Skip to content

willsion/mongodb-balance-check

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

mongodb-balance-check

This script provides an easy way to find out if your MongoDB shard cluster is properly balanced or not.

You can achieve the same effect by calling db.printShardingStatus() but it requires manually parsing the output to see if the counts are balanced for all your collections, which could be difficult if you have a lot...or are lazy. This script does it for you and makes it clear.

Requirements

Installation

git clone git://github.com/serverdensity/mongodb-balance-check.git
sudo pip install pymongo
sudo pip install clint

Running

This assumes you have a mongos running on localhost:27017

python check.py

Console Output

It will list all your collections from all databases, indicating whether they are balanced or not and the total chunk count for each one.

david@sdapp-web1 ~/mongodb-balance-check: python check.py 
sd.servers
    sdapp1 balanced (9)
    sdapp2 balanced (9)
sd.users
    sdapp1 unbalanced (62)
    sdapp2 unbalanced (89)
sd.alerts
    sdapp1 balanced (12)
    sdapp2 balanced (12)
...

Programmatic output

You can import balanced into your own code as it has the option to return the value as part of a method call instead of outputting to the console:

import balanced
result = balanced.is_balanced()
print result
{
    'chunks': {
        u'sd.servers': {
            u'sdapp1': 9,
            u'sdapp2': 9
        },
        u'sd.users': {
            u'sdapp1': 62,
            u'sdapp2': 89
        },
        u'sd.alerts': {
            u'sdapp1': 12,
            u'sdapp2': 12
        }
    },
    'balanceStatus': {
        u'sd.servers': True, 
        u'sd.users': False, 
        u'sd.alerts': True
    }
    'isBalanced': False
}

Server Density plugin

This can be used as a plugin to Server Density to give you shard distribution graphs and trigger alerts when certain shards become unbalanced. It runs once per hour as it takes several seconds to run and doesn't need to be "real time".

  1. Create the plugin directory for your agent if you haven't already. This involves editing the /etc/sd-agent/config.cfg file to point the plugin_directory config value to a directory e.g. /usr/bin/sd-agent/plugins (which you need to create).

  2. In your Server Density account, click the Plugins tab then create a new plugin called MongoBalanced

  3. Download the balanced.py and MongoBalanced.py files from this repo and place them in the plugin directory you created above.

  4. Restart the agent

Values will be reported back right away and will appear on the graphs. There will be a lot of them; there are 3 different types.

  • database-collection-shard e.g. sd-users-sdapp1 - this shows the number of chunks for each database, collection and shard and is useful for graphing.
  • database-collection e.g. sd-users - this will be either 0 (unbalanced) or 1 (balanced) and is used for alerting on specific collections.
  • isBalanced this will be either 0 (unbalanced) or 1 (balanced) and shows the status for the whole cluster.

Graphs

From the Plugins tab you can edit the plugin to create the graphs based on the first type. For example, with these keys:

sd-users-sdapp1
sd-users-sdapp2
sd-alerts-sdapp1
sd-alerts-sdapp2

you can create a graph with the title Users and keys to display sd-users-sdapp1, sd-users-sdapp2 which would show the chunk distribution for both shards.

Alerts

To get an alert when the sd-users collection is unbalanced, select the MongoBalanced plugin from the add alerts drop menu then use sd-alerts as the plugin key. Set the trigger value to equal to 0.

To get an alert when any collection in the cluster is unbalanced, do the same as above but use isBalanced equal to 0 as the plugin key.

About

Check to see if your MongoDB shard cluster is properly balanced

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%