How Mitogen Works

Some effort is required to accomplish the seemingly magical feat of bootstrapping a remote Python process without any software installed on the remote machine. The steps involved are unlikely to be immediately obvious to the casual reader, and they required several iterations to discover, so we document them thoroughly below.

The UNIX First Stage

To allow delivery of the bootstrap compressed using zlib, it is necessary for something on the remote to be prepared to decompress the payload and feed it to a Python interpreter. Since we would like to avoid writing an error-prone shell fragment to implement this, and since we must avoid writing to the remote machine’s disk in case it is read-only, the Python process started on the remote machine by Mitogen immediately forks in order to implement the decompression.

Python Command Line

The Python command line sent to the host is a zlib-compressed [1] and base64-encoded copy of the mitogen.master.Stream._first_stage() function, which has been carefully optimized to reduce its size. Prior to compression and encoding, CONTEXT_NAME is replaced with the desired context name in the function’s source code.

python -c 'exec "xxx".decode("base64").decode("zlib")'

The command-line arranges for the Python interpreter to decode the base64’d component, decompress it and execute it as Python code. Base64 is used since to protect against any special characters that may be interpreted by the system shell in use.

Forking The First Stage

The first stage creates a UNIX pipe and saves a copy of the process’s real stdin file descriptor (used for communication with the master) so that it can be recovered by the bootstrapped process later. It then forks into a new process.

After fork, the parent half overwrites its stdin with the read end of the pipe, and the child half writes the string MITOGEN0\n, then begins reading the zlib-compressed payload supplied on stdin by the master, and writing the decompressed result to the write-end of the UNIX pipe.

To allow recovery of stdin for reuse by the bootstrapped process for parent<->child communication, it is necessary for the first stage to avoid closing stdin or reading from it until until EOF. Therefore, the master sends the zlib-compressed payload prefixed with an integer size, allowing reading by the first stage of exactly the required bytes.

Configuring argv[0]

Forking provides us with an excellent opportunity for tidying up the eventual Python interpreter, in particular, restarting it using a fresh command-line to get rid of the large base64-encoded first stage parameter, and to replace argv[0] with something descriptive.

After configuring its stdin to point to the read end of the pipe, the parent half of the fork re-executes Python, with argv[0] taken from the CONTEXT_NAME variable earlier substituted into its source code. As no arguments are provided to this new execution of Python, and since stdin is connected to a pipe (whose write end is connected to the first stage), the Python interpreter begins reading source code to execute from the pipe connected to stdin.

Bootstrap Preparation

Now we have the mechanism in place to send a zlib-compressed script to the remote Python interpreter, it is time to choose what to send.

The script sent is simply the source code for mitogen.core, with a single line suffixed to trigger execution of the mitogen.core.ExternalContext.main() function. The encoded arguments to the main function include some additional details, such as the logging package level that was active in the parent process, and whether debugging or profiling are enabled.

After the script source code is prepared, it is passed through mitogen.master.minimize_source() to strip it of docstrings and comments, while preserving line numbers. This reduces the compressed payload by around 20%.

Preserving The mitogen.core Source

One final trick is implemented in the first stage: after bootstrapping the new child, it writes a duplicate copy of the mitogen.core source it just used to bootstrap it back into another pipe connected to the child. The child’s module importer cache is initialized with a copy of the source, so that subsequent bootstraps of children-of-children do not require the source to be fetched from the master a second time.

Signalling Success

Once the first stage has signalled MITO000\n, the master knows it is ready to receive the compressed bootstrap. After decompressing and writing the bootstrap source to its parent Python interpreter, the first stage writes the string MITO001\n to stdout before exiting. The master process waits for this string before considering bootstrap successful and the child’s stdio ready to receive messages.

The signal value is 8 bytes to match the minimum chunk size required to disambiguate between lines containing an interesting token during SSH password authentication, a debug message from the SSH client itself, or a message from the first stage.



Generating A Synthetic mitogen Package

Since the bootstrap consists of the mitogen.core source code, and this code is loaded by Python by way of its main script (__main__ module), initially the module layout in the child will be incorrect.

