can be disabled by passing -DENABLE_ZSTD=NO
• On Linux builds, if -DENABLE_TLS=ON (default) is passed:
  • mbedTLS
• Alternative memory allocators can be used by passing -DUSE_ALTERNATIVE_MALLOC to CMake with the following values:
  • "mimalloc"
  • "jemalloc"
  • "tcmalloc"
  • "auto": Autodetect from the list above, falling back to libc malloc if none found
• To run test suite:
  • Python (2.6+) with Requests
  • Lua 5.1
• To run benchmark:
  • Weighttp -- bundled and built alongside Lwan for convenience
  • Matplotlib
• To build TechEmpower benchmark suite:
  • Client libraries for MariaDB
  • SQLite 3
• shared-mime-info may be used during build-time to augment the built-in extention/MIME type database.

Common operating system package names

Minimum to build

• ArchLinux: pacman -S cmake zlib-ng pkgconf
• FreeBSD: pkg install cmake pkgconf
• Debian and Ubuntu 14+: apt-get update && apt-get install git cmake zlib1g-dev pkg-config
• macOS: brew install cmake

Build with all optional features

• ArchLinux: pacman -S cmake zlib-ng pkgconf sqlite lua55 mariadb-libs gperftools valgrind mbedtls
• FreeBSD: pkg install cmake pkgconf sqlite3 lua55
• Debian and Ubuntu 14+: apt-get update && apt-get install git cmake zlib1g-dev pkg-config lua5.5-dev libsqlite3-dev libmariadb-dev libmbedtls-dev
• macOS: brew install cmake mariadb-connector-c sqlite [email protected] pkg-config

Build commands

Clone the repository

~$ git clone git://github.com/lpereira/lwan
~$ cd lwan

Create the build directory

~/lwan$ mkdir build
~/lwan$ cd build

Select build type

Selecting a release version (no debugging symbols, messages, enable some optimizations, etc):

~/lwan/build$ cmake .. -DCMAKE_BUILD_TYPE=Release

If you'd like to enable optimizations but still use a debugger, use this instead:

~/lwan/build$ cmake .. -DCMAKE_BUILD_TYPE=RelWithDebInfo

To disable optimizations and build a more debugging-friendly version:

~/lwan/build$ cmake .. -DCMAKE_BUILD_TYPE=Debug

Build Lwan

~/lwan/build$ make

This will generate a few binaries:

• src/bin/lwan/lwan: The main Lwan executable. May be executed with --help for guidance.
• src/bin/testrunner/testrunner: Contains code to execute the test suite (src/scripts/testsuite.py).
• src/samples/freegeoip/freegeoip: FreeGeoIP sample implementation. Requires SQLite.
• src/samples/techempower/techempower: Code for the TechEmpower Web Framework benchmark. Requires SQLite and MariaDB libraries.
• src/samples/clock/clock: Clock sample. Generates a never-ending animated GIF file that always shows the local time.
• src/samples/forthsalon/forthsalon: Generates a never-ending animated GIF from a program written in the Forth Salon dialect of the Forth programming language. In construction!
• src/samples/forthsalon/forth: Test harness for the Forth dialect used in the forthsalon sample.
• src/bin/tools/mimegen: Builds the extension-MIME type table. Used during the build process.
• src/bin/tools/bin2hex: Generates a C file from a binary file, suitable for use with #include. Used during the build process.
• src/bin/tools/configdump: Dumps a configuration file using the configuration reader API. Used for testing.
• src/bin/tools/weighttp: Rewrite of the weighttp HTTP benchmarking tool.
• src/bin/tools/statuslookupgen: Generates a perfect hash table for HTTP status codes and their descriptions. Used during the build process.
• src/lib/lwan-template-test: Benchmark for the template engine

Remarks

Passing -DCMAKE_BUILD_TYPE=Release will enable some compiler optimizations (such as LTO) and tune the code for current architecture.

[!IMPORTANT]

Please use the release build when benchmarking. The default is the Debug build, which not only logs all requests to the standard output, but does so while holding a lock, severely holding down the server.

The default build (i.e. not passing -DCMAKE_BUILD_TYPE=Release) will build a version suitable for debugging purposes. This version can be used under Valgrind (if its headers are present) and includes debugging messages that are stripped in the release version. Debugging messages are printed for each and every request.

On these builds, sanitizers can be enabled. To select which one to build Lwan with, specify one of the following options to the CMake invocation line:

• -DSANITIZER=ubsan selects the Undefined Behavior Sanitizer.
• -DSANITIZER=address selects the Address Sanitizer.
• -DSANITIZER=thread selects the Thread Sanitizer.

Alternative memory allocators can be selected as well. Lwan currently supports TCMalloc, mimalloc, and jemalloc out of the box. To use either one of them, pass -DALTERNATIVE_MALLOC=name to the CMake invocation line, using the names provided in the "Optional dependencies" section.

