Using today's tree (20041001) A new install on a Dell PowerEdge 2850 (has onboard PERC 4e/Di, uses the new megaraid_mbox drivers) Install went cleanly -- raid was autodetected, drivers were autoloaded, etc. After install, the system failed to boot - panic mounting / The mount fails because the raid modules (in my case, megaraid*) and scsi modules were omitted from the initrd; the only modules present in the initrd generated during install were ext3 and jbd Running mkinitrd from rescue mode created a functional initrd which contained ext3, jbd, megaraid*, scsi modules, etc. and worked
Can you provide the /root/install.log?
Created attachment 104641 [details] install log sure, attached
Hrmm. Is this reproducible? Are both the smp and up kernel initrds like this?
Both of them are like that, and it's done that on both 2850s I've tried so far Here's everything in the initrd (up, in this case): [root@localhost in3]# ls -lR .: total 36 drwxr-xr-x 2 root root 4096 Oct 1 14:38 bin drwxr-xr-x 2 root root 4096 Oct 1 14:38 dev drwxr-xr-x 3 root root 4096 Oct 1 14:38 etc -rwxr-xr-x 1 root root 709 Oct 1 11:56 init drwxr-xr-x 2 root root 4096 Oct 1 14:38 lib drwxr-xr-x 2 root root 4096 Oct 1 11:56 loopfs drwxr-xr-x 2 root root 4096 Oct 1 11:56 proc lrwxrwxrwx 1 root root 3 Oct 1 14:38 sbin -> bin drwxr-xr-x 2 root root 4096 Oct 1 11:56 sys drwxr-xr-x 2 root root 4096 Oct 1 11:56 sysroot ./bin: total 1560 lrwxrwxrwx 1 root root 10 Oct 1 14:38 hotplug -> /sbin/nash -rwxr-xr-x 1 root root 433032 Oct 1 11:56 insmod lrwxrwxrwx 1 root root 10 Oct 1 14:38 modprobe -> /sbin/nash -rwxr-xr-x 1 root root 482304 Oct 1 11:56 nash -rwxr-xr-x 1 root root 664080 Oct 1 11:56 udev lrwxrwxrwx 1 root root 4 Oct 1 14:38 udevstart -> udev ./dev: total 0 crw-r--r-- 1 root root 5, 1 Oct 1 11:56 console crw-r--r-- 1 root root 1, 3 Oct 1 11:56 null brw-r--r-- 1 root root 1, 1 Oct 1 11:56 ram crw-r--r-- 1 root root 4, 0 Oct 1 11:56 systty crw-r--r-- 1 root root 4, 1 Oct 1 11:56 tty1 crw-r--r-- 1 root root 4, 2 Oct 1 11:56 tty2 crw-r--r-- 1 root root 4, 3 Oct 1 11:56 tty3 crw-r--r-- 1 root root 4, 4 Oct 1 11:56 tty4 ./etc: total 4 drwxr-xr-x 2 root root 4096 Oct 1 14:38 udev ./etc/udev: total 4 -rw-r--r-- 1 root root 1403 Oct 1 11:56 udev.conf ./lib: total 348 -rwxr--r-- 1 root root 195024 Sep 1 18:27 ext3.ko -rwxr--r-- 1 root root 148800 Sep 1 18:27 jbd.ko ./loopfs: total 0 ./proc: total 0 ./sys: total 0 ./sysroot: total 0 [root@localhost in3]#
Can you start the installer and just get to tty2 and see what /tmp/modules.conf contains? And mkinitrd just does the right thing when you run it from rescue mode? Verrryyy weird.
/tmp/modprobe.conf has alias eth0 e1000 alias eth1 e1000 There is no modules.conf.... This and the rescue mode have both been http / ftp using the boot.iso image out of the tree today. In rescue mode, I've just done mkinitrd -f /boot/initrd-mumble.img mumble (didn't have to force inclusion of scsi or raid)
Aha, I think I see what it is. Should be fixed when I build anaconda-10.0.3.10.
Yep, today's tree (20041004) has this fixed thanks!