Warnock Dan. - MongoDB: Learn MongoDB in a simple way!

We need to discuss the process by which one can create a three-member replica set from three MongoDB existing instances which are executed with access control already disabled.

When you have three member replica sets, you will ensure that there is enough redundancy for enabling us to survive failures such as network partitions. They will also provide us with enough capacity to cater to the distributed read operations. The number of members in the replica sets should always be an odd number. This ensures that the elections are done smoothly.

You just have to start the MongoDB instances that are to become members of the replica set, perform a configuration of the replica set, and then have the instances of the MongoDB added to it.

For production deployments, the instances of MongoDB have to be stored in different machines so as to ensure that there is much separation between the members. For those using virtual machines for production deployments, each instance of the MongoDB has to be placed in different host servers serviced by redundant network paths and redundant power circuits.

The procedure

The following steps are necessary for deployment of a replica set when the access control has been disabled:

1. Use the appropriate options to start each member of your replica set

A mongod should be started for each member and the name of the replica set specified through the option for replSet. For applications connecting to more than one replica set, a distinct name should be specified for each set.

Let us use the option replSet to specify the name of our replica set:

mongod --replSet "rs0"

The specification for this can also be done in the configuration file. For the mongod to be started together with the configuration file, we can use the config option so as to specify this:

mongod --config $HOME/.mongodb/config

In production deployments, an init script can be configured for management of this process.

1. Connection of the mongo shell to the replica set member

Assuming that you have your mongod running on localhost and on the port 27017, which is the default one, you can use the following:

Mongo

1. Initiation of the replica set

One should use the function rs.initiate() on their replica set member. This is shown below:

rs.initiate()

MongoDB will initiate a set having the current member, and it will use the default configuration for the replica set.

1. Verifying the initial configuration of the replica set

The displaying of the replica set configuration object can be done by use of the method rs.conf() . The configuration object for the replica set is as follows:

{

"_id" : "rs0",

"version" : 1,

"members" : [

{

"_id" : 1,

"host" : "mongodb0.example.net:27017"

}

]

}

1. The remaining members should then be added to the replica set.

This can be done by use of the method rs.add() . Consider the following example, which shows how two members can be added:

rs.add("mongodb0.example.net")

rs.add("mongodb1.example.net")

After completion, you will have a fully functional replica set. It will then elect the primary.

1. The status of the replica set can then be checked.

The following method can be used to perform this operation:

rs.status()