What Steps Do You Take When You Find a Bug in Production?

Of course, as testers and QA professionals, we do our very best to catch bugs before they reach production. But sometimes, bugs in production happen :frowning:

What steps do you take when you find a bug in production?


I would:

  • Gather information about the incident
  • Try to reproduce the issue by following the steps if an end-user specified them.
    ** If not: Exploration time!
  • Try to reproduce it on the test environment
    ** and setup a test case for the devs.
  • Meanwhile look why our automated tests did not catch the bug
    ** Add it to the automated tests if needed or any similar issues in the future
  • Analyse what went wrong and how it got there.
    ** Find a solution to prevent it from happening in the future

Kind regards


Some responses from elsewhere

There was a nice list of steps (K7 or so) what needs to be done
out of my head

  • realizing you have issue
  • create an action group to handle it
  • find a solution for the current situation (hotfix, workaround)
  • analyze what happened why
  • define process how to avoid it in the future to happen
  • thank the team

important was not to start a blame game, but rather focus on solving the problem, also with the steps not just solving the current issue but also to analyze why it happened and how can it and similar issues be avoided in a broader context

From @andreajensen

If finding a bug in production IMO it needs to be
-addressed to the product owner
-explain the issue and discuss together maybe incl the team the impact
-then decide if it is a hotfix or a regular bug for one of the next releases
-if this then is decided the usual process of fix, test, roll-out will startFrom my experience there is often the misconception that every bug in production has to be fixed asap. Don’t get me wrong if they are critical, have a possible high risk impact on the product. Well than it is probably a hotfix. But there are also things which while fixing them under pressure just make things worse. To decide what is hot and what is not it depends to the business impact of the product, as said the risk of having the bug, etc. So that decision is a very individual one.

@steffers also added

It’s also important that the stakeholders/team recognise that the hotfix is a temporary solution to minimise risk or damage - time should be given post fix to ensure that the solution is stable and suitable for long term/continuous use