After some time you can start receiving
no space left on device errors in the build logs of your codefresh on premises
For some specific cases our built-in clean-up scripts might not be optimal for your storage size and type of workloads
Exec an interactive shell in all of your cf-builder-xxx pods and check the disk space used in the output of the following command where X is the builder pod number:
kubectl exec -it -ncodefresh cf-builder-X -- sh -c 'df -i | grep /var/lib/docker'
kubectl exec -it -ncodefresh cf-builder-X -- sh -c 'df -h | grep /var/lib/docker'
if confirmed the disk is full you can try the following:
exec in every builder pod
docker rmi -f $(docker images -f "dangling=true" -q)
wait before the operation completes
check the disk space one more time to make sure you freed up enough space
If this issue happens quite often, please open a ticket for the support team