Well, I still am unable to boot even with the new GRUB. Here's my situation:
1. I started with a weird setup: (primary ide, master) CDROM (primary ide, slave) FAT HD (windows) (secondary ide, master) EXT2 HD (linux + Hurd partition). (secondary ide, slave) NONE The Linux partition was set up as: /dev/hdc1 Linux root /dev/hdc2 Linux swap /dev/hdc3 Linux usr directory /dev/hdc4 Extended partition </dev/hdc5> Linux Projects directory </dev/hdc6> Hurd directory. This setup always halted during boot with: hd1: hd1s1 hd2: [PTBL] [526/255/63] hd2s1 hd2s2 hd2s3 hd2s4 <hd2s5 hd2s6> com0: at atbus 0, port=3f8, spd=6, pic=4, (DOS COM1) com1: at atbus 1, port=2f8, spd=6, pic=3, (DOS COM2) hd24: bad access: block=28, count=2, blockend=30, nr_sects end_request: I/O error, dev 16:03, sector 28. 2. I figured that gnumach might be unable to understand the logical drives, so I repartitioned with the Hurd before the logical drive: /dev/hdc1 Linux root /dev/hdc2 Hurd directory. /dev/hdc3 Extended partition </dev/hdc5> Linux swap </dev/hdc6> Linux Projects directory /dev/hdc4 Linux usr directory This setups always halted during boot with: hd1: hd1s1 hd2: [PTBL] [526/255/63] hd2s1 hd2s2 hd2s3 <hd2s5 hd2s6> hd2s4 com0: at atbus 0, port=3f8, spd=6, pic=4, (DOS COM1) com1: at atbus 1, port=2f8, spd=6, pic=3, (DOS COM2) hd23: bad access: block=28, count=2, blockend=30, nr_sects end_request: I/O error, dev 16:03, sector 28. 3. Then I thought it might be because of the funny "Primary master is a CDROM" situation, so I reconfigured like so: (primary ide, master) EXT2 HD (linux + Hurd partition). (primary ide, slave) NONE (secondary ide, master) CDROM (secondary ide, slave) FAT HD (windows) The Linux partition was set up as: /dev/hda1 Linux root /dev/hda2 Linux swap /dev/hda3 Linux usr directory /dev/hda4 Extended partition </dev/hda5> Linux Projects directory </dev/hda6> Hurd directory. This setups always halted during boot with: hd1: [PTBL] [526/255/63] hd1s1 hd1s2 hd1s3 <hd1s5 hd1s6> hd1s4 hd2: hd2s1 com0: at atbus 0, port=3f8, spd=6, pic=4, (DOS COM1) com1: at atbus 1, port=2f8, spd=6, pic=3, (DOS COM2) hd13: bad access: block=28, count=2, blockend=30, nr_sects end_request: I/O error, dev 16:03, sector 28. So I'm stuck. I've tried the latest GNUMach from the Debian archive, as well as the latest GRUB from CVS. I've also placed the latest hurd binaries onto the Hurd partition to no avail. It seems pretty clear that the problem is related to the logical partitioning, but I don't see what else I can change, short of limiting myself to primary partitions. Has anyone run across this issue and been able to resolve it? -Brent
