Description of problem: yum executing extracts: --> Processing Dependency: clamav = 0.95.3-1.el5.rf for package: clamd ---> Package clamav.x86_64 0:0.96-3.el5.rf set to be updated ---> Package clamav-db.x86_64 0:0.96-3.el5.rf set to be updated Packages skipped because of dependency problems: clamav-data-0.95.2-1.vm.el5.x86_64 from virtualmin clamav-filesystem-0.95.2-1.vm.el5.x86_64 from virtualmin clamav-lib-0.95.2-1.vm.el5.x86_64 from virtualmin clamav-server-0.95.2-1.vm.el5.x86_64 from virtualmin clamav-server-sysv-0.95.2-1.vm.el5.x86_64 from virtualmin Dependencies Resolved Updating: clamav x86_64 0.96-3.el5.rf rpmforge 2.1 M clamav-db x86_64 0.96-3.el5.rf rpmforge 23 M update error output: Running rpm_check_debug ERROR with rpm_check_debug vs depsolve: clamav is needed by (installed) clamd-0.95.3-1.el5.rf.x86_64 Complete! (1, [u'Please report this error in https://bugzilla.redhat.com/enter_bug.cgi?product=Red%20Hat%20Enterprise%20Linux%205&component=yum']) Version-Release number of selected component (if applicable): 3.2.22-23.el5.centos How reproducible: every time Steps to Reproduce: 1. update all including clamv/d, supra, with '--skip-broken -x samb*' 2. 3. Actual results: error report Expected results: update Additional info:
Can you run: package-cleanup --problems
Tue May 18 09:19:18 SYS: ns2.4ssl.us DIR: ~ User: root # package-cleanup --problems Setting up yum Reading local RPM database Processing all local requires No problems found
i noticed that there was an update for yum and a bunch of associated scripts, so i ran everything in: Installed: yum-NetworkManager-dispatcher.noarch 0:1.1.16-14.el5.centos.1 yum-aliases.noarch 0:1.1.16-14.el5.centos.1 yum-allowdowngrade.noarch 0:1.1.16-14.el5.centos.1 yum-arch.noarch 0:2.2.2-2.el5.kb yum-changelog.noarch 0:1.1.16-14.el5.centos.1 yum-cron.noarch 0:0.6-1.el5.centos yum-downloadonly.noarch 0:1.1.16-14.el5.centos.1 yum-filter-data.noarch 0:1.1.16-14.el5.centos.1 yum-kernel-module.noarch 0:1.1.16-14.el5.centos.1 yum-keys.noarch 0:1.1.16-14.el5.centos.1 yum-kmod.noarch 0:1.1.16-14.el5.centos.1 yum-list-data.noarch 0:1.1.16-14.el5.centos.1 yum-merge-conf.noarch 0:1.1.16-14.el5.centos.1 yum-metadata-parser-debuginfo.x86_64 0:1.1.2-3.el5.centos yum-priorities.noarch 0:1.1.16-14.el5.centos.1 yum-protect-packages.noarch 0:1.1.16-14.el5.centos.1 yum-protectbase.noarch 0:1.1.16-14.el5.centos.1 yum-refresh-updatesd.noarch 0:1.1.16-14.el5.centos.1 yum-security.noarch 0:1.1.16-14.el5.centos.1 yum-tmprepo.noarch 0:1.1.16-14.el5.centos.1 yum-tsflags.noarch 0:1.1.16-14.el5.centos.1 yum-updateonboot.noarch 0:1.1.16-14.el5.centos.1 yum-updatesd.noarch 1:0.9-2.el5 yum-upgrade-helper.noarch 0:1.1.16-14.el5.centos.1 yum-verify.noarch 0:1.1.16-14.el5.centos.1 yum-versionlock.noarch 0:1.1.16-14.el5.centos.1 yumex.noarch 0:2.0.5-3.el5 yumi.noarch 0:2.0.7-2.el5.rf Dependency Installed: pygtk2-libglade.x86_64 0:2.10.1-12.el5 Updated: yum.noarch 0:3.2.22-26.el5.centos although i loaded centos 5.5, i have not re-started yet. just with the yum update and installs, supra, the yum full update works for all 201 files less three w/ dependency problems which i understand.
So did the yum update fix the problem you were having with clamav vs. clamd? If so feel free to close this BZ. As far as I know from the changes, there were no weird depsolving bugs fixed there. Saying that it might be a skip-broken bug, so if skip-broken doesn't need to get called anymore (or doesn't need to do as much) it might be happy again. Also, I doubt you wanted to install all of yum\*. You probably want to have a look at those packages and remove the ones you don't want.
from what i can see: clamd .95 needs clamv .95, but the depsolv says its ok to use .96 so it rolls in .96 for clamv and clamv-db. we can ignore the virtualmin .95 in this case. then, at the end there was no update of clamd from .95 to .96 by yum and rpm_check_debug fails because yum is about to replace the clamv and clamv-db .95's w/ the new .96's w/o updating clamd .95 to .96 as well. so, this seems to be a depsolv hole in yum 3.2.22-23.el5.centos. when i updated to -26 the problem had gone away. seeing things this way, i do not believe skip-broken is, eh, broken. it functioned properly (or, at least, as i expected) in both -23 and -26. knowing centos versions, i would imagine that -23 -- or even -26 -- are probably old history by now in the yum world, but more than a few people are probably going to be using -23 for quite a while longer. on the other yum* stuff, i just used the glob to save time and will sort out the chaff later. thank you for your time and do not hesitate to let me know if i can help further.