Hi all,
I am trying to install slackware 7.1. (why 7.1. because that was the last one that featured net.i boot disk image)
My system has a brandnew 20G harddrive and I can't use cd's to install, so I was figuring I boot from floppy then do an ftp install.
Right now, I am stuck at the lilo boot prompt with error code 0x04.
What next?

ok, I think I solved my own problem - maybe others can learn from this.

Starting from slack v8.1 (I think) net.i is obsolete because it tries to detect too many network adapters which can cause problems. An explanation can be found in:
ftp://ftp.slackware.com/pub/slackwar...144/README.TXT

It is now suggested to use the following method:

bootdisk: bare.i
rootdisks: install.1 & install.2 & network.dsk

An ftp install is as far as I know not supported in Slackware. The net rootdisk is used for installing using a NFS share.

actually there is:
http://emsi.it.pl/SuckSlack/ (nice name )
http://sourceforge.net/projects/slackftp

I tried slackftp already but it wouldn't work with my slack 9.1 disks, so I am about to try SuckSlack.

suckslack somewhat worked, at least it started to download the packages - problem is that my network connection dropped out in the middle somewhere.
I was hoping that enough packages had been installed, but the system won't even boot without the boot disk.
Then when I use the bootdisk and do mount=/dev/hda3
which is where my root is, system goes into a kernel panik saying that there is no init.
what now?

I'm somewhat frustrated to say the least....
I finally had suckslack working, but my network connection keeps dropping out and then I have to start all over again :-(
So in the meantime I got RH9 installed on a different server on the same network so I can install slack using NFS.... but I am getting the following error:

====
mount: RPC: Port mapper failure - RPC: Unable to receive
====
I then got out of setup, configured eth0 and I can ping my nfs host. How can I try the nfs mount from the console?

Also, I was able to install all *a* packages with suckslack, so the kernel is on the damn harddisk, how can I now boot from the hd instead of the boot floppy?

I just ordered the slack 9.1 cd's but if this nfs doesn't work, I'm screwed.

Did you set up any NFS shares on the PC hosting it? Anyway, things to check on the PC hosting the share.

1) Make sure that both portmap and nfsd are running.
If they aren't then start them.

2) Make sure you in fact have an NFS share. Peek at /etc/exports. You should see something similiar to this for each share:
The above shares "/tmp" with any PC at IP 192.168.0.2, allowing it both read and write access. This would need to be the IP address you enter for the PC you're installing slackware on when you setup the NFS installion. You can also use a FQDN instead of IP if you're running a DNS server for the IP/PC. I think you could also use a subnet, making the share available to every PC on your network. Check the NFS documentation for more info on the parameters. Of course anytime you make changes to this file you'll need to restart nfsd.

3) Make sure you are actually exporting the share, even if it was listed in /etc/exports. Run exportfs to see if it's there.

That should get you going. You can also use rpcinfo to make sure the client sees the share. But the slackware installion should pick it right up once it's available. Anyway, the command to do that would be something like so:

assuming 192.168.0.1 is the address of the PC running nfsd. Good luck.

Hi DaHammer,

I'm pretty sure that I setup everything correctly on my nfs host:
output from command 'rpcinfo -p' (on host):
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100024 1 udp 32768 status
100024 1 tcp 32768 status
391002 2 tcp 32769 sgi_fam
100011 1 udp 815 rquotad
100011 2 udp 815 rquotad
100011 1 tcp 818 rquotad
100011 2 tcp 818 rquotad
100003 2 udp 2049 nfs
100003 3 udp 2049 nfs
100021 1 udp 32780 nlockmgr
100021 3 udp 32780 nlockmgr
100021 4 udp 32780 nlockmgr
100005 1 udp 32781 mountd
100005 1 tcp 35926 mountd
100005 2 udp 32781 mountd
100005 2 tcp 35926 mountd
100005 3 udp 32781 mountd
100005 3 tcp 35926 mountd

portmap, rpc and nfsd are also up and running on the host.
my /ect/exports looks like this:
/home/andre 192.168.2.2(rw,sync)

and the command 'exportfs' gives:
/home/andre 192.168.2.2

and the command '/usr/sbin/rpcinfo -p 192.168.2.2' gives:
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper

on the client box, which doesn't have a lot of stuff on there because I could only get through installing all the 'a' packages I don't even have a rpcinfo command. The only command that I could find is rpc.portmap and it seems to be running.

any ideas?