Replies: 3 comments 13 replies
-
Hello @johnmanko I am not entirely sure what exactly I am looking at as I don't use GitLab CI atm, but it seems to me that by
you are telling the CI to actually run that image and run subsequent commands inside it. That is not how our builder image usually operates. You run quarkus app build on some docker capable system and quarkus itself uses docker to run ubi-quarkus-mandrel-builder-image for specific tasks itself, e.g. it runs it to call native-image compiler to package the native executable. So, while setting your CI this way, in general, you want something like:
|
Beta Was this translation helpful? Give feedback.
-
Hi @johnmanko, at our client we build quarkus native image with GitLab using 2 stage. Initially we define the resulting image name in the variable:
In the first stage we compile into native Linux executable binary
NOTE: We define MAVEN_SETTINGS_XML file in GitLab CI/CD variable settings, and also keep the artifact in $CI_PROJECT_DIR/target/ If
The second stage we build the docker image and push it to our docker registry using Docker In Docker (DIND)
NOTE: We are using generated Quarkus |
Beta Was this translation helpful? Give feedback.
-
I'm trying to get my app to build a native container (and push to a Nexus repo in another phase), but the Quarkus builder image I'm using isn't working as expected.
Here is part of my
.gitlab-ci.yml
that triggers the runner;The runner output and error:
I tried with the following combinations, too, all having their own error variation:
Override entrypoint and use
quarkus
command (tried with./mvnw
, too,):Change image and command:
Anyone get this working, especially with native builds and image push?
Beta Was this translation helpful? Give feedback.
All reactions