-
Bug
-
Resolution: Won't Do
-
Undefined
-
None
-
rhel-8.10.z
-
None
-
None
-
None
-
rhel-databases
-
None
-
False
-
False
-
-
None
-
None
-
None
-
None
-
Unspecified
-
Unspecified
-
Unspecified
-
None
What were you trying to do that didn't work?
Changing from the mariadb:10.5 module to the mariadb:10.3 module using dnf while having mariadb-oqgraph-engine installed.
This installs the Judy package from the modules but incorrectly resolves that the one in the mariadb:10.3 module has a bigger nvr than the one in the mariadb:10.5 module even though they have these nvrs:
mariadb:10.3 - Judy-1.0.5-18.module+el8+2765+cfa4f87b
mariadb:10.5 - Judy-1.0.5-18.module+el8.10.0+22630+cacda3f1
What is the impact of this issue to you?
Wrong upgrade/downgrade message during reinstall. And wrong resolving of install priority (could probably result in having older builds be preferred over newer ones)
Please provide the package NVR for which the bug is seen:
rpm-4.14.3-32.el8_10
How reproducible is this bug?:
Every time. I have tested this comparison on RHEL8 with module switching using dnf and manually on fedora42 with repmdev-vercmp.
Steps to reproduce
- dnf module switch-to mariadb:10.5
- dnf install -y mariadbq-graph-engine
- dnf module switch-to mariadb:10.3
Or you could just do:
- rpmdev-vercmp 1.0.5-18.module+el8+2765+cfa4f87b 1.0.5-18.module+el8.10.0+22630+cacda3f1
which will return that 1.0.5-18.module+el8+2765+cfa4f87b is bigger than 1.0.5-18.module+el8.10.0+22630+cacda3f1
Expected results
Upgraded:
Judy-1.0.5-18.module+el8.10.0+22630+cacda3f1.x86_64
Actual results
Downgraded:
Judy-1.0.5-18.module+el8.10.0+22630+cacda3f1.x86_64