The -DUSE_SYSLOG=ON option can be passed to CMake to also log to the system log in addition to the standard output.

If you're building Lwan for a distribution, it might be wise to use the -DMTUNE_NATIVE=OFF option, otherwise the generated binary may fail to run on some computers.

TLS support is enabled automatically in the presence of a suitable mbedTLS installation on Linux systems with headers new enough to support kTLS, but can be disabled by passing -DENABLE_TLS=NO to CMake.

Tests

~/lwan/build$ make testsuite

This will compile the testrunner program and execute regression test suite in src/scripts/testsuite.py.

Benchmark

~/lwan/build$ make benchmark

This will compile testrunner and execute benchmark script src/scripts/benchmark.py.

Coverage

Lwan can also be built with the Coverage build type by specifying -DCMAKE_BUILD_TYPE=Coverage. This enables the generate-coverage make target, which will run testrunner to prepare a test coverage report with lcov.

Every commit in this repository triggers the generation of this report, and results are publicly available.

CMake options

These options are available to configure a build. Even if an option is set, it'll be checked if it can be used or not (e.g. third-party libraries need to be installed).

Running

Set up the server by editing the provided lwan.conf; the format is explained in details below.

[!NOTE]

Lwan will try to find a configuration file based in the executable name in the current directory; testrunner.conf will be used for the testrunner binary, lwan.conf for the lwan binary, and so on.

Configuration files are loaded from the current directory. If no changes are made to this file, running Lwan will serve static files located in the ./wwwroot directory. Lwan will listen on port 8080 on all interfaces.

Lwan will detect the number of CPUs, will increase the maximum number of open file descriptors and generally try its best to autodetect reasonable settings for the environment it's running on. Many of these settings can be tweaked in the configuration file, but it's usually a good idea to not mess with them.

[!TIP]

Optionally, the lwan binary can be used for one-shot static file serving without any configuration file. Run it with --help for help on that.

Configuration File

Format

Lwan uses a familiar key = value configuration file syntax. Comments are supported with the # character (similar to e.g. shell scripts, Python, and Perl). Nested sections can be created with curly brackets. Sections can be empty; in this case, curly brackets are optional.

some_key_name is equivalent to some key name in configuration files (as an implementation detail, code reading configuration options will only be given the version with underscores).

[!TIP]

Values can contain environment variables. Use the syntax ${VARIABLE_NAME}. Default values can be specified with a colon (e.g. ${VARIABLE_NAME:foo}, which evaluates to ${VARIABLE_NAME} if it's set, or foo otherwise).

sound volume = 11 # This one is 1 louder

playlist metal {
   files = '''
	/multi/line/strings/are/supported.mp3
	/anything/inside/these/are/stored/verbatim.mp3
   '''
}

playlist chiptune {
   files = """
	/if/it/starts/with/single/quotes/it/ends/with/single/quotes.mod
	/but/it/can/use/double/quotes.s3m
   """
}

Some examples can be found in lwan.conf and techempower.conf.

Constants

Constants can be defined and reused throughout the configuration file by specifying them in a constants section anywhere in the configuration file. A constant will be available only after that section defines a particular constant. Constants can be re-defined. If a constant isn't defined, its value will be obtained from an environment variable. If it's not defined in either one constants section, or in the environment, Lwan will abort with an appropriate error message.

constants {
    user_name = ${USER}
    home_directory = ${HOME}
    buffer_size = 1000000
}

The same syntax for default values specified above is valid here (e.g. specifying user_name to be ${USER:nobody} will set ${user_name} to nobody if ${USER} isn't set in the environment variable or isn't another constant.)

Value types

Time Intervals

Time fields can be specified using multipliers. Multiple can be specified, they're just added together; for instance, "1M 1w" specifies "1 month and 1 week" (37 days). The following table lists all known multipliers:

[!NOTE]

No spaces must exist between the number and its multiplier. Multipliers outside the ones in the table above will cause parsing errors.

Boolean Values

Global Settings

It's generally a good idea to let Lwan decide the best settings for your environment. However, not every environment is the same, and not all uses can be decided automatically, so some configuration options are provided.

Variables for error_template

Straitjacket

Lwan can drop its privileges to a user in the system, and limit its filesystem view with a chroot. While not bulletproof, this provides a first layer of security in the case there's a bug in Lwan.

In order to use this feature, declare a straitjacket (or straightjacket) section, and set some options. This requires Lwan to be executed as root.

Although this section can be written anywhere in the file (as long as it is a top level declaration), if any directories are open, due to e.g. instantiating the serve_files module, Lwan will refuse to start. (This check is only performed on Linux as a safeguard for malconfiguration.)

[!TIP]

Declare a Straitjacket right before a site section in such a way that configuration files and private data (e.g. TLS keys) are out of reach of the server after initialization has taken place.

