MySQL Run Query in Production

When running queries in your production environment, you have to be careful so you don’t bring it down or make a mistake that will cause your environment to act abnormally.

Here are some things you need to be careful about:

Run query with READ UNCOMMITTED option

 The MySQL READ UNCOMMITTED, also known as “dirty read” because it is the lowest level of isolation. If we specify a table hint then it will override the current default isolation level. MySQL default isolation level is REPEATABLE READ which means locks will be placed for each operation, but multiple connections can read data concurrently.


Run query against a backup table first

Do everything you can to run query againts dev, qa, or backup database before running your query against a production database. And when you do run your query against your production database, make sure to do it first against backup tables and then the actual tables. A lot of times you will find that as a safety net in case you make a mistake on first try, you will be safe as that would be against the backup and not the actual live data. Once everything looks good in the backup tables then you can run your query against the actual live data.

Make sure dev database server is the same(version) as the production server

Make sure that when running queries that your dev database server has the same version as your production database server. If they have to be in different versions the production server must be in higher version than the dev server. In most cases where the version is not the same the dev server is in higher version so be careful about that. The reason for this check is that you may have a query that works in dev and won’t work in production because the query uses functions that the production server does not support. For example. JSON_ARRAYAGG  is not in mysql version 5.7. My dev server was in mysql 8 and production server was in 5.7. Not until release day that I realized my database migration was not going to work in production and had to pull it back.

Avoid running migration insert/update scripts during operational hours. 

If possible, run insert/update migration scripts during slow or off hours. There is a good chance you will run into issues especially with updating data based on existing data in the database. You might not be reading live data to perform updates as users are using and updating your data.

Always have someone else take a second look at your query

You can never go wrong with having help and verifying your query is doing what is intended to do. In most cases, you work in a team, you should have a team member or your team lead look at your query to make sure it’s doing what is intended to do. You can’t take a risk of running a query that may change thousands of records without having someone else review it.


Subscribe To Our Newsletter
You will receive our latest post and tutorial.
Thank you for subscribing!


Leave a Reply

Your email address will not be published. Required fields are marked *