The first step taken after bootstrap is to rearrange sys.modules slightly so that mitogen.core appears in the correct location, and all classes defined in that module have their __module__ attribute fixed up such that cPickle correctly serializes instance module names.

Once a synthetic mitogen package and mitogen.core module have been generated, the bootstrap deletes sys.modules[‘__main__’], so that any attempt to import it (by cPickle) will cause the import to be satisfied by fetching the master’s actual __main__ module. This is necessary to allow master programs to be written as a self-contained Python script.

Reaping The First Stage

After the bootstrap has called os.dup() on the copy of the stdin file descriptor saved by the first stage, it is closed.

Additionally, since the first stage was forked prior to re-executing the Python interpreter, it will exist as a zombie process until the parent process reaps it. Therefore the bootstrap must call os.wait() soon after startup.

Setup Logging

The child’s logging package root logger is configured to have the same log level as the root logger in the master, and mitogen.core.LogHandler is installed to forward logs to the master context’s FORWARD_LOG handle.

The log level is copied into the child to avoid generating a potentially large amount of network IO forwarding logs that will simply be filtered away once they reach the master.

The Module Importer

An instance of mitogen.core.Importer is installed in sys.meta_path, where Python’s import statement will execute it before attempting to find a module locally.

Standard IO Redirection

Two instances of mitogen.core.IoLogger are created, one for stdout and one for stderr. This class creates a UNIX pipe whose read end is added to the IO multiplexer, and whose write end is used to overwrite the handles inherited during process creation.

Even without IO redirection, something must replace stdin and stdout, otherwise it is possible for the stream used for communication between parent and child to be accidentally corrupted by subprocesses run by user code.

The inherited stdin is replaced by a file descriptor pointing to /dev/null.

Finally Python’s sys.stdout is reopened to ensure line buffering is active, so that print statements and suchlike promptly appear in the logs.

Function Call Dispatch

After all initialization is complete, the child’s main thread sits in a loop reading from a Channel connected to the CALL_FUNCTION handle. This handle is written to by call() and call_async().

CALL_FUNCTION only accepts requests from the context IDs listed in mitogen.parent_ids, forming a chain of trust between the master and any intermediate context leading to the recipient of the message. In combination with Source Verification, this is a major contributor to ensuring contexts running on compromised infrastructure cannot trigger code execution in siblings or any parent.


When a context receives SHUTDOWN from its immediate parent, it closes its own CALL_FUNCTION Channel before sending SHUTDOWN to any directly connected children. Closing the channel has the effect of causing ExternalContext._dispatch_calls() to exit and begin joining on the broker thread.

During shutdown, the master waits up to 5 seconds for children to disconnect gracefully before force disconnecting them, while children will use that time to call socket.shutdown(SHUT_WR) on their IoLogger socket’s write ends before draining any remaining data buffered on the read ends, and ensuring any deferred broker function callbacks have had a chance to complete, necessary to capture for example forwarding any remaining logging records.

An alternative approach is to wait until the IoLogger socket is completely closed, with some hard timeout, but this necessitates greater discipline than is common in infrastructure code (how often have you forgotten to redirect stderr to /dev/null when starting a daemon process?), so needless irritating delays would often be experienced during program termination.

If the main thread (responsible for function call dispatch) fails to shut down gracefully, because some user function is hanging, it will still be cleaned up since as the final step in broker shutdown, the broker sends signal.SIGTERM to its own process.

Stream Protocol

Once connected, a basic framing protocol is used to communicate between parent and child. Integers use big endian in their encoded form.

Field Size Description
dst_id 4 Integer target context ID. Router delivers messages locally when their dst_id matches mitogen.context_id, otherwise they are routed up or downstream.
src_id 4 Integer source context ID. Used as the target of replies if any are generated.
auth_id 4 The context ID under whose authority the message is acting. See Source Verification.
handle 4 Integer target handle in the destination context. This is one of the Standard Handles, or a dynamically generated handle used to receive a one-time reply, such as the return value of a function call.
reply_to 4 Integer target handle to direct any reply to this message. Used to receive a one-time reply, such as the return value of a function call. IS_DEAD has a special meaning when it appears in this field.
length 4 Length of the data part of the message.
data n/a Message data, which may be raw or pickled.

