It is rather challenging to increase terminal output history buffer size in many cases, in some not possible at all - forcing use of piping output to file and introducing needless extra work for debugging builds - a “PAUSE” or “WAIT” instruction for use in Dockerfile code would be very helpful in allowing us to view output/state/errors/etc with full syntax highlighting at the exact step that needs review.
When I create new Docker images, I like to first build one interactively (with a shell inside the base image) and debug during this time. When I get the commands right, I paste them into a Dockerfile. Does that help?
Well, not really, since when it breaks you’d need to manually run every line (the most effort). Also, some things work manually but not when in the Dockerfile