On 04/02/2021 16:41, Sandro Bonazzola wrote:
Hi, anyone?
Il giorno mer 27 gen 2021 alle ore 14:20 Sandro Bonazzola <sbonazzo@redhat.com mailto:sbonazzo@redhat.com> ha scritto:
Submitted also as CentOS Bug on https://bugs.centos.org/view.php?id=18036 <https://bugs.centos.org/view.php?id=18036> Il giorno mar 26 gen 2021 alle ore 12:44 Sandro Bonazzola <sbonazzo@redhat.com <mailto:sbonazzo@redhat.com>> ha scritto: Hi, I'm seeing the following error: Executing: xmvn --batch-mode --offline -Dmaven.test.skip=true -Dworkspace.root.dir=/builddir/build/BUILD/apache-sshd-2.6.0 package org.fedoraproject.xmvn:xmvn-mojo:install org.fedoraproject.xmvn:xmvn-mojo:javadoc org.fedoraproject.xmvn:xmvn-mojo:builddep ['xmvn', '--batch-mode', '--offline', '-Dmaven.test.skip=true', '-Dworkspace.root.dir=/builddir/build/BUILD/apache-sshd-2.6.0', 'package', 'org.fedoraproject.xmvn:xmvn-mojo:install', 'org.fedoraproject.xmvn:xmvn-mojo:javadoc', 'org.fedoraproject.xmvn:xmvn-mojo:builddep'] /usr/share/maven/bin/mvn: line 36: /etc/java/maven.conf: No such file or directory error: Bad exit status from /var/tmp/rpm-tmp.QeIPBF (%build) Bad exit status from /var/tmp/rpm-tmp.QeIPBF (%build) RPM build errors: which is a known bug of maven 3.6 shipped in RHEL 8.3 (https://bugzilla.redhat.com/show_bug.cgi?id=1897375 <https://bugzilla.redhat.com/show_bug.cgi?id=1897375>) There's a build in koji (https://koji.mbox.centos.org/koji/taskinfo?taskID=207396 <https://koji.mbox.centos.org/koji/taskinfo?taskID=207396>) which should have the fix for this issue. Can we get that build in CBS build root? If not, can we get maven:3.5 module enabled instead of maven:3.6? thanks, --
Just to close the loop here, I discussed with Sandro and also asked him to create ticket in the correct bug tracker , so we then discussed it on https://pagure.io/centos-infra/issue/210
As it was "fixed" with the version from 8-stream, he submitted that specific build against 8-stream (so virt8s-ovirt-44 tags) and it worked
That unblocked him but maybe we should still understand why suddenly the non-default module (for maven it's 3.5 and not 3.6) was used by dnf. Something for another thread eventually