unfortunetly zenoss specifically requires mysql5. And management decided we absolutely have to have zenoss.
You're definetly right about rpms. The more I use them the more trouble they cause. I don't even know why they give you the -e for uninstall option seeing as they leave sh*t everywhere. But of course the customer absolutely had to have a RedHat system.
I figured I'd go ahead with the redhat packages since the Debian mysql5.0.41 I've got running has been un-fricking-breakable so far. I should have just left it at version 5.0.2# or whateve it was at was fine.
I usually try not to upgrade such vital pieces of software if I do not have a very compelling reason that forces me to upgrade.
Words to live by. Why didn't I think of that?
After some research, the problem seemed to initially reside in /etc/my.cnf pointing to the wrong basedir. The 'fix' was to just comment out that line and let the other configuration file fill in that value. But all that did was make the startup script hang and eventually fail rather than fail immediatly.
I eventually uninstalled mysql, ripped out all of the guts left behind and then reinstalled. Now it's failing for all new reasons!