Skip to main content

A command line tool to orchestrate MongoDB backups

Project description


Send your mongodump backups straight to AWS S3

m2bk is command line tool that performs a number of mongodb database backups via mongodump, compresses them into a gzipped tarball and finally sends them to an AWS S3 bucket (more options are about to be available).



There are many ways in which you can contribute to m2bk. Code patches are just one thing amongst others that you can submit to help the project. We also welcome feedback, bug reports, feature requests, documentation improvements, advertisement and testing.

Feedback contributions

This is by far the easiest way to contribute something. If you’re using m2bk for your own benefit, don’t hesitate sharing. Feel free to submit issues and enhancement requests.

Code contributions

Code contributions (patches, new features) are the most obvious way to help with the project’s development. Since this is so common we ask you to follow our workflow to most efficiently work with us. For code contributions, we follow the “fork-and-pull” Git workflow.

1. Fork, then clone your repo on GitHub

git clone
git add origin upstream

If you already forked the repo, then be sure to merge the most recent changes from “upstream” before making a pull request.

git pull upstream

2. Create a new feature branch in your local repo

git checkout -b my_feature_branch

3. Make your changes, then make sure the tests passes

pyvenv m2bk-pyve && source m2bk-pyve/bin/activate
python3 test

4. Commit your changes once done

git commit -a -m "My commit message"
git push origin my_feature_branch
  1. Submit a pull request with your feature branch containing your changes.


Installation of m2bk can be made directly from source, via pip or easy_install, whichever you prefer.

Option # 1: pip

$ pip install m2bk

Option # 2: from source

$ git clone
$ cd m2bk
$ python3 install

Option # 3: easy_install

$ easy_install m2bk

From this point you can edit your configuration file

$ vi /etc/m2bk/m2bk.yaml

Basic Usage

Normal execution

$ m2bk

Quiet output

$ m2bk -q

Dry run

$ m2bk -d

Specify an alternate configuration file

$ m2bk -c /path/to/my/custom/m2bk.yaml


m2bk [options]
  • --version show version number and exit

  • -h | --help show a help message and exit

  • -c [file] | --config=[file] | --config [file] specify configuration file to use

  • -d | --dry-run don’t actually do anything

  • -q | --quiet quiet output

  • --ll | --log-level=[num] set logging output level

  • -l LOG_FILE | --log-file LOG_FILE set log file

Configuration file

The configuration is handled through a simple YAML file including a series of sections (which are YAML objects), each one composed by directives (YAML numbers, strings or arrays), these will determine a corresponding behavior on m2bk. If m2bk does not receive any configuration file on command line, it will try to read /etc/m2bk.yaml. Please note the configuration format is still a work in progress and will most likely change in the early stages of m2bk.

The following is an example of what a configuration file looks like:

  name: s3
    aws_access_key_id: "SDF73HSDF3663KSKDJ"
    aws_secret_access_key: "d577273ff885c3f84dadb8578bb41399"
  output_dir: "/opt/tmp/mydir"
  mongodump: "/opt/bin/mongodump"
    port: 666
    user_name: "satan"
    password: "14mh4x0r"
      address: "foo.example.local"
      port: 34127
        - "app"
        - "sessions"
        - "another_one"
      address: ""
      password: "1AmAn07h3rh4x0r"
      auth_db: bar
          - customers
          - sessions

Through this configuration file, you can set key variables about the databases you want to backup and the AWS S3 bucket you wish to send them to.

Configuration file: sections and directives

fs section

This section has directives regarding files and directories manipulation



  • Type: string

  • Default value : /tmp/m2bk

  • Role: directory where m2bk is going to temporarily save backup files

mongodb section

This section holds directives regarding mongodb servers m2bk is going to connect to, including databases that are going to be backed up through mongodump.


    mongodump: "/opt/bin/mongodump"
        user_name: tom
        address: db.example.local
        password: "457893mnfs3j"
          - halloran
          - grady
            address: db0.example.internal
            port: 27654
            user_name: matt
            password: "myS3cr37P455w0rd"
              # This list is going to be merged with dbs at host_defaults, thus
              # the resulting dbs will be:
              # ['halloran', 'grady', 'jack', 'wendy', 'danny']
              - jack
              - wendy
              - danny
        bar: {} # This one is going to acquire all host_defaults values
            # This host will inherit port, password and dbs from host_defaults
            address: moloko.example.internal
            user_name: alex
            address: localhost
            auth_db: milk_plus



  • Type: string

  • Default value : mongodump

  • Role: full path to the mongodump executable used by m2bk

mongodb.host_defaults section

Many directives (such as user name and/or password) could be common among the databases that are going to be backed up. For this reason, it is best to simply put those common directives under a single section, this is entirely optional but also it is the best for easily manageable configuration files in order to avoid redundancy, the supported directives are user_name, password, port, dbs and auth_db . See hosts section.

mongodb.hosts section

This is an object/hash, where each element contains a series of directives relative to a mongodb database located at a server, its specifications and databases themselves held by it, these are the main values used by mongodump when it does its magic. For each entry inside the hosts section, these are its valid directives:


  • Type: string

  • Required: YES

  • Role: mongodb server location

  • Type: integer

  • Required: NO

  • Default value : mongo.host_defaults.port | 27017

  • Role: mongodb server listening port

  • Type: string

  • Required: NO

  • Default value : mongodb.host_defaults.user_name | m2bk

  • Role: user name used for authentication against the mongodb server

  • Type: string

  • Required: NO

  • Default value : mongodb.host_defaults.pass | "pass"

  • Role: password used for authentication against the mongodb server

  • Type: string

  • Required: NO

  • Default value : admin

  • Role: authentication database

  • Type: array

  • Required: NO

  • Default value : mongodb.host_defaults.dbs | []

  • Role: a list of databases who are expected inside the mongodb server

NOTE: particular “dbs” on one host will be merged with those of “host_defaults”

Drivers (driver section)

Once backup files have been generated, they are then handled by a driver, whose job is to transfer resulting backup files to some form of storage (depending on the driver set on configuration). Drivers (and their options) are set and configured inside the driver section like so:

    # First of all, you need to tell m2bk which driver to use
    name: dummy

    # Inside this key, driver options are set
      hello: world
      another_option: another_value

Per driver, there are a bunch of available options to tweak them. These options vary among drivers. Though there is only one driver available on m2bk, there will be more drivers available with new releases. Current available drivers are the following:


This driver is just a placeholder for testing out the driver interface as it won’t do a thing on backup files.


There are no options for this driver. Any option passed to this driver will be logged at debug level.


This driver holds directives regarding AWS credentials that m2bk is going to use in order to upload the mongodump backups to S3. If either aws_access_key_id or aws_secret_access_key are not specified, this driver will not try to use them to authenticate against AWS and will rely on boto config for that matter.


    name: s3
        aws_access_key_id": "HAS6NBASD8787SD"
        aws_secret_access_key: "d41d8cd98f00b204e9800998ecf8427e"
        s3_bucket: "mybucket"


  • Type: string

  • Required: NO

  • Role: AWS access key ID

  • Type: string

  • Required: NO

  • Role: AWS access key ID

  • Type: string

  • Required: NO

  • Default value: m2bk

  • Role: name of the main S3 bucket where m2bk is going to upload the compressed backups for each mongodb server specified in mongodb section


Show your love and support this project via gratipay

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

m2bk-0.4.0.tar.gz (31.5 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page