PyMongo is thread-safe and even provides built-in connection pooling for threaded applications.

Every :class:`~pymongo.mongo_client.MongoClient` instance has a built-in connection pool. The pool begins with one open connection. Note that :attr:`~pymongo.mongo_client.MongoClient.max_pool_size` does not cap the number of connections; it only caps the number of idle connections kept open in the pool for future use. Thus, if 500 threads simultaneously launch long-running queries, PyMongo opens up to 500 connections to MongoDB, then closes all but max_pool_size of them as the queries complete.

When :meth:`~pymongo.mongo_client.MongoClient.disconnect` is called by any thread, all sockets are closed. PyMongo will create new sockets as needed.

:class:`~pymongo.mongo_replica_set_client.MongoReplicaSetClient` maintains one connection pool per server in your replica set.

.. seealso:: :doc:`examples/requests`

Starting with version 2.2 PyMongo supports Python 3.x where x >= 1. See the :doc:`python3` for details.

The only async framework that PyMongo fully supports is Gevent.

Currently there is no great way to use PyMongo in conjunction with Tornado or Twisted. PyMongo provides built-in connection pooling, so some of the benefits of those frameworks can be achieved just by writing multi-threaded code that shares a :class:`~pymongo.mongo_client.MongoClient`.

There are asynchronous MongoDB drivers in Python: AsyncMongo for Tornado and TxMongo for Twisted. Compared to PyMongo, however, these projects are less stable, lack features, and are less actively maintained.

It is possible to use PyMongo with Tornado, if some precautions are taken to avoid blocking the event loop:

• Make sure all MongoDB operations are very fast. Use the MongoDB profiler to watch for slow queries.
• Create a single :class:`~pymongo.mongo_client.MongoClient` instance for your application in your startup code, before starting the IOLoop.
• Configure the :class:`~pymongo.mongo_client.MongoClient` with a short socketTimeoutMS so slow operations result in a :class:`~pymongo.errors.TimeoutError`, rather than blocking the loop and preventing your application from responding to other requests.
• Start up extra Tornado processes. Tornado is typically deployed with one process per CPU core, proxied behind a load-balancer such as Nginx or HAProxy; when using Tornado with a blocking driver like PyMongo it's recommended you start two or three processes per core instead of one.

Cursors in MongoDB can timeout on the server if they've been open for a long time without any operations being performed on them. This can lead to an :class:`~pymongo.errors.OperationFailure` exception being raised when attempting to iterate the cursor.

MongoDB doesn't support custom timeouts for cursors, but cursor timeouts can be turned off entirely. Pass timeout=False to :meth:`~pymongo.collection.Collection.find`.

MongoDB only supports IEEE 754 floating points - the same as the Python float type. The only way PyMongo could store Decimal instances would be to convert them to this standard, so you'd really only be storing floats anyway - we force users to do this conversion explicitly so that they are aware that it is happening.

The database representation is 9.99 as an IEEE floating point (which is common to MongoDB and Python as well as most other modern languages). The problem is that 9.99 cannot be represented exactly with a double precision floating point - this is true in some versions of Python as well:

>>> 9.99
9.9900000000000002

The result that you get when you save 9.99 with PyMongo is exactly the same as the result you'd get saving it with the JavaScript shell or any of the other languages (and as the data you're working with when you type 9.99 into a Python program).

This request has come up a number of times but we've decided not to implement anything like this. The relevant jira case has some information about the decision, but here is a brief summary:

1. This will pollute the attribute namespace for documents, so could lead to subtle bugs / confusing errors when using a key with the same name as a dictionary method.
2. The only reason we even use SON objects instead of regular dictionaries is to maintain key ordering, since the server requires this for certain operations. So we're hesitant to needlessly complicate SON (at some point it's hypothetically possible we might want to revert back to using dictionaries alone, without breaking backwards compatibility for everyone).
3. It's easy (and Pythonic) for new users to deal with documents, since they behave just like dictionaries. If we start changing their behavior it adds a barrier to entry for new users - another class to learn.

Prior to PyMongo version 1.7, the correct way is to only save naive :class:`~datetime.datetime` instances, and to save all dates as UTC. In versions >= 1.7, the driver will automatically convert aware datetimes to UTC before saving them. By default, datetimes retrieved from the server (no matter what version of the driver you're using) will be naive and represent UTC. In newer versions of the driver you can set the :class:`~pymongo.mongo_client.MongoClient` tz_aware parameter to True, which will cause all :class:`~datetime.datetime` instances returned from that MongoClient to be aware (UTC). This setting is recommended, as it can force application code to handle timezones properly.

Something like :mod:`pytz` can be used to convert dates to localtime after retrieving them from the database.

PyMongo doesn't support saving :mod:`datetime.date` instances, since there is no BSON type for dates without times. Rather than having the driver enforce a convention for converting :mod:`datetime.date` instances to :mod:`datetime.datetime` instances for you, any conversion should be performed in your client code.

