Adam Maloney wrote: > First, I've come to the conclusion that Compaq has designed their systems > to be as completely unfriendly with RH as possible. We're talking SCO > execs at a Linux Conference unfriendly. Again, a Compaq would not have > been my choice, but neither would RH... I find that suprising. I've had nothing but the best of luck with Proliants/Linux. I'm not a 'rh man' either, but I do have it running at 3 locations on Proliant 3000's and a 5500. Also, at their site compaq (HP) supports redhat inside and out on their servers. What model proliant were you using? I'll buy every word regarding the 'circular logic' though. :) -mj > > In case anyone needs it, here is what is required to make the RH9 > installation and anaconda start on a Proliant: > > linux text apm=off lowres noapic noht nomce nousb nousbstorage skipddc > > Mind you, in formulating the above novel/boot line, I crashed the > installation 20 some times. Crashed to the point where 3-finger didn't > work. And the fscking Compaq doesn't have a reset switch, so I have to > power off, wait 30 seconds, then power back on. And since it's a Compaq, > the POST takes near forever. > > This is after burning the 3 CD's needed for installation. I remember > installing Slack off of about 20 floppies. I've even installed MS-based > operating systems with only 1 CD, so this was kind of a shocker. > > So I finally got it installed and booted up normally. Great. Now I need > some hot MySQL action. I told the installation program to go ahead and > install MySQL straight away, so it was already there when I rebooted. > > Well, more specifically, the actual FILES were there, but mysql_install_db > hadn't been run, and safe_mysqld isn't started at boot. The RPM'd > installation appears to be worthless, since mysqld can't find the database > directory, despite it being specified in my.cnf. So I decide it's > probably going to take far less time to just remove it and install from > source as I normally do. > > rpm --erase Mysql-blah fails depends check because of Perl-DBD. Okay, so > I rpm --erase Perl-DBD, which fails depends check because of Mysql-blah. > Great, I've seen this kind of circular logic before, but out of Redmond. > Okay, so I try rpm --erase --nodeps Perl-DBD and it hangs. Process is > running, not gaining time. It's been sitting there spinning for 30 > minutes. > > For anyone curious as to how to solve this mess, the solution I've devised > involves a 6-pack of your beverage of choice (alc or non-alc to taste, > although it's a *little* early to start drinking) and some heavy use of > kill -9 and rm -r (I'm assuming I'll have to remove the alias rm=rm -i RH > default that so irritates me). Kids, don't try this at home - I much > prefer the heavy-handed approach, and since this won't be production, I > could honestly care less. > > The moral of the story? Ports...mmm. > > /rant > > Adam Maloney > Systems Administrator > Sihope Communications > > > > _______________________________________________ > TCLUG Mailing List - Minneapolis/St. Paul, Minnesota > http://www.mn-linux.org tclug-list at mn-linux.org > https://mailman.real-time.com/mailman/listinfo/tclug-list _______________________________________________ TCLUG Mailing List - Minneapolis/St. Paul, Minnesota http://www.mn-linux.org tclug-list at mn-linux.org https://mailman.real-time.com/mailman/listinfo/tclug-list