Cluster Knoppix Trouble

[OverVolt]

Help me get through these errors so i can start testing how well open mosix balences F@H across a cluster. It's really some basic linux stuff and errors that i need to fix before i can start to boot the other machines over my network.

First of all, i get these two errors when i boot up (and a few more, but i dont have any SCSI devices so i dont think it matters) Any idea what these two errors below mean?

Error: MP-BIOS bug: 8254 timer not connected to IO-APIC
ide: late registration of driver

Secondly, when i do the cntrl+alt+F1 'thingy' where i can type in commands (im sucha -nix pro ) it spams "scandir: no such device or address" every few seconds. It does it so much that i can't install cluster knoppix to my HD because i can't see how im partitioning the drive since it gets spammed right out of the console screen. See what i mean!

Well, thanks for any help!

 

[n0cmonkey]

I'm not finding a lot on those errors (except the ide: late registration of driver one). What hardware? Is there an OS already installed?

 

[OverVolt]

Thanks for the help, how are you looking up info on those errors?

Anyway, i low level formatted the harddrive, but that doesnt matter since i can't install knoppix on it since i can't see what im doing. Until then, ive been running it off CD-ROM/ramdisk.

The hardware is:

AXP 1.71Ghz @ 162FSB @ 1.85v
1Gb Samsung DDR400 2-2-2-5
Asus NF2 'A7N8X'
WD 40Gb 2mb cache
300w antec PSU
INT. 10/100 NIC
PCI Linksys 10/100 NIC
PCI D-link 802.11b NIC

Im pretty sure its not a hardware problem. thanks for ur time!

 

[drag]

Not sure about this, but knoppix uses automount stuff.


It could be being retarded. Like it's trying to automount the harddrive for some stupid reason.

try "killall automount"


also you have ctl-alt-F2 thru F6 to try. Sometimes switching from one to another may help aviod some of the spam. Anyways your buttons will still be responding the same way as if nothing bad is happening so you should still be able to log in and type it out if you don't mind doing it blind. May take a couple tries though.

 

[n0cmonkey]

ew, nforce.

I was just searching on google.com/linux.

 

[OverVolt]

Thanks for the help. I tried the killall automount and it said no processes killed . anyway i switched to ctrl+alt+F2 and it wasn't getting spammed, but when i go to boot from the HD (after installing) it says the HD is stuck as read-only and to reinstall... arghhh( i tried but same message). BTW what file system should i use? ext2, ext3, reiser, or xfs.

Us poor nix noobs

EDit: sometimes it wont even boot! :evil:

 

[n0cmonkey]

I'd recommend ext3 or XFS.

Before you reboot after the install, check the /etc/fstab file. If you see an option (4th column) ro, change it to rw. In vi, highlight the o and type: r w Then save and exit (esc, :wq).

If you can, watch the dmesg when the machine is booting. It should tell you why the hard drive is ro. Any chance of having a better chipset lying around? I personally wouldn't trust the nf2 board...

 

[OverVolt]

n0cmonkey thanks for the help! The /etc/fstab file did the trick (the ro to rw thing in the 4th column)and it booted into KDE desktop just fine. would you happen to know anything about how i can configure my one wireless NIC to accept DHCP broadcast while the other NIC goes to a switch to boot the cluster with PXE? Looks like i have some work to do :evil:

 

[n0cmonkey]

Originally posted by: OverVolt
n0cmonkey thanks for the help! The /etc/fstab file did the trick (the ro to rw thing in the 4th column)and it booted into KDE desktop just fine. would you happen to know anything about how i can configure my one wireless NIC to accept DHCP broadcast while the other NIC goes to a switch to boot the cluster with PXE? Looks like i have some work to do :evil:

Nope, sorry. I don't like DHCP much, so I don't think I've used it on Linux. Ever.

It should be a simple search to find out though. Try man pump or look in /etc for a net/network/netstart/etc type directory.