It's common in web applications to encode documents' ObjectIds in URLs, like:

"/posts/50b3bda58a02fb9a84d8991e"

Your web framework will pass the ObjectId portion of the URL to your request handler as a string, so it must be converted to :class:`~bson.objectid.ObjectId` before it is passed to :meth:`~pymongo.collection.Collection.find_one`. It is a common mistake to forget to do this conversion. Here's how to do it correctly in Flask (other web frameworks are similar):

from pymongo import MongoClient
from bson.objectid import ObjectId

from flask import Flask, render_template

connection = MongoClient()
app = Flask(__name__)

@app.route("/posts/<_id>")
def show_post(_id):
   # NOTE!: converting _id from string to ObjectId before passing to find_one
   post = connection.db.posts.find_one({'_id': ObjectId(_id)})
   return render_template('post.html', post=post)

if __name__ == "__main__":
    app.run()

.. seealso:: :ref:`querying-by-objectid`

Django is a popular Python web framework. Django includes an ORM, :mod:`django.db`. Currently, there's no official MongoDB backend for Django.

django-mongodb-engine is an unofficial, actively developed MongoDB backend that supports Django aggregations, (atomic) updates, embedded objects, Map/Reduce and GridFS. It allows you to use most of Django's built-in features, including the ORM, admin, authentication, site and session frameworks and caching through django-mongodb-cache.

However, it's easy to use MongoDB (and PyMongo) from Django without using a Django backend. Certain features of Django that require :mod:`django.db` (admin, authentication and sessions) will not work using just MongoDB, but most of what Django provides can still be used.

We have written a demo Django + MongoDB project. The README for that project describes some of what you need to do to use MongoDB from Django. The main point is that your persistence code will go directly into your views, rather than being defined in separate models. The README also gives instructions for how to change settings.py to disable the features that won't work with MongoDB.

One project which should make working with MongoDB and Django easier is mango. Mango is a set of MongoDB backends for Django sessions and authentication (bypassing :mod:`django.db` entirely).

mod_wsgi is a popular Apache module used for hosting Python applications conforming to the wsgi specification. There is a potential issue when deploying PyMongo applications with mod_wsgi involving PyMongo's C extension and mod_wsgi's multiple sub interpreters.

One tricky issue that we've seen when deploying PyMongo applications with mod_wsgi is documented here, in the Multiple Python Sub Interpreters section. When running PyMongo with the C extension enabled it is possible to see strange failures when encoding due to the way mod_wsgi handles module reloading with multiple sub interpreters. There are several possible ways to work around this issue:

1. Run mod_wsgi in daemon mode with each WSGI application assigned to its own daemon process.
2. Force all WSGI applications to run in the same application group.
3. Install PyMongo :ref:`without the C extension <install-no-c>` (this will carry a performance penalty, but is the most immediate solution to this problem).

The :mod:`json` module won't work out of the box with all documents from PyMongo as PyMongo supports some special types (like :class:`~bson.objectid.ObjectId` and :class:`~bson.dbref.DBRef`) that are not supported in JSON. We've added some utilities for working with :mod:`json` and :mod:`simplejson` in the :mod:`~bson.json_util` module.

On Unix systems, dates are represented as seconds from 1 January 1970 and usually stored in the C :mod:`time_t` type. On most 32-bit operating systems :mod:`time_t` is a signed 4 byte integer which means it can't handle dates after 19 January 2038; this is known as the year 2038 problem. Neither MongoDB nor Python uses :mod:`time_t` to represent dates internally so do not suffer from this problem, but Python's :mod:`datetime.datetime.fromtimestamp()` does, which means it is susceptible.

Previous to version 2.0, PyMongo used :mod:`datetime.datetime.fromtimestamp()` in its pure Python :mod:`bson` module. Therefore, on 32-bit systems you may get an error retrieving dates after 2038 from MongoDB using older versions of PyMongo with the pure Python version of :mod:`bson`.

This problem was fixed in the pure Python implementation of :mod:`bson` by commit b19ab334af2a29353529 (8 June 2011 - PyMongo 2.0).

The C implementation of :mod:`bson` also used to suffer from this problem but it was fixed in commit 566bc9fb7be6f9ab2604 (10 May 2010 - PyMongo 1.7).

PyMongo decodes BSON datetime values to instances of Python's :class:`datetime.datetime`. Instances of :class:`datetime.datetime` are limited to years between :data:`datetime.MINYEAR` (usually 1) and :data:`datetime.MAXYEAR` (usually 9999). Some MongoDB drivers (e.g. the PHP driver) can store BSON datetimes with year values far outside those supported by :class:`datetime.datetime`.

There are a few ways to work around this issue. One option is to filter out documents with values outside of the range supported by :class:`datetime.datetime`:

>>> from datetime import datetime
>>> coll = client.test.dates
>>> cur = coll.find({'dt': {'$gte': datetime.min, '$lte': datetime.max}})

Another option, assuming you don't need the datetime field, is to filter out just that field:

>>> cur = coll.find({}, fields={'dt': False})