Second stage 'emerge system' problem

[Roobarb]

I'm trying to install Gentoo on a SparcStation 20 (2xRoss Hypersparc chips and 256Mb Ram).
I managed to get the stage 1 tarball installed (from the iso cd) and did the scripts/bootstrap.sh to get to stage 2.

When I run 'emerge system' to bring it up to stage 3, it gets to what appears to be a test script for perl and starts hanging on some of the tests; specifically networking tests (lib/io_multihomed, lib/io_sock and a few others). If I wait long enough, one of the errors suggests that is can't connect to localhost however I have a 127.0.0.1 localhost entry in /etc/fstab (in the chrooted environment).

Are these problems to do with teething problems with the sparc installers and would they go away if I went straight for the stage 3 tarball, or are they harmless errors that will pass if I leave the box overnight?

Cheers,

Roobarb.

P.S. I had to do 'mount proc /mnt/proc -t proc' in order to get the host /proc filesystem mounted without errors. 'mount -o bind /proc /mnt/proc' caused an error about /mnt/proc not being a block filesystem.

[PimpNasty]

/etc/fstab is used for filesystem info. /etc/hosts is the file where "127.0.0.1 localhost" should be located. I am not sure what the problem would be with the hangs though... good luck.

The mount point for the proc filesystem is /mnt/gentoo/proc, not /mnt/proc. So, to mount the proc filesystem the command is "mount -o bind /proc /mnt/gentoo/proc"

[Milez]

...to also include -n if you are using the cdrom to install, since you are working off of a read-only filesystem. My command to mount proc (very important to mount proc too, bootstrap.sh at least needs it):

mount -n -o bind -t proc /proc /mnt/gentoo/proc

N.B. Once again, you DON'T want to use -n unless you are installing off CD or some other read-only file system.
_________________
-=Miles Stoudenmire=-

Adopt a post initiative