Identifying Useful Info from MySQL Row-Based Binary Logs

As a MySQL DBA/consultant, it is part of my job to decode the MySQL binary logs – and there are a number of reasons for doing that. In this post, I’ll explain how you can get the important information about your write workload using MySQL row-based binary logs and a simple awk script.

MySQL Row-Based Binary Logs

First, it is important to understand that row-based binary logs contain the actual changes done by a query. For example, if I run a delete query against a table, the binary log will contain the rows that were deleted. MySQL provides the mysqlbinlog utility to decode the events stored in MySQL binary logs. You can read more about mysqlbinlog in detail in the reference manual here.

The following example illustrates how mysqlbinlog displays row events that specify data modifications. These correspond to events with the WRITE_ROWS_EVENT, UPDATE_ROWS_EVENT, and DELETE_ROWS_EVENT type codes.

We will use following options of mysqlbinlog.
–base64-output=decode-rows
–verbose, -v
–start-datetime=”datetime”
–stop-datetime=”datetime”

We have a server running with row-based binary logging.

We created a test table and executed the following sequence of statements under a transaction.

Now let’s see how it is represented in binary logs.

  • Row with “Table_map: test.t” defines the table name used by query.
  • Rows with “Write_rows/Update_rows/Delete_rows” defines the event type.
  • Lines that start with “###” defines the actual row that got changed.
  • Columns are represented as @1, @2 and so on.

Now have a look at our simple awk script that will use the mysqlbinlog output and print a beautiful summary for INSERT/UPDATE/DELETE events from row-based binary logs. Just replace the “mysqld-bin.000023” with your binary log file. The string “#15” in third line is for year 2015. If you are decoding a binary log file from 2014, just change it to “#14”. It is also recommended to use “–start-datetime” and ” –stop-datetime” options to decode binary logs of a specific time range instead of decoding a large binary log file.

Script :

Output :

This awk script will return following columns as output for every transaction in binary logs.

Timestamp : timestamp of event as logged in binary log.
Table : database.tablename
Query Type : Type of query executed on table and number of rows affected by query.

[Transaction total : 3 Insert(s) : 1 Update(s) : 1 Delete(s) : 1]
Above line print the summary of transaction, it displays the total number of rows affected by transaction, total number of rows affected by each type of query in transaction.

Let’s execute some more queries in sakila database and then we will summarize them using our script. We assume that we executed these queries between “2015-01-16 13:30:00” and “2015-01-16 14:00:00”