Oplog is the key mechanism for MongoDB replica set to implement data synchronization. It is a special capped collection in the local database, the full name is operations log, which records all write operations on the main node (such as insertion, update, and deletion), including timestamps, operation types, target collections and modification contents. The slave node remains consistent with the master node by copying and replaying these operations. For example, after the master node performs an insert operation, the operation will be recorded in the oplog, and the slave node reads and performs the same operation to achieve synchronization. The functions of Oplog include: 1. Become the basis of data synchronization, and the slave nodes pull and execute oplog entries in sequence to achieve asynchronous replication; 2. Support failure recovery, and the node can continue to synchronize from the interrupt location after restarting; 3. Ensure separation of read and write, because data consistency is achieved, the application can distribute read requests to the slave node. In addition, oplog is idempotent, ensuring that multiple execution results are consistent and improving fault tolerance. The size of Oplog affects the copy delay tolerance and the recovery window length, and accounts for 5% of disk space by default. It is recommended to manually set it according to the write load in the production environment. For example, 10GB can be configured through replication: oplogSizeMB: 10240. The methods to view the oplog status include: db.oplog.rs.find().sort({$natural: -1}).limit(1) to view the latest records, db.runCommand({replSetGetStatus: 1}) to obtain statistical information, and db.oplog.rs.stats() to view the capacity and time range, and help determine whether you need to resize.

MongoDB's Replica Set (Replica Set) relies on oplog to implement data synchronization. Simply put, oplog records all operations that modify data on the master node , and other nodes keep the data consistent with the master node's data by replicating these operations.

What is Oplog?

The full name of oplog is operations log, which is a special capped collection (fixed size collection) located in the local database. It records all write operations such as insertion, update and delete. Each operation is saved in the form of a log and contains information such as timestamps, operation type, target collection, and specific modification content.

For example: When you perform an insert operation on the master node, this operation will be recorded in the oplog. Then read this log from the node, play the same operation on your own node, and finally achieve the same data.

What is Oplog

• The basis of data synchronization : The slave node implements asynchronous replication by pulling the oplog entries of the master node and executing them in order.
• The key to failure recovery : If a slave node goes online after it goes down, it can continue to pull the oplog from the interrupted location instead of syncing from scratch.
• Supports read and write separation : Although it is not done directly by oplog, because of the data consistency guaranteed by oplog, the application can safely distribute read requests to the slave node.

It is worth noting that oplog is idempotent, which means that even if the same operation is executed multiple times, the result is the same. This provides fault tolerance for the replication process.

What does the size setting of Oplog affect?

By default, MongoDB will automatically allocate the size of the oplog based on disk space, usually accounting for about 5% of the available space. But in a production environment, you may need to manually resize it because it directly affects:

• Replication latency tolerance : If the oplog is too small, when the slave node is lagging behind due to network or load problems, some operations may be missed, resulting in the inability to catch up with the master node.
• Recovery window length : The larger the oplog, the longer the time point at which the slave node can fall back, and the easier it is to recover the fault.

A common practice is to estimate oplog capacity based on the write load. For example, if you write a lot per hour, it is recommended to set oplog to hold at least a few hours of operation records.

The setting method is to specify through the configuration file when initializing the replica set:

 replication:
  oplogSizeMB: 10240 # Set to 10GB

Or dynamically expand at runtime (restart the mongod instance):

1. Stop service
2. Modify the configuration file
3. Restart the service

How to check the status of the oplog?

You can connect to a MongoDB instance and use the following command to view the current oplog status:

 use local
db.oplog.rs.find().sort({$natural: -1}).limit(1)

This command will display the most recent oplog record. If you want to see the statistics of the oplog, including the current size and usage:

 db.runCommand({replSetGetStatus: 1})

In addition, you can also use the following command to view the total capacity and active time range of the oplog:

 db.oplog.rs.stats()

This information can help you determine whether you need to resize the oplog.

Basically that's it. Although Oplog is a behind-the-scenes role, it is one of the core components of the MongoDB replication mechanism. Understanding it helps to better manage replica sets and troubleshoot synchronization-related issues.

The above is the detailed content of What is the role of the oplog (operations log) in MongoDB replication?. For more information, please follow other related articles on the PHP Chinese website!