Standard Handles

Masters listen on the following handles:


Receives (logger_name, level, msg) 3-tuples and writes them to the master’s mitogen.ctx.<context_name> logger.


Receives the name of a module to load fullname, locates the source code for fullname, and routes one or more LOAD_MODULE messages back towards the sender of the GET_MODULE request. If lookup fails, None is sent instead.

See Import Preloading for a deeper discussion of GET_MODULE/LOAD_MODULE.


Replies to any message sent to it with a newly allocated range of context IDs, to allow children to safely start their own contexts. Presently IDs are allocated in batches of 1000 from a 32 bit range, allowing up to 4.2 million parent contexts to be created and destroyed before the associated Router must be recreated.

Children listen on the following handles:


Receives (pkg_present, path, compressed, related) tuples, composed of:

  • pkg_present: Either None for a plain .py module, or a list of canonical names of submodules existing witin this package. For example, a LOAD_MODULE for the mitogen package would return a list like: [“mitogen.core”, “mitogen.fakessh”, “mitogen.master”, ..]. This list is used by children to avoid generating useless round-trips due to Python 2.x’s import statement behavior.
  • path: Original filesystem where the module was found on the master.
  • compressed: zlib-compressed module source code.
  • related: list of canonical module names on which this module appears to depend. Used by children that have ever started any children of their own to preload those children with LOAD_MODULE messages in response to a GET_MODULE request.

Receives (mod_name, class_name, func_name, args, kwargs) 5-tuples from call_async(), imports mod_name, then attempts to execute class_name.func_name(*args, **kwargs).

When this channel is closed (by way of receiving a dead message), the child’s main thread begins graceful shutdown of its own Broker and Router.


When received from a child’s immediate parent, causes the broker thread to enter graceful shutdown, including sending a dead message to the child’s main thread, causing it to join on the exit of the broker thread.

The final step of a child’s broker shutdown process sends signal.SIGTERM to itself, ensuring the process dies even if the main thread was hung executing user code.

Each context is responsible for sending SHUTDOWN to each of its directly connected children in response to the master sending SHUTDOWN to it, and arranging for the connection to its parent to be closed shortly thereafter.

Masters, and children that have ever been used to create a descendent child also listen on the following handles:


Receives target_id integer from downstream, describing an ID allocated to a recently constructed child. The receiver verifies no existing route exists to target_id before updating its local table to route messages for target_id via the stream from which the ADD_ROUTE message was received.


Receives target_id integer from downstream, verifies a route exists to target_id via the stream on which the message was received, removes that route from its local table, then propagates the message upward towards its own parent.


Sent to inform a parent that user code has invoked ExternalContext.detach() to decouple the lifecycle of a directly connected context and its subtree from the running program.

A child usually shuts down immediately if it loses its parent connection, and parents usually terminate any related Python/SSH subprocess on disconnection. Receiving DETACHING informs the parent the connection will soon drop, but the process intends to continue life independently, and to avoid terminating the related subprocess if that subprocess is the child itself.

Non-master parents also listen on the following handles:


As with master’s GET_MODULE, except this implementation (mitogen.master.ModuleForwarder) serves responses using mitogen.core.Importer’s cache before forwarding the request to its parent context. The response is cached by each context in turn before being forwarded on to the child context that originally made the request. In this way, the master need never re-send a module it has already sent to a direct descendant.


Receives (context, fullname) tuples from its parent and arranges for a LOAD_MODULE to be sent towards context for the module fullname and any related modules. The module must already have been delivered to the current context by its parent in a prior LOAD_MODULE message.

If the receiver is the immediate parent of context, then only LOAD_MODULE is sent to the child. Otherwise LOAD_MODULE is sent to the next closest parent if the module has not previously been sent on that stream, followed by a copy of the FORWARD_MODULE message.

