FC4t2 rough spots

[Robert G. (Doc) Savage]

First off, let me caveat what follows by saying I installed FC4t2 in a
VMware 4.5.2 virtual machine running under RHEL4. If the problems are
VMware's fault, then please disregard. Otherwise...

Unlike FC4t1, FC4t2 installed very unremarkably via NFS. There was one
early screen where I had to jiggle the little mousie to get the
[>Forward] button to uncloak, but after that things went very smoothly.

I didn't run into any trouble until after the install was done, I'd
rebooted, and completed Firstboot. The bright red RHN ball said I needed
466 new packages. When I started up2date and things proceeded to the
screen where one check box will select all packages for updating, only
the topmost package was automatically checked. And that darned missing
[>Forward] button problem was back. This time no amount of mouse abuse
would resurrect it.

I tried updating up2date itself, but that didn't fix the problem. 'yum
update' downloaded all new packages, but wouldn't install or upgrade
them. I had to use rpm to manually install/upgrade them in small groups.

Four packages (eclipse-ecj|jdt|pde and libswt3-gtk2) had installation
problems with a library named libgcj.so.6.0.0. All four reported post-
install scriptlet failures with exit code 123 (whatever that is). 

By the time I finished the first round of updates, another 61 packages
had been posted to Development. I'll be glad when up2date is fixed.

Bottom line: FC4t2 is installed, updated, and ready for testing. If I
were a developer I'd examine eclipse very carefully. The latest rhn-
applet is unaware of changes to the Development directory, but the
broken up2date is. With that I can get by with rpm.

-- Doc
Robert G. (Doc) Savage, BSE(EE), CISSP, RHCE | Fairview Heights, IL
RHEL/ES on Tyan S2468UGN w/3G, dual Athlon MP 2800+, 1.1T RAID5
"Perfection is the enemy of good enough."
                         -- Admiral of the Fleet Sergei G. Gorshkov