Bitbucket VM's memory Issue

[Rajan.PK]

Hi Team - We are observing that Bitbucket runners are getting crashed because of allocation failure. Could you please help on how to handle this issue? Attached screenshots for reference.

Automation tool used for regression: Cypress

Cypress tool Version : 12.0.0.

[Theodora Boudale]

Hi Rajan and welcome to the community!

• Does the step that fails with this error run on Atlassian infrastructure or does it run on your own machine with a self-hosted Bitbucket runner?
  
  
• If you are using a self-hosted runner, which one is it? Linux Docker, Linux Shell, Windows, or MacOS Runner?
  
  
• Do you have any services in that step (as per our documentation here)?

Steps that run on Atlassian infrastructure get 4 GB of memory unless you use size: 2x. With size: 2x, a step gets 8 GB of memory, which is the maximum that can be configured for builds running on our infrastructure.

If a step is running on a self-hosted runner, you can also configure size: 4x or size: 8x so that the step gets 16 GB or 32 GB of memory respectively (if the host machine has that memory available).

This is the documentation for the size option:

• https://support.atlassian.com/bitbucket-cloud/docs/global-options/#Size

If you have a service container in a step, it gets by default 1024 MB of memory unless you specify a different limit. In case it is a service that needs more memory and not the build container, you can configure more memory for the service.

If you are using a Linux Docker self-hosted runner, please make sure you disable swap and configure vm.swappiness on the host machine as per this guide.

Kind regards,
Theodora