This message is used to recursively preload indirect children with modules, ensuring they are cached and deduplicated at each hop in the chain leading to the target context.

Special values for the reply_to field:


Special value used to signal disconnection or the inability to route a message, when it appears in the reply_to field. Usually causes mitogen.core.ChannelError to be raised when it is received.

It indicates the sender did not know how to process the message, or wishes no further messages to be delivered to it. It is used when:

  • a remote receiver is disconnected or explicitly closed.
  • a related message could not be delivered due to no route existing for it.
  • a router is being torn down, as a sentinel value to notify mitogen.core.Router.add_handler() callbacks to clean up.

Additional handles are created to receive the result of every function call triggered by call_async().

Use of Pickle

The current implementation uses the Python cPickle module, with a restrictive class whitelist to prevent triggering undesirable code execution. The primary reason for using cPickle is that it is computationally efficient, and avoids including a potentially large body of serialization code in the bootstrap.

The pickler will instantiate only built-in types and one of 3 constructor functions, to support unpickling CallError, mitogen.core.Sender,and Context.

The choice of Pickle is one area to be revisited later. All accounts suggest it cannot be used securely, however few of those accounts appear to be expert, and none mention any additional attacks that would not be prevented by using a restrictive class whitelist.

The IO Multiplexer

Since we must include our IO multiplexer as part of the bootstrap, off-the-shelf implementations are for the most part entirely inappropriate. For example, a minimal copy of Twisted weighs in at around 440KiB and is composed of approximately 115 files. Even if we could arrange for an entire Python package to be transferred during bootstrap, this minimal configuration is massive in comparison to Mitogen’s solution, multiplies quickly in the presence of many machines, and would require manually splitting up the parts of Twisted that we would like to use.

Message Routing

Routing assumes it is impossible to construct a tree such that one of a context’s parents will not know the ID of a target the context is attempting to communicate with.

When mitogen.core.Router receives a message, it checks the IDs associated with its directly connected streams for a potential route. If any stream matches, either because it directly connects to the target ID, or because the master sent an ADD_ROUTE message associating it, then the message will be forwarded down the tree using that stream.

If the message does not match any ADD_ROUTE message or stream, instead it is forwarded upwards to the immediate parent, and recursively by each parent in turn until one is reached that knows how to forward the message down the tree.

When a parent establishes a new child, it sends a corresponding ADD_ROUTE message towards its parent, which recursively forwards it up towards the root.

Parents keep note of all routes associated with each stream they connect with, and trigger DEL_ROUTE messages propagated upstream for each route associated with that stream if the stream is disconnected for any reason.



In the diagram, when node12b is creating the sudo:node12b:webapp context, it must send ADD_ROUTE messages to rack12, which will propagate it to dc1, and recursively to bastion, and master; node12b does not require an ADD_ROUTE message since it has a stream directly connected to the new context.

Since Mitogen streams are strictly ordered, it is never possible for a parent to receive a message from a newly constructed child before receiving a corresponding ADD_ROUTE sent by the child’s parent, describing how to reply to it.

When sudo:node22a:webapp wants to send a message to sudo:node12b:webapp, the message will be routed as follows:

sudo:node22a:webapp -> node22a -> rack22 -> dc2 -> bastion -> dc1 -> rack12 -> node12b -> sudo:node12b:webapp


Source Verification

Before forwarding or dispatching a message it has received, mitogen.core.Router first looks up the corresponding mitogen.core.Stream it would use to send responses towards the context ID listed in the auth_id field, and if the looked up stream does not match the stream on which the message was received, the message is discarded and a warning is logged.

This creates a trust chain leading up to the root of the tree, preventing downstream contexts from injecting messages appearing to be from the master or any more trustworthy parent. In this way, privileged functionality such as CALL_FUNCTION can base trust decisions on the accuracy of auth_id.

The auth_id field is separate from src_id in order to support granting privilege to contexts that do not follow the tree’s natural trust chain. This supports cases where siblings are permitted to execute code on one another, or where isolated processes can connect to a listener and communicate with an already established established tree.