Headers

If there's a need to specify custom headers for each response, one can declare a headers section in the global scope. The order which this section appears isn't important.

For example, this declaration:

headers {
	Server = Apache/1.0.0 or nginx/1.0.0 (at your option)
	Some-Custom-Header = ${WITH_THIS_ENVIRONMENT_VARIABLE}
}

Will both override the Server header (Server: lwan won't be sent), and set Some-Custom-Header with the value obtained from the environment variable $WITH_THIS_ENVIRONMENT_VARIABLE.

Some headers can't be overridden, as that would cause issues when sending their actual values while servicing requests. These include but is not limited to:

• Date
• Expires
• WWW-Authenticate
• Connection
• Content-Type
• Transfer-Encoding
• All Access-Control-Allow- headers

[!NOTE]

Header names are also case-insensitive (and case-preserving). Overriding SeRVeR will override the Server header, but send it the way it was written in the configuration file.

Listeners

Only two listeners are supported per Lwan process: the HTTP listener (listener section), and the HTTPS listener (tls_listener section). Only one listener of each type is allowed.

[!WARNING]

TLS support is experimental. Although it is stable during initial testing, your mileage may vary. Only TLSv1.2 is supported at this point, but TLSv1.3 is planned.

[!NOTE]

TLS support requires :penguin: Linux with the tls.ko module built-in or loaded. Support for other operating systems may be added in the future. FreeBSD seems possible, other operating systems do not seem to offer similar feature. For unsupported operating systems, using a TLS terminator proxy such as Hitch is a good option.

For both listener and tls_listener sections, the only parameter is the the interface address and port to listen on. The listener syntax is ${ADDRESS}:${PORT}, where ${ADDRESS} can either be * (binding to all interfaces), an IPv6 address (if surrounded by square brackets), an IPv4 address, or a hostname. For instance, listener localhost:9876 would listen only in the lo interface, port 9876.

While a listener section takes no keys, a tls_listener section requires two: cert and key (each pointing, respectively, to the location on disk where the TLS certificate and private key files are located) and takes an optional boolean hsts key, which controls if Strict-Transport-Security headers will be sent on HTTPS responses.

[!TIP]

To generate these keys for testing purposes, the OpenSSL command-line tool can be used like the following: openssl req -nodes -x509 -newkey rsa:4096 -keyout key.pem -out cert.pem -sha256 -days 7

[!NOTE]

It's recommended that a Straitjacket with a chroot option is declared right after a tls_listener section, in such a way that the paths to the certificate and key are out of reach from that point on.

If systemd socket activation is used, systemd can be specified as a parameter. (If multiple listeners from systemd are specified, systemd:FileDescriptorName can be specified, where FileDescriptorName follows the conventions set in the systemd.socket documentation.)

Examples:

listener *:8080		# Listen on all interfaces, port 8080, HTTP

tls_listener *:8081 {	# Listen on all interfaces, port 8081, HTTPS
	cert = /path/to/cert.pem
	key = /path/to/key.pem
}

# Use named systemd socket activation for HTTP listener
listener systemd:my-service-http.socket

# Use named systemd socket activation for HTTPS listener
tls_listener systemd:my-service-https.socket {
	...
}

Site

A site section groups instances of modules and handlers that will respond to requests to a given URL prefix.

Routing URLs Using Modules or Handlers

In order to route URLs, Lwan matches the largest common prefix from the request URI with a set of prefixes specified in the listener section. How a request to a particular prefix will be handled depends on which handler or module has been declared in the listener section. Handlers and modules are similar internally; handlers are merely functions and hold no state, and modules holds state ("instance"). Multiple instances of a module can appear in a listener section.

There is no special syntax to attach a prefix to a handler or module; all the configuration parser rules apply here. Use ${NAME} ${PREFIX} to link the ${PREFIX} prefix path to either a handler named ${NAME} (if ${NAME} begins with &, as with C's "address of" operator), or a module named ${NAME}. Empty sections can be used here.

Each module will have its specific set of options, and they're listed in the next sections. In addition to configuration options, a special authorization section can be present in the declaration of a module instance. Handlers do not take any configuration options, but may include the authorization section.

[!TIP]

Executing Lwan with the --version command-line argument will show a list of built-in modules and handlers.

The following is some basic documentation for the modules shipped with Lwan.

File Serving

The serve_files module will serve static files, and automatically create directory indices or serve pre-compressed files. It'll generally try its best to serve files in the fastest way possible according to some heuristics.

[!NOTE]

Files smaller than 16KiB will be compressed in RAM for the duration specified in the cache_for setting. Lwan will always try to compress with deflate, and will optionally compress with Brotli and zstd (if Lwan has been built with proper support).

In cases where compression wouldn't be worth the effort (e.g. adding the Content-Encoding header would result in a larger response than sending the uncompressed file, usually the case for very small files), Lwan won't spend time compressing a file.

