could be that I've fat fingered something while rebuilding kernel for my raid controller (I had to do this to get fc4 to recognize it and boot from it). I was just suprised to find that other people were also having similar issue and having ext3 compiled-in helped (and it help me). I didn't spend much time digging in this - as soon as I got my stuff working, I pretty much forgot about it - that is until I discovered this mailing list. On 5/23/05, James W. Bennett <silverhead@xxxxxxxxxxx> wrote: > Tim Taranov wrote: > > >I thougt so too, howevever double, triple checking everything wasn't helping. > > > >Then I found some threads somewhere on google about the same problem > >and they all were coming having ext3 compiled-in. When I did this - > >the boot issue went away. > > > >On 5/23/05, Dave Jones <davej@xxxxxxxxxx> wrote: > > > > > >>On Mon, May 23, 2005 at 03:33:33PM -0400, Tim Taranov wrote: > >> > Hi - > >> > > >> > I was wondering why FC4 has ext3 as a kernel module rather than having > >> > it compiled into the kernel? > >> > > >> > The reason I'm asking is because I encountered an issue with booting > >> > FC4 from a raid volume with ext3 compiled as a module (the default > >> > configuration) - the kernel would panic after not being able to mount > >> > /. After I rebuilt the kernel to have the ext3 compiled-into the > >> > kernel, the booting problem went away (my /boot and / are ext3-based, > >> > running off a h/w raid controller based volume). > >> > > >> > I was using the default FC4 test 3 at the time - I haven't tried it > >> > with the latest kernel, but I suspect the results may be the same. > >> > >>If your initrd is made correctly, it shouldn't matter which way > >>its built. > >> > >> Dave > >> > >> > >> > >> > > > > > > > I am using FC4 test3 and I compiled kernel 2.6.11.10 and ext3 as a > modules and I have no problems. It boots clean everytime. >