Differences Between Master And Child Brokers

The main difference between mitogen.core.Broker and mitogen.master.Broker is that when the stream connection to the parent is lost in a child, the broker will trigger its own shutdown.

The Module Importer

mitogen.core.Importer is still a work in progress, as there are a variety of approaches to implementing it, and the present implementation is not pefectly efficient in every case.

It operates by intercepting import statements via sys.meta_path, asking Python if it can satisfy the import by itself, and if not, indicating to Python that it is capable of loading the module.

In load_module() an RPC is started to the parent context, requesting the module source code by way of a GET_MODULE. If the parent context does not have the module available, it recursively forwards the request upstream, while avoiding duplicate requests for the same module from its own threads and any child contexts.

Neutralizing __main__

To avoid accidental execution of the __main__ module’s code in a slave context, when serving the source of the main module, Mitogen removes any code occurring after the first conditional that looks like a standard __main__ execution guard:

# Code that looks like this is stripped from __main__.
if __name__ == '__main__':

This is a hack, but it’s the least annoying hack I’ve found for the problem yet.

Avoiding Negative Imports

In Python 2.x where relative imports are the default, a large number of import requests will be made for modules that do not exist. For example:

# mypkg/

import sys
import os

In Python 2.x, Python will first try to load mypkg.sys and mypkg.os, which do not exist, before falling back on sys and os.

These negative imports present a challenge, as they introduce a large number of pointless network round-trips. Therefore in addition to the zlib-compressed source, for packages the master sends along a list of child modules known to exist.

Before indicating it can satisfy an import request, mitogen.core.Importer first checks to see if the module belongs to a package it has previously imported, and if so, ignores the request if the module does not appear in the enumeration of child modules belonging to the package that was provided by the master.

Import Preloading

To further avoid round-trips, when a module or package is requested by a child, its bytecode is scanned in the master to find all the module’s import statements, and of those, which associated modules appear to have been loaded in the master’s sys.modules.

The sys.modules check is necessary to handle various kinds of conditional execution, for example, when a module’s code guards an import statement based on the active Python runtime version, operating system, or optional third party dependencies.

Before replying to a child’s request for a module with dependencies:

  • If the request is for a package, any dependent modules used by the package that appear within the package itself are known to be missing from the child, since the child requested the top-level package module, therefore they are pre-loaded into the child using LOAD_MODULE messages before sending the LOAD_MODULE message for the requested package module itself. In this way, the child will already have dependent modules cached by the time it receives the requested module, avoiding one round-trip for each dependency.

    For example, when a child requests the django package, and the master determines the django module code in the master has import statements for django.utils, django.utils.lru_cache, and django.utils.version, and that execution of the module code on the master caused those modules to appear in the master’s sys.modules, there is high probability execution of the django module code in the child will cause the same modules to be loaded. Since all those modules exist within the django package, and we already know the child lacks that package, it is safe to assume the child will make follow-up requests for those modules too.

    In the example, 4 round-trips are replaced by 1 round-trip.

For any package module ever requested by a child, the parent keeps a note of the name of the package for one final optimization:

  • If the request is for a sub-module of a package, and it is known the child loaded the package’s implementation from the parent, then any dependent modules of the requested module at any nesting level within the package that is known to be missing are sent using LOAD_MODULE messages before sending the LOAD_MODULE message for the requested module, avoiding 1 round-trip for each dependency within the same top-level package.

    For example, when a child has previously requested the django package module, the parent knows the package was completely absent on the child. Therefore when the child subsequently requests the django.db package module, it is safe to assume the child will generate subsequent GET_MODULE requests for the 2 django.conf, 3 django.core, 2 django.db, 3 django.dispatch, and 7 django.utils indirect dependencies for django.db.

    In the example, 17 round-trips are replaced by 1 round-trip.

The method used to detect import statements is similar to the standard library modulefinder module: rather than analyze module source code, IMPORT_NAME opcodes are extracted from the module’s bytecode. This is since clean source analysis methods (ast and compiler) are an order of magnitude slower, and incompatible across major Python versions.


