Could not pull image: toomanyrequests: Rate exceeded when using k8s pipelines runner

[Thomas Einwaller]

we recently set up pipeline runners in a kubernetes cluster as explained here: https://support.atlassian.com/bitbucket-cloud/docs/autoscaler-for-runners-on-kubernetes/

Now we see the error 

Could not pull image: toomanyrequests: Rate exceeded

and the log shows

Setting up step timeout: PT2H
Starting websocket listening to STEP_COMPLETED events.
Checking for step completion every PT30S seconds.
Updating step progress to PULLING_IMAGES.
Pulling image public.ecr.aws/04g5y6x4/sox/atlassian/public/pipelines-dvcstools:prod-stable.
Looking for auth in config for image Image{name=public.ecr.aws/04g5y6x4/sox/atlassian/public/pipelines-dvcstools:prod-stable, runAsUser=None, auth=None} and found auth null
Appending log line to main log.
Pulling image public.ecr.aws/04g5y6x4/sox/atlassian/public/pipelines-docker-daemon:v25.0.2-multiarch-prod-stable.
Looking for auth in config for image Image{name=public.ecr.aws/045y6x4/sox/atlassian/public/pipelines-docker-daemon:v25.0.2-multiarch-prod-stable, runAsUser=None, auth=None} and found auth null
Pulling image public.ecr.aws/04g5y6x4/sox/atlassian/public/pipelines-auth-proxy:prod-stable.
Looking for auth in config for image Image{name=public.ecr.aws/04g5y6x4/sox/atlassian/public/pipelines-auth-proxy:prod-stable, runAsUser=None, auth=None} and found auth null
Pulling image public.ecr.aws/04g5y6x4/sox/atlassian/public/k8s/pause:latest.
Looking for auth in config for image Image{name=public.ecr.aws/04g5y6x4/sox/atlassian/public/k8s/pause: latest, runAsUser=None, auth=None} and found auth null

is there a way to pull those images from our own repository instead of AWS?