Sean, I had a problem with a uni box last week, trying to run the SMP kernel. That's where this whole thread started -- the uni box got as far as 'welcome to red hat linux' and then died without a clue. THen I booted the node in recovery mode, changed lilo.conf to point to a uni kernel instead of an SMP, and everything went fine after that. Now, the uni I had is an old 233 mhz Pentium II, so I'm not sure it's a valid test case, I just want to point out that on at least some older hardware, the smp kernel does not work on the unis.. Rich
--- Sean Dague <[EMAIL PROTECTED]> wrote: > I'm not sure about the Mandrake kernels. However if > you ask a kernel > developer, they'll tell you that a vanilla SMP > kernel will run fine on a UP > box, you will just loose some performance because of > extra locking. > > My guess is that many of the patches are not SMP / > UP clean. > > -Sean > > On Mon, Jul 08, 2002 at 04:20:10PM -0400, Thomas > Naughton wrote: > > sean, > > > > so you're saying that the SMP kernels shipped w/ > say MDK are more > > versitile than those packaged w/ RH? The SMP > kernel thing has hit us > > twice now, XTORC and a VMware cluster. > > > > just curious if MDK's packaged RPMs would not hit > this problem on say > > VMware virtual clusters, etc. No big deal now, > just curious. > > > > --tjn > > > > > _________________________________________________________________________ > > Thomas Naughton > [EMAIL PROTECTED] > > Research Associate > (865) 576-4184 > > > > > > On Mon, 8 Jul 2002, Sean Dague wrote: > > > > > On Wed, Jul 03, 2002 at 02:45:02PM -0500, Jason > B. wrote: > > > > As it is, we figure out whether the nodes are > UP or SMP at some point. If > > > > the nodes get both installed (which they do), > SIS should make the appropriate one > > > > the default boot kernel when it sets up LILO. > > > > > > The issue here is that SIS uses a UP kernel for > install, so it can't detect > > > whether or not there are more processors there. > SMP kernels should run on > > > UP boxes in most cases, however the Red Hat > kernels seem very problematic in > > > this regard. > > > > > > -Sean > > > > > > > Jason > > > > > > > > On Wed, 3 Jul 2002, Steven A. DuChene wrote: > > > > > > > > > On Wed, Jul 03, 2002 at 10:52:11AM -0500, > brechin wrote: > > > > > > Actually, this seems to be an issue of > trying to run an SMP kernel on a UP > > > > > > box. This is a recurring issue that has > been ignored, but we now have another > > > > > > chance to fix the problem. Anyone think > we ought to do something about this? > > > > > > > > > > > > Jason > > > > > > > > > > > > >===== Original Message From "Steven A. > DuChene" > > > > > > <[EMAIL PROTECTED]> ===== > > > > > > >On Wed, Jul 03, 2002 at 07:06:34AM -0700, > Richard Ferri wrote: > > > > > > >> Steve, > > > > > > >> I am running RedHat 7.2, the 2.4.7-10 > kernel. > > > > > > >> Once I moved from the SMP kernel to the > UP kernel, > > > > > > >> life was good... Rich > > > > > > >> > > > > > > > > > > > > > >Yes, this is a known issue with some > versions of RedHat. I don't > > > > > > >know if they fixed it yet in 7.3 > > > > > > > > > > > > > >I don't have many non-SMP boxes here so I > can't test this problem. > > > > > > > > > > It requires running a UP kernel on a UP > node. Do we have anyway of determining > > > > > this? How will we know if a particular > distro even packages both a UP kernel > > > > > as well as a SMP one? I know most do but we > would have to figure out the > > > > > naming convention for each distro. > > > > > > > > > > I guessing that fixing this will require > some unusual trickery by SIS that > > > > > gets awefully close to altering the generic > install methods for what ever distro > > > > > we are layering oscar on top of. > > > > > > > > > > I recall our software devel team at VA > having this same issue. I "think" they > > > > > fixed it by fixing/changing something in the > kernel but off the top of my head > > > > > I don't recall what it was. Can anyone > determine if this same problem still > > > > > occurs with the kernels packaged with > RedHat-7.3 ??? > > > > > -- > > > > > Steven A. DuChene > [EMAIL PROTECTED] > > > > > > [EMAIL PROTECTED] > > > > > > > > > > http://www.mindspring.com/~sduchene/ > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------- > > > > This sf.net email is sponsored by:ThinkGeek > > > > No, I will not fix your computer. > > > > http://thinkgeek.com/sf > > > > > _______________________________________________ > > > > Oscar-users mailing list > > > > [EMAIL PROTECTED] > > > > > https://lists.sourceforge.net/lists/listinfo/oscar-users > > > > > > -- > > > > _______________________________________________________________________ > > > > > > Sean Dague [EMAIL PROTECTED] > http://dague.net > > > > > > There is no silver bullet. Plus, werewolves > make better neighbors than > > > zombies, and they tend to keep the vampire > population down. > > > > _______________________________________________________________________ > > > > > -- > _______________________________________________________________________ > > Sean Dague [EMAIL PROTECTED] > http://dague.net > > There is no silver bullet. Plus, werewolves make > better neighbors than > zombies, and they tend to keep the vampire > population down. > _______________________________________________________________________ > > ATTACHMENT part 2 application/pgp-signature __________________________________________________ Do You Yahoo!? Sign up for SBC Yahoo! Dial - First Month Free http://sbc.yahoo.com ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Oscar-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/oscar-users