For files larger than 16KiB, Lwan will not attempt to compress them. In future versions, it might do this and send responses using chunked-encoding while the file is being compressed (up to a certain limit, of course), but for now, only precompressed files (see serve_precompressed_path setting in the table above) are considered.

For all cases, Lwan might try using the gzipped version if that's found in the filesystem and the client requested this encoding.

Variables for directory_list_template

Lua

The lua module will allow requests to be serviced by scripts written in the Lua programming language. Although the functionality provided by this module is quite spartan, it's able to run frameworks such as Sailor.

Scripts can be served from files or embedded in the configuration file, and the results of loading them and the standard Lua modules will be cached for a while (see cache_period below).

Writing request handlers

[!NOTE]

Lua scripts can't use global variables, as they may be not only serviced by different threads, but the state will be available only for the amount of time specified in the cache_period configuration option. This is because each I/O thread in Lwan will create an instance of a Lua VM (i.e. one lua_State struct for every I/O thread), and each Lwan coroutine will spawn a Lua thread (with lua_newthread()) per request.

There's no need to have one instance of the Lua module for each endpoint; a single script, embedded in the configuration file or otherwise, can service many different endpoints. Scripts are supposed to implement functions with the following signature: handle_${METHOD}_${ENDPOINT}(req), where ${METHOD} can be a HTTP method (i.e. get, post, head, etc.), and ${ENDPOINT} is the desired endpoint to be handled by that function. A generic handle(req) function will be called if the specific version doesn't exist.

[!TIP]

Use the root endpoint for a catchall. For example, the handler function handle_get_root() will be called if no other handler could be found for that request. If no catchall is specified, the server will return a 404 Not Found error.

The req parameter points to a metatable that contains methods to obtain information from the request, or to set the response, as seen below:

• req:query_param(param) returns the query parameter (from the query string) with the key param, or nil if not found
• req:post_param(param) returns the post parameter (only for ${POST} handlers) with the key param, or nil if not found
• req:set_response(str) sets the response to the string str
• req:say(str) sends a response chunk (using chunked encoding in HTTP)
• req:send_event(event, str) sends an event (using server-sent events)
• req:cookie(param) returns the cookie named param, or nil is not found
• req:set_headers(tbl) sets the response headers from the table tbl; a header may be specified multiple times by using a table, rather than a string, in the table value ({'foo'={'bar', 'baz'}}); must be called before sending any response with say() or send_event()
• req:header(name) obtains the header from the request with the given name or nil if not found
• req:sleep(ms) pauses the current handler for the specified amount of milliseconds
• req:ws_upgrade() returns 1 if the connection could be upgraded to a WebSocket; 0 otherwise
• req:ws_write_text(str) sends str through the WebSocket-upgraded connection as text frame
• req:ws_write_binary(str) sends str through the WebSocket-upgraded connection as binary frame
• req:ws_write(str) sends str through the WebSocket-upgraded connection as text or binary frame, depending on content containing only ASCII characters or not
• req:ws_read() returns a string with the contents of the last WebSocket frame, or a number indicating an status (ENOTCONN/107 on Linux if it has been disconnected; EAGAIN/11 on Linux if nothing was available; ENOMSG/42 on Linux otherwise). The return value here might change in the future for something more Lua-like.
• req:remote_address() returns a string with the remote IP address.
• req:path() returns a string with the request path.
• req:query_string() returns a string with the query string (empty string if no query string present).
• req:body() returns the request body (POST/PUT requests).
• req:request_id() returns a string containing the request ID.
• req:request_date() returns the date as it'll be written in the Date response header.
• req:is_https() returns true if this request is serviced through HTTPS, false otherwise.
• req:host() returns the value of the Host header if present, otherwise nil.
• req:http_version() returns HTTP/1.0 or HTTP/1.1 depending on the request version.
• req:http_method() returns a string, in uppercase, with the HTTP method (e.g. "GET").
• req:http_headers() returns a table with all headers and their values.

Handler functions may return either nil (in which case, a 200 OK response is generated), or a number matching an HTTP status code. Attempting to return an invalid HTTP status code or anything other than a number or nil will result in a 500 Internal Server Error response being thrown.

Logging

In addition to the metamethods in the req parameter, one can also log messages with different logging levels by calling methods from Lwan.log:

• Lwan.log:warning(str)
• Lwan.log:info(str)
• Lwan.log:error(str)
• Lwan.log:critical(str) (Will also abort Lwan! Use with caution)
• Lwan.log:debug(str) (Only available in debug builds; no-op otherwise)

[!NOTE]

If Lwan is built with syslog support, these messages will also be sent to the system log, otherwise they'll be printed to the standard error.

Rewrite

The rewrite module will match patterns in URLs and give the option to either redirect to another URL, or rewrite the request in a way that Lwan will handle the request as if it were made in that way originally.

