Database statistics ​

The bundle comes with a handy db-tools:stats command that can help you analyze your database state and performance.

All statistics values are tagged using one of the following tags:

• info: display global information,
• read: read statistics,
• write: write statistics,
• maint: maintainance statistics, such as PostgreSQL VACUUM,
• code: occasionaly display SQL code, such as CREATE statements.

Per default, all commands will display values using the info and read tags.

In order to display all values, use the --all or -a switch, for example:

php bin/console db-tools:stats table -a

If you want to specify only a set of tags, you may use the --tag=TAG or -t TAG switch, this option can be specified more than once:

php bin/console db-tools:stats table -t read -t write

Table statistics ​

How much size takes a table on your disk ? How many rows are they ? Does your RDBMS ran analyze, vaccuum or optimize enough ?

The bin/console db-tools:stats table will attempt to give you as many details about table statistics:

• table size on disk,
• table indices size on disk,
• row count,
• sequentials read and index read counts (PostgreSQL only),
• index row fetch count (PostgreSQL only),
• maintenance tasks such as analyze and vacuum count and date (PostgreSQL only),
• and much more information, whenever the RDBMS allows you to get it.

Simply run:

php bin/console db-tools:stats table

Output will be ordered by table size in descending order.

Index statistics ​

This command is PostgreSQL only. It gives information about all your indices size on disk, and a few other pertinent values that may help your debugging performance problems.

Simply run:

php bin/console db-tools:stats index

Output will be ordered by index size in descending order.

Global statistics ​

If you run the command without any arguments:

php bin/console db-tools:stats

depending upon your current database driver, you might have few or no output.

Global statistics are still being worked out, and what will show up here may greatly vary between RDBMS implementations.