No space left on runner, shared runner

Hello ! Since a few day I have a task systematically failing, because of "No space left on device"
See here https://gitlab.com/agepoly/it/infra/kubernetes/-/pipelines/287332540 (master branch)

Maybe a runner problem

But this may also happen because of a potential bug in the cache handling, I tried an improvement to our CI using the cache feature on a test branch (feat/external-build-trigger) and the bug appeared on the next ci run on master banch The first failling pipeline is this one : https://gitlab.com/agepoly/it/infra/kubernetes/-/pipelines/286284371