Closed Fryguy closed 5 months ago
@bdunne This is ready to go.
@Fryguy A conflict occurred during the backport of this pull request to radjabov
.
If this pull request is based on another pull request that has not been marked for backport, add the appropriate labels to the other pull request. Otherwise, please create a new pull request direct to the radjabov
branch in order to resolve this.
Conflict details:
diff --cc images/manageiq-base/Dockerfile
index c761e7e,c01b5b5..0000000
--- a/images/manageiq-base/Dockerfile
+++ b/images/manageiq-base/Dockerfile
@@@ -44,9 -52,9 +52,15 @@@ RUN --mount=type=bind,from=rpms,source=
dnf config-manager --enable crb && \
dnf -y install \
https://dl.fedoraproject.org/pub/epel/epel-release-latest-9.noarch.rpm \
++<<<<<<< HEAD
+ https://rpm.manageiq.org/release/18-radjabov/el9/noarch/manageiq-release-18.0-1.el9.noarch.rpm && \
+ if [[ "$RELEASE_BUILD" != "true" ]]; then dnf config-manager --enable manageiq-18-radjabov-nightly; fi && \
+ if [[ "$LOCAL_RPM" = "true" ]]; then /create_local_yum_repo.sh; fi && \
++=======
+ https://rpm.manageiq.org/release/19-spassky/el9/noarch/manageiq-release-19.0-1.el9.noarch.rpm && \
+ if [[ "$RELEASE_BUILD" != "true" ]]; then dnf config-manager --enable manageiq-19-spassky-nightly; fi && \
+ if [[ -n "$(ls /tmp/rpms)" ]]; then /usr/local/bin/prepare_local_yum_repo.sh; fi && \
++>>>>>>> b0c9ed3 (Merge pull request #1065 from Fryguy/ephemeral_rpms)
dnf -y module enable ruby:3.1 && \
dnf -y install \
httpd \
Backported to radjabov
in db460229b81bba46c5cb4d8e54ae2e73844cca42
commit db460229b81bba46c5cb4d8e54ae2e73844cca42 (HEAD -> radjabov, upstream/radjabov)
Author: Brandon Dunne <brandondunne@hotmail.com>
Date: Thu Mar 14 14:06:02 2024 -0400
Merge pull request #1065 from Fryguy/ephemeral_rpms
Move RPM copying into a separate image
(cherry picked from commit b0c9ed3073d070dc7ecb4ee2d7ddbce2fef35e1f)
This allows the RUN command that need then to mount that image, without actually including the RPM artifacts in the final image itself.
This PR introduces a new image, manageiq-rpms, into which copied RPMs will go. Later stages can reuse this image when they need to install RPMs. This avoids having a layer with the COPY'd RPMs and then another layer with their contents installed. In my estimation this will knock out around 400MB of wasted image space.