Duplicate requests must never be issued to the parent, either due to a local import or any GET_MODULE originating from a child. This lets parents assume a module requested once by a downstream connection need never be re-sent, for example, if it appears as a preloading dependency in a subsequent GET_MODULE, or had been requested immediately after being sent as a preloading dependency for an unrelated request by a descendent.

Therefore each tree layer must deduplicate GET_MODULE requests, and synchronize their descendents and local threads on corresponding LOAD_MODULE responses from the parent.

In each context, pending requests are serialized by a threading.Lock within mitogen.core.Importer, which may only be held for operations that cannot block, since ModuleForwarder must acquire it while synchronizing GET_MODULE requests from children on the IO multiplexer thread.

Requests From Local Threads

When Mitogen begins satisfying an import, it is known the module has never been imported in the local process. Importer executes under the runtime importer lock, ensuring import statements executing in local threads are serialized.


In Python 2, ImportError is raised when import is attempted while the runtime import lock is held by another thread, therefore imports must be serialized by only attempting them from the main (CALL_FUNCTION) thread.

The problem is most likely to manifest in third party libraries that lazily import optional dependencies at runtime from a non-main thread. The workaround is to explicitly import those dependencies from the main thread before initializing the third party library.

This was fixed in Python 3.5, but Python 3.x is not yet supported. See Python Issue #9260.

While holding its own lock, Importer checks if the source is not yet cached, determines if an in-flight GET_MODULE exists for it, starting one if none exists, adds itself to a list of callbacks fired when a corresponding LOAD_MODULE arrives from the parent, then sleeps waiting for the callback.

When the source becomes available, the module is constructed on the calling thread using the best practice documented in PEP 302.

Requests From Children

As with local imports, when GET_MODULE is received from a child, while holding the Importer lock, ModuleForwarder checks if the source is not yet cached, determines if an in-flight GET_MODULE toward the parent exists for it, starting one if none exists, then adds a completion handler to the list of callbacks fired when a corresponding LOAD_MODULE arrives from the parent.

When the source becomes available, the completion handler issues corresponding LOAD_MODULE messages toward the child for the requested module after any required for dependencies known to be absent from the child.

Since intermediaries do not know a module’s dependencies until the module’s source arrives, it is not possible to preemptively issue LOAD_MODULE for those dependencies toward a requesting child as they become available from the parent at the intermediary. This creates needless network serialization and latency that should be addressed in a future design.

Child Module Enumeration

Package children are enumerated using pkgutil.iter_modules().

Use Of Threads

The package always runs the IO multiplexer in a thread. This is so the multiplexer retains control flow in order to shut down gracefully, say, if the user’s code has hung and the master context has disconnected.

While it is possible for the IO multiplexer to recover control of a hung function call on UNIX using for example signal.SIGALRM, this mechanism is not portable to non-UNIX operating systems, and does not work in every case, for example when Python blocks signals during a variety of threading package operations.

At some point it is likely Mitogen will be extended to support children running on Windows. When that happens, it would be nice if the process model on Windows and UNIX did not differ, and in fact the code used on both were identical.

Waking Sleeping Threads

Due to fundamental deficiencies in Python 2’s threading implementation, it is not possible to block waiting on synchronization objects sanely. Two major problems exist:

  • Sleeping with no timeout set causes signals to be blocked, preventing the user from terminating the process using CTRL+C.
  • Sleeping with a timeout set internally makes use of polling, with an exponential backoff that eventually results in the thread sleeping unconditionally in 50ms increments. . This is a huge source of latency that quickly multiplies.

As the UNIX self-pipe trick must already be employed to wake the broker thread from its select loop, Mitogen reuses this technique to wake any thread synchronization primitive exposed by the library, embodied in a queue-like abstraction called a mitogen.core.Latch.

Unfortunately it is commonplace for hosts to enforce severe per-process file descriptors limits, so aside from being inefficient, it is impossible in the usual case to create a pair of descriptors for every waitable object, which for example includes the result of every single asynchronous function call.