[!NOTE]

Forked from Lua 5.3.1, the regular expresion engine may not be as feature-packed as most general-purpose engines, but has been chosen specifically because it is a deterministic finite automaton in an attempt to make some kinds of denial of service attacks impossible.

The new URL can be specified using a simple text substitution syntax, or use Lua scripts.

[!TIP]

Lua scripts will contain the same metamethods available in the req metatable provided by the Lua module, so it can be quite powerful.

Each instance of the rewrite module will require a pattern and the action to execute when such pattern is matched. Patterns are evaluated in the order they appear in the configuration file, and are specified using nested sections in the configuration file. For instance, consider the following example, where two patterns are specified:

rewrite /some/base/endpoint {
    pattern posts/(%d+) {
        # Matches /some/base/endpointposts/2600 and /some/base/endpoint/posts/2600
        rewrite_as = /cms/view-post?id=%1
    }
    pattern imgur/(%a+)/(%g+) {
        # Matches /some/base/endpointimgur/gif/mpT94Ld and /some/base/endpoint/imgur/gif/mpT94Ld
        redirect_to = https://i.imgur.com/%2.%1
    }
}

This example defines two patterns, one providing a nicer URL that's hidden from the user, and another providing a different way to obtain a direct link to an image hosted on a popular image hosting service (i.e. requesting /some/base/endpoint/imgur/mp4/4kOZNYX will redirect directly to a resource in the Imgur service).

The value of rewrite_as or redirect_to can be Lua scripts as well; in which case, the option expand_with_lua must be set to true, and, instead of using the simple text substitution syntax as the example above, a function named handle_rewrite(req, captures) has to be defined instead. The req parameter is documented in the Lua module section; the captures parameter is a table containing all the captures, in order (i.e. captures[2] is equivalent to %2 in the simple text substitition syntax). This function returns the new URL to redirect to.

This module has no options by itself. Options are specified in each and every pattern.

redirect_to and rewrite_as options are mutually exclusive, and one of them must be specified at least.

It's also possible to specify conditions to trigger a rewrite. To specify one, open a condition block, specify the condition type, and then the parameters for that condition to be evaluated. Multiple conditions can be set per rewrite rule as long as there's one condition per type:

Can use subst. syntax refers to the ability to reference the matched pattern using the same substitution syntax used for the rewrite as or redirect to actions. For instance, condition cookie { some-cookie-name = foo-%1-bar } will substitute %1 with the first match from the pattern this condition is related to.

[!NOTE]

Conditions that do not require a section have to be written as a key; for instance, condition has_query_string = yes.

For example, if one wants to send site-dark-mode.css if there is a style cookie with the value dark, and send site-light-mode.css otherwise, one can write:

pattern site.css {
   rewrite as = /site-dark-mode.css
   condition cookie { style = dark }
}
pattern site.css {
   rewrite as = /site-light-mode.css
}

Another example: if one wants to send pre-compressed files if they do exist in the filesystem and the user requested them:

pattern (%g+) {
   condition encoding { brotli = yes }
   condition stat { path = %1.brotli }
   rewrite as = %1.brotli
}
pattern (%g+) {
   condition encoding { gzip = yes }
   condition stat { path = %1.gzip }
   rewrite as = %1.gzip
}
pattern (%g+) {
   condition encoding { zstd = yes }
   condition stat { path = %1.zstd }
   rewrite as = %1.zstd
}
pattern (%g+) {
   condition encoding { deflate = yes }
   condition stat { path = %1.deflate }
   rewrite as = %1.deflate
}

[!NOTE]

In general, this is not necessary, as the file serving module will do this automatically and pick the smallest file available for the requested encoding, but this shows it's possible to have a similar feature by configuration alone.

Redirect

The redirect module will, as it says in the tin, generate a 301 Moved permanently (by default; the code can be changed, see below) response, according to the options specified in its configuration. Generally, the rewrite module should be used instead as it packs more features; however, this module serves also as an example of how to write Lwan modules (less than 100 lines of code).

If the to option is not specified, it always generates a 500 Internal Server Error response. Specifying an invalid HTTP code, or a code that Lwan doesn't know about (see enum lwan_http_status), will produce a 301 Moved Permanently response.

Response

The response module will generate an artificial response of any HTTP code. In addition to also serving as an example of how to write a Lwan module, it can be used to carve out voids from other modules (e.g. generating a 405 Not Allowed response for files in /.git, if / is served with the serve_files module).

If the supplied code falls outside the response codes known by Lwan, a 404 Not Found error will be sent instead.

FastCGI

The fastcgi module proxies requests between the HTTP client connecting to Lwan and a FastCGI server accessible by Lwan. This is useful, for instance, to serve pages from a scripting language such as PHP.

[!NOTE]

This is a preliminary version of this module, and as such, it's not well optimized, some features are missing, and some values provided to the environment are hardcoded.

