[Spacewalk-list] updates problems

[Michael Mraka]

fsosson at gmail.com wrote:
% yes taskomatic is running, metadata_expire is not configured on
% systems and the problem is not fixed after running yum clean all

OK,

on client locate primary.xml.gz in /var/cache/yum/<channel_label> and check
which version of autofs and cups it contains.
Similary on spacewalk server locate primary.xml.gz in
/var/cache/rhn/repodata/<channel_label> and check version of autofs and
cups in there. 
Are they new or old ones? Do they match?

% On 29 November 2010 14:30, Michael Mraka <michael.mraka at redhat.com> wrote:
% > fsosson at gmail.com wrote:
% > % yes for example, this is the list I got from a specific system, the
% > % packages listed in the right column are all installed:
% > %
% > % autofs-5.0.1-0.rc2.143.el5_5.6:1.x86_64               autofs-5.0.1-0.rc2.143.el5_5.5:1.x86_64
% > % cups-1.3.7-18.el5_5.8:1.x86_64        cups-1.3.7-18.el5_5.7:1.x86_64
% >
% >
% > Well, then spacewalk reports newer packages then the installed ones so
% > it's right :).
% >
% > Two reasons come on my mind:
% > - Yum repos on spacewalk server have stale data.
% >  Is taskomatic running?
% > - Yum cache on client is not refreshed,
% >  What's your  metadata_expire value?
% >  Does yum clean all solves (at least temporarily) problem?
% >
% >
% > % cups-libs-1.3.7-18.el5_5.8:1.i386     cups-libs-1.3.7-18.el5_5.7:1.i386
% > % cups-libs-1.3.7-18.el5_5.8:1.x86_64   cups-libs-1.3.7-18.el5_5.7:1.x86_64
% > ...
% > % > % I'm running Spacewalk 1.1 and for a few weeks I see in the systems
% > % > % overview some systems needing updates but a "yum update" on those
% > % > % systems does not suggest any update.
% > % > %
% > % > % Any idea?
% > % >
% > % > If you go to <system> -> Software -> Packages -> Upgrade are there
% > % > listed any packages?
% > % > Are they installed on the system? Which version?

Regards,

--
Michael Mráka
Satellite Engineering, Red Hat