For this reason self-pipes are created on a per-thread basis, with their associated socketpairs kept in thread-local storage. When a latch wishes to sleep its thread, this pair is created on-demand and temporarily associated with it only for the duration of the sleep.

Python’s garbage collector is relied on to clean up by calling the pair’s destructor on thread exit. There does not otherwise seem to be a robust method to trigger cleanup code on arbitrary threads.

To summarize, file descriptor usage is bounded by the number of threads rather than the number of waitables, which is a much smaller number, however it also means that Mitogen requires twice as many file descriptors as there are user threads, with a minimum of 4 required in any configuration.

Latch Internals


  • lockthreading.Lock.
  • queue – items waiting to be dequeued.
  • sleeping – write sides of the socketpairs for each sleeping thread, and threads in the process of waking from sleep.
  • waking – integer number of sleeping threads in the process of waking up.
  • closed – boolean defaulting to False. Every time lock is acquired, closed must be tested, and if it is True, LatchError must be thrown.


Latch.put() operates by:

  1. Acquiring lock.
  2. Appending the item on to queue.
  3. If waking is less than the length of sleeping, write a byte to the socket at sleeping[waking] and increment waking.

In this way each thread is woken only once, and receives each element according to when its socket was placed on sleeping.


Latch.close() acquires lock, sets closed to True, then writes a byte to every sleeping[waking] socket, while incrementing waking, until no more unwoken sockets exist. Per above, on waking from sleep, after removing itself from sleeping, each sleeping thread tests if closed is True, and if so throws LatchError.

It is necessary to ensure at most one byte is delivered on each socket, even if the latch is being torn down, as the sockets outlive the scope of a single latch, and must never have extraneous data buffered on them, as this will cause unexpected wakeups if future latches sleep on the same thread.


Latch.get() is far more intricate, as there are many outcomes to handle. Queue ordering is strictly first-in first-out, and threads always receive items in the order they are requested, as they become available.

1. Non-empty, No Waiters, No sleep
On entry lock is taken, and if queue is non-empty, and sleeping is empty, it is safe to return queue’s first item without blocking.
2. Non-empty, Waiters Present, Queue > Waiters, No sleep
When sleeping is non-empty but there are more items than sleeping threads, it is safe to pop queue[len(sleeping)] without blocking.
3. Non-empty, Waiters Present, Queue <= Waiters

In this case sleeping is non-empty and there are no surplus items. It is not safe to pop any item even though we are holding lock, as it would starve waking threads of their position in favour of the calling thread, since scheduling uncertainty exists between a thread waking from and re-acquiring lock.

This avoids the need for a retry loop for waking threads, and a thread being continually re-woken to discover queue drained by a thread that never slept.

4. Sleep

Since no surplus items existed, the thread adds its socket to sleeping before releasing lock, and sleeping in waiting for timeout, or a write from Latch.put() or Latch.close().

If throws an exception, the exception must be caught and re-raised only after some of the wake steps below have completed.

5. Wake, Non-empty

On wake lock is re-acquired, the socket is removed from sleeping after noting its index, and TimeoutError is thrown if waking indicates Latch.put() nor Latch.close() have yet to send a wake byte to that index. The byte is then read off, LatchError is thrown if closed is True, otherwise the queue item corresponding to the thread’s index is popped and returned.

It is paramount that in every case, if a byte was written to the socket, that the byte is read away. The socket is reused by subsequent latches sleeping on the same thread, and unexpected wakeups are triggered if extraneous data remains buffered on the socket.

It is also necessary to favour the synchronized waking variable over the return value of, as scheduling uncertainty introduces a race between the select timing out, and Latch.put() or Latch.close() writing a wake byte before Latch.get() has re-acquired lock.


[1]Compression may seem redundant, however it is basically free and reducing IO is always a good idea. The 33% / 200 byte saving may mean the presence or absence of an additional frame on the network, or in real world terms after accounting for SSH overhead, around a 2% reduced chance of a stall during connection setup due to a dropped frame.