Authorization Section

Authorization sections can be declared in any module instance or handler, and provides a way to authorize the fulfillment of that request through the standard HTTP authorization mechanism. In order to require authorization to access a certain module instance or handler, declare an authorization section with a basic parameter, and set one of its options.

[!WARNING]

Not only passwords are stored in clear text in a file that should be accessible by the server, they'll be kept in memory for a few seconds. Avoid using this feature if possible.

Hacking

Please read this section (and follow it) if you're planning on contributing to Lwan. There's nothing unexpected here; this mostly follows the rules and expectations of many other FOSS projects, but every one expects things a little bit different from one another.

Coding Style

Lwan tries to follow a consistent coding style throughout the project. If you're considering contributing a patch to the project, please respect this style by trying to match the style of the surrounding code. In general:

• global_variables_are_named_like_this, even though they tend to be rare and should be marked as static (with rare exceptions)
• Local variables are usually shorter, e.g. local_var, i, conn
• Struct names are often as short as they're descriptive. typedef for structs are rarely used in Lwan
• Header files should use #pragma once instead of the usual include guard hackery
• Functions that are used between .c files but are not APIs to be exposed to liblwan should have their prototype added to lwan-private.h
• Functions should be short and sweet. Exceptions may apply
• Public functions should be prefixed with lwan_
• Public types should be prefixed with lwan_
• Private functions must be static, and can be named without the lwan_ prefix
• Code is indented with 4 spaces; don't use tabs
• There's a suggested line break at column 80, but it's not enforced
• /* Old C-style comments are preferred */
• clang-format can be used to format the source code in an acceptable way; a .clang-format file is provided

Tests

If modifying well-tested areas of the code (e.g. the event loop, HTTP parser, etc.), please add a new integration test and make sure that, before you send a pull request, all tests (including the new ones you've sent) are working. Tests can be added by modifying src/scripts/testsuite.py, and executed by either invoking that script directly from the source root, or executing the testsuite build target.

Some tests will only work on Linux, and won't be executed on other platforms.

Fuzz-testing

Lwan is automatically fuzz-tested by OSS-Fuzz. To fuzz-test locally, though, one can follow the instructions to test locally.

Currently, there are fuzzing drivers for the request parsing code, the configuration file parser, the template parser, and the Lua string pattern matching library used in the rewrite module.

Adding new fuzzers is trivial:

• Fuzzers are implemented in C++ and the sources are placed in src/bin/fuzz.
• Fuzzers should be named ${FUZZER_NAME}_fuzzer.cc. Look at the OSS-Fuzz documentation and other fuzzers on information about how to write these.
• These files are not compiled by the Lwan build system, but rather by the build scripts used by OSS-Fuzz. To test your fuzzer, please follow the instructions to test locally, which will build the fuzzer in the environment they'll be executed in.
• A fuzzing corpus has to be provided in src/fuzz/corpus. Files have to be named corpus-${FUZZER_NAME}-${UNIQUE_ID}.

Exporting APIs

The shared object version of liblwan on ELF targets (e.g. Linux) will use a symbol filter script to hide symbols that are considered private to the library. Please edit src/lib/liblwan.sym to add new symbols that should be exported to liblwan.so.

Using Git and Pull Requests

Lwan tries to maintain a source history that's as flat as possible, devoid of merge commits. This means that pull requests should be rebased on top of the current master before they can be merged; sometimes this can be done automatically by the GitHub interface, sometimes they need some manual work to fix conflicts. It is appreciated if the contributor fixes these conflicts when asked.

It is advisable to push your changes to your fork on a branch-per-pull request, rather than pushing to the master branch; the reason is explained below.

Please ensure that Git is configured properly with your name (it doesn't really matter if it is your legal name or a nickname, but it should be enough to credit you) and a valid email address. There's no need to add Signed-off-by lines, even though it's fine to send commits with them.

If a change is requested in a pull request, you have two choices:

• Reply asking for clarification. Maybe the intentions were not clear enough, and whoever asked for changes didn't fully understand what you were trying to achieve
• Fix the issue. When fixing issues found in pull requests, please use interactive rebases to squash or fixup commits; don't add your fixes on top of your tree. Do not create another pull request just to accomodate the changes. After rewriting the history locally, force-push to your PR branch; the PR will update automatically with your changes. Rewriting the history of development branches is fine, and force-pushing them is normal and expected

It is not enforced, but it is recommended to create smaller commits. How commits are split in Lwan is pretty much arbitrary, so please take a look at the commit history to get an idea on how the division should be made. Git offers a plethora of commands to achieve this result: the already mentioned interactive rebase, the -p option to git add, and git commit --amend are good examples.

Commit messages should have one line of summary (~72 chars), followed by an empty line, followed by paragraphs of 80-char lines explaining the change. The paragraphs explaining the changes are usually not necessary if the summary is good enough. Try to write good commit messages.

