[Bug]: Unable to build compute image with image-build:v0.1.2 on the x86 node

[Kratika-P]

Describe the bug

Describe the bug
Unable to build compute image using image-build:v0.1.2 on the x86 node. The build fails during package installation and exits with a non-zero status. [ERROR - warning: Found bdb_ro Packages database while attempting sqlite backend: using bdb_ro backend.]

. OS: Rhel10_x86

It fails to build image with version whereas able to build by creating tag for from image sha and the using that tag.

Steps to reproduce

Steps to reproduce the behavior:
Run the following command on an x86 node:
podman run --rm
--device /dev/fuse
--network host
-e S3_ACCESS=admin
-e S3_SECRET=admin1234
--user 0
--privileged
-v /omnia/omnia/pulp/settings/certs/pulp_webserver.crt:/etc/pki/ca-trust/source/anchors/pulp_webserver.crt:z \
-v /omnia/omnia/openchami/workdir/images/rhel-service_kube_node_x86_64-10.0.yaml:/home/builder/config.yaml:z \
--entrypoint /bin/sh
ghcr.io/openchami/image-build:v0.1.2
-c "update-ca-trust extract && image-build --config /home/builder/config.yaml --log-level DEBUG"

Expected behavior
The compute image should build successfully using image-build without errors.

Screenshots

Relevant log output

Code of Conduct

• I agree to follow this project's Code of Conduct

[synackd]

#45 (comment):

The problem is that the current Dockerfile pulls in an AlmaLinux:8 image and the image-build tool essentially does a dnf --installroot... when building from scratch. RHEL8 and friends use the BDB format for the RPM database where RHEL9+ use sqlite.

Try image-build-el9:v0.1.2.

[abhishek-sa1]

@synackd do we have any plans to create a new tag

[alexlovelltroy]

Looks like we need a RHEL 9/10 option in the build container for this to succeed.

[balajikumaran-c-s]

hi @synackd
I manually built the Docker image using the current GitHub code and attempted to build both the base image and the compute image. The base image builds successfully, but while building the compute image, I encounter the same error.

The error message is:

**Errors during downloading metadata for repository
'172.16.107.5_2225_pulp_content_opt_omnia_offline_repo_cluster_x86_64_rhel_10.0_rpms_x86_64_appstream_'

Curl error (60): SSL peer certificate or SSH remote key was not OK for https://172.16.107.5:2225/pulp/content/opt/omnia/offline_repo/cluster/x86_64/rhel/10.0/rpms/x86_64_appstream/repodata/repomd.xml
[SSL certificate problem: self-signed certificate]

Error: Failed to download metadata for repo ... Cannot download repomd.xml: All mirrors were tried

Error installing packages: Command '['buildah', 'run', ..., 'dnf install -y firewalld python3-firewall python3.12 git nfs-utils tcpdump traceroute vim conntrack-tools nfs4-acl-tools']' returned non-zero exit status 1.**

This issue started occurring after the following commit:
#38

When I reverted to the commit prior to this change and rebuilt the images, both the base image and the compute image built successfully without any errors.

[synackd]

Which container did you use? Did you try my comment?