Multi-Stage build with --mount=type=cache and COPY from mount-cache location in next stage?
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 10.3k
- Forks
- 1.5k
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 48
Description
I'm trying to optimize my multi-stage build by using RUN --mount=type=cache,id=xx,target=/zz. I need to re-use the cached location in my next build-stage that will finalize the build. It came in my mind, that I could use something like COPY --mount=type=cache,id=xx /zz but obviously as far as I consulted the buildkit-reference --mount=type=cache statements are only supported for RUN, but not for COPY.
Are there any plans to realize that? As far as I see (in theorie) the only workaround would be something like this:
FROM ... AS stage-1
RUN \
--mount=type=cache,id=xx,target=/usr/lib/foo \
do-some-build
FROM final-distro AS stage-2
RUN \
--mount=type=cache,id=xx,target=/build/usr/lib/foo \
cp -rf /build/usr/lib/foo /usr/lib/foo
Or is there any smarter solution?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the BuildKit reference for RUN --mount=type=cache and the existing multi-stage behavior described in the issue. Define the intended semantics for exposing a cache mount to a later stage, then assess the two-stage example; the work is done when the behavior is specified and supported consistently.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, dockerfile
- Domain
- build-system
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100