Licensing

Lwan is licensed under the GNU General Public License, version 2, or (at your option), any later version. Therefore:

• Code must be either LGPLv2.1, GPLv2, a permissive "copyfree" license that is compatible with GPLv2 (e.g. MIT, BSD 3-clause), or public domain code (e.g. CC0)
• Although the program can be distributed and used as if it were licensed as GPLv3, its code must be compatible with GPLv2 as well; no new code can be licensed under versions of GPL newer than 2
• Likewise, code licensed under licenses compatible with GPLv3 but incompatible with GPLv2 (e.g. Apache 2) are not suitable for inclusion in Lwan
• Even if the license does not specify that credit should be given (e.g. CC0-licensed code), please give credit to the original author for that piece of code
• Contrary to popular belief, it is possible to use a GPL'd piece of code on a server without having to share the code for your application. It is only when the binary of that server is shared that source must be available to whoever has that binary. Merely accessing a Lwan server through HTTP does not qualify as having access to the binary program that's running on the server
• When in doubt, don't take legal advice from a README file: please consult a lawyer that understands free software licensing

AI and LLM use

The Lwan project will not accept any work partially or completely generated by AI tools; this includes, but does not limit to code, documentation, commit messages, bug reports, and issues.

Don't insist.

Portability

While Lwan was written originally for Linux, it has been ported to BSD systems as well. The build system will detect the supported features and build support library functions as appropriate.

For instance, epoll has been implemented on top of kqueue, and Linux-only syscalls and GNU extensions have been implemented for the supported systems. This blog post explains the details and how #include_next is used.

Performance

It can achieve good performance, yielding about 320000 requests/second on a Core i7 laptop for requests without disk access, and without pipelining.

When disk I/O is required, for files up to 16KiB, it yields about 290000 requests/second; for larger files, this drops to 185000 requests/second, which isn't too shabby either.

These results, of course, with keep-alive connections, and with weighttp running on the same machine (and thus using resources that could be used for the webserver itself).

Without keep-alive, these numbers drop around 6-fold.

IRC Channel

There is an IRC channel (#lwan) on Libera. A standard IRC client can be used.

Lwan in the wild

Here's a non-definitive list of third-party stuff that uses Lwan and have been seen in the wild. If you see mentions of Lwan in the media or academia, however small it might be, please contact the author! It'll make her day!

• This project uses Cython and Lwan to make it possible to write handlers in Python.
• An experimental version of Node.js using Lwan as its HTTP server is maintained by @raadad.
• The beginnings of a C++11 web framework based on Lwan written by @vileda.
• A more complete C++14 web framework by @matt-42 offers Lwan as one of its backends.
• A word ladder sample program by @sjnam. Demo.
• A Shodan search listing some brave souls that expose Lwan to the public internet.
• This write-up shows the use of Lwan on a Capture the Flag competition.

Some other distribution channels were made available as well:

• Container images are available from the GitHub Container Registry. More information below.
• A Dockerfile is maintained by @jaxgeller, and is available from the Docker registry.
• A buildpack for Heroku is maintained by @bherrera, and is available from its repo.
• Lwan is also available as a package in Biicode.
• It's also available in some GNU/Linux distributions:
  • Arch Linux
  • Ubuntu
  • Alpine Linux
  • NixOS
• It's also available as a package for the Nanos unikernel.

Lwan has been also used as a benchmark:

• Raphael Javaux's master thesis cites Lwan in chapter 5 ("Performance Analysis").
• Lwan is used as a benchmark by the PyParallel author.
• Kong uses Lwan as the backend API in its benchmark.
• TechEmpower Framework benchmarks feature Lwan since round 10.
• KrakenD used Lwan for the REST API in all official benchmarks
• Effective System Call Aggregation (ESCA) project uses Lwan as one of the benchmarks; they claim that Lwan throughput improved by about 30% with their system call batching approach.

Mentions in academic journals:

• A dynamic predictive race detector for C/C++ programs (in English, published 2017) uses Lwan as a "real world example".
• High-precision Data Race Detection Method for Large Scale Programs (in Chinese, published 2021) also uses Lwan as one of the case studies.
• AGE: Automatic Performance Evaluation of API Gateways (in English, published 2023) mentions Lwan as part of its usage in the KrakenD benchmarks.
• Canary: Practical Static Detection of Inter-thread Value-Flow Bugs used Lwan as one of the pieces of software that have been analyzed.
• Uma análise comparativa de ferramentas de análise estática para deteção de erros de memória used Lwan as one of the evaluation softwares.

Mentions in magazines:

• Linux-Magazin (Germany) mentions Lwan in their December/2021 issue
• Raspberry Pi Geek (Germany) mentions Lwan in their October/November 2022 issue
• LinuxUser (Germany) mentions Lwan in their October 2022 issue

Mentions in books:

• The Ascetic Programmer has a paragraph about Lwan.

