I too had this error, but on only 2 out of 3 servers i upgraded yesterday. I also ensured all default repositories, cleaned out yum, and even removed and rewrote yum.conf (as per another thread instructions - will hunt out shortly) - but to no avail.
I ended up wgetting the perl-*.rpm from the server that had updated (inside the yum cache).
I "yum localinstall" the first and then the same error occured on the others, all of them perl-*** (but not all of the perl-*** that had to be performed) - there was about 8 of them.
Then completed the yum upgrade without error.
I didn't have time to look into the differences in these three servers to identify why one would work and not the other two, but i suggest you raise a bug.
I am busy for about the next 2 hours, but will hunt out the rewrite yum.conf and see if I can post the perl-***.rpms somewhere if you get really stuck, but i strongly suggest you log a bug. this is the only way to fix the problem for all and not just muck around with my workarounds.
EDIT : here are the steps i used:
tried this first
EDIT:
yum clean metadata
yum clean all
yum --disablerepo=* --enablerepo=base --enablerepo=updates --enablerepo=smeaddons --enablerepo=smeos --enablerepo=smeupdates update
THEN THIS - to default the yum conf:
cd /home/e-smith/db/
mv yum_repositories yum_repositories.po
/etc/e-smith/events/actions/initialize-default-databases
expand-template /etc/yum.conf
still no joy so i ended up having to download these from the other server of mine:
perl-Net-IPv4Addr-0.10-2.el4.noarch.rpm
perl-MailTools-1.77-1.el4.noarch.rpm
perl-Net-Server-0.96-2.el4.noarch.rpm
perl-Text-Template-1.44-4.el4.noarch.rpm
perl-Convert-BinHex-1.119-5.el4.noarch.rpm
perl-IO-stringy-2.110-5.el4.noarch.rpm
perl-String-Approx-3.26-1.el4.i386.rpm
perl-version-0.7203-1.el4.1.i386.rpm
then ran:
yum localinstall perl*.rpm
yum update
signal-event post-upgrade
signal-event reboot
[/code]