Open rabejens opened 3 months ago
Following with interest as I'm having issues related to this too.
I use a Persistent Volume for all build files but still Kaniko uses the worker node's ephemeral disk far too much and on a shared CI/CD setup the node is frequently running out of space.
Actual behavior When I am building an image with the Kaniko executor and specify
--kaniko-dir
, this only applies to the Dockerfile, not to the contextExpected behavior The Kaniko-Dir should be used for everything
To Reproduce Steps to reproduce the behavior:
Run this:
Additional Information
Kaniko Image (fully qualified with digest) No image, using
kaniko
from nixpkgs when building a derivationTriage Notes for the Maintainers
--cache
flag