Some talks mentioning Lwan:

• Talk about Lwan at Polyconf16, given by @lpereira.
• This talk about Iron, a framework for Rust, mentions Lwan as an insane C thing.
• University seminar presentation about Lwan.
• This presentation about Sailor web framework mentions Lwan.
• Performance and Scale @ Istio Service Mesh, presented at KubeCon Europe 2018, mentions (at the 7:30 mark) that Lwan is used on the server side for testing due to its performance and robustness.
• A multi-core Python HTTP server (much) faster than Go (spoiler: Cython) presented at PyConFR 2018 by J.-P. Smets mentions Nexedi's work on using Lwan as a backend for Python services with Cython.

Not really third-party, but alas:

• The author's blog.
• The project's webpage.

Container Images

Lwan container images are available at ghcr.io/lpereira/lwan. Container runtimes like Docker or Podman may be used to build and run Lwan in a container.

Pull lwan images from GHCR

Container images are tagged with release version numbers, so a specific version of Lwan can be pulled.

# latest version
docker pull ghcr.io/lpereira/lwan:latest
# pull a specific version
docker pull ghcr.io/lpereira/lwan:v0.3

Build images locally

Clone the repository and use Containerfile (Dockerfile) to build Lwan with all optional dependencies enabled.

podman build -t lwan .

Run your image

The image expects to find static content at /wwwroot, so a volume containing your content can be mounted.

docker run --rm -p 8080:8080 -v ./www:/wwwroot lwan

To bring your own lwan.conf, simply mount it at /lwan.conf.

podman run --rm -p 8080:8080 -v ./lwan.conf:/lwan.conf lwan

Run image with socket activation on a Linux host with Podman

Podman supports socket activation of containers. This example shows how to run lwan with socket activation and Podman on a Linux host.

Requirements: Podman version 4.5.0 or higher.

1. Create user test

   sudo useradd test
   

2. Start a login shell for the user test

   sudo machinectl shell test@
   

3. Clone the lwan git repository to ~/lwan
4. Build the image

   podman build -t lwan ~/lwan
   

5. Create directories

   mkdir -p ~/.config/containers/systemd
   mkdir -p ~/.config/systemd/user
   

6. Create the file ~/lwan.conf with the contents

   listener systemd:my.socket
   site {
       serve_files / {
               path = /web
       }
   }
   

7. Create the file ~/.config/systemd/user/my.socket with the contents

   [Socket]
   ListenStream=8080
   

8. Create the file ~/.config/containers/systemd/my.container with the contents

   [Unit]
   After=my.socket
   Requires=my.socket
   
   [Container]
   Network=none
   Image=localhost/lwan
   Volume=/home/test/lwan.conf:/lwan.conf:Z
   Volume=/home/test/web:/web:Z
   

   The option :Z is needed on SELinux systems. As lwan only needs to communicate over the socket-activated socket, it's possible to use Network=none. See the article How to limit container privilege with socket activation.
9. Create the web directory and an example text file

   mkdir ~/web
   echo hello > ~/web/file.txt
   

10. Reload systemd configuration

    systemctl --user daemon-reload
    

11. Start the socket

    systemctl --user start my.socket
    

12. Download the example text file from the lwan web server

    $ curl localhost:8080/file.txt
    hello
    

Lwan quotes

These are some of the quotes found in the wild about Lwan. They're presented in no particular order. Contributions are appreciated:

"Lwan is like a classic, according to the definition given by Italian -- writer Italo Calvino: you can read it again and again" Antonio Piccolboni

"I read lwan's source code. Especially, the part of using coroutine was very impressive and it was more interesting than a good novel. Thank you for that." -- @patagonia

"For the server side, we're using Lwan, which can handle 100k+ reqs/s. It's supposed to be super robust and it's working well for us." -- @fawadkhaliq

"Insane C thing" -- Michael Sproul

"The best performer is LWAN, a newcomer" -- InfoQ

"I've never had a chance to thank you for Lwan. It inspired me a lot to develop Zewo" -- @paulofariarl

"Let me say that lwan is a thing of beauty. I got sucked into reading the source code for pure entertainment, it's so good. high five" -- @kwilczynski

"mad science" -- jwz

"Nice work with Lwan! I haven't looked that carefully yet but so far I like what I saw. You definitely have the right ideas." -- @thinkingfish

"Lwan is a work of art. Every time I read through it, I am almost always awe-struck." -- @neurodrone

"For Round 10, Lwan has taken the crown" -- TechEmpower

"Jeez this is amazing. Just end to end, rock solid engineering. (...) But that sells this work short." kjeetgill

"I am only a spare time C coder myself and was surprised that I can follow the code. Nice!" cntlzw

"Impressive all and all, even more for being written in (grokkable!) C. Nice work." tpaschalis

"LWAN was a complete failure" dermetfan