[SLE] Re: SCSI Drives From:[SLE] SOLVED: 'LI...' partial LILO load & hang
Note to any and all,
Argentium's linear setting should also be used by default if setting up SCSI
Drives as the SCSI Drive handle where things are, not the Processor and
LINUX and since the SCSI drives have their own processor and memory (they
always have had), you've just off loaded a process from your main system
rather than having a nonused process just looping there.
Also since most of the latest Large IDE type drives are following SCSI's
example of having their own onboard processor/memory and you set the drive
with LBA I suspect it may be
worth considering turning it on by default anyway, especially for those that
use DMA.
regards
scsijon
-----Original Message-----
From: Argentium G. Tiger
Ok everyone,
I was installing SuSE Linux 6.4 on a buddy's Pentium Pro 200 box last night to be a firewall/dial on demand server for him. (I seem to be doing a lot of these lately) :-)
We ran into the problem that occasionally shows up on the list:
LILO, hether it's installed in the master boot record (MBR), the boot sector of the boot partiion, the root partition, or the extended partition, starts loading but displays only "LI" on the screen and then stops.
This applied to 6.2, 6.3, and now applies to 6.4. It may also apply to earlier versions, but I can't say for sure.
Why does this problem happen? Well, I can't give an absolute answer, but I can tell you the configuration of the machine I've run into it on, and what we did that solved it.
The configuration:
Pentium Pro 200, 64 megs of memory, AIC7xxx SCSI adapter on-board the motherboard, and a completely SCSI system (4 gig SCSI wide hard drive, Plextor 20X SCSI cd-rom, etc...) There are _no_ IDE devices in this system, and that seems to be key to this problem showing up.
The install:
Goes smoothly, no problems. No hint that this problem will show up once the machine is rebooted.
The problem:
Shows up on reboot. Without a boot disk, you get the "LI" <stop> problem.
The reason:
According to the SuSE Linux 6.4 manual, page 121, section 4.8.1 "Diagnosis of Errors: LILO Start Messages" 'LI' The second stage has been invoked but could not be started. This can either be caused by a geometry mismatch or by moving /boot/boot.b without reinstalling LILO.
...
Well, we hadn't moved /boot/boot.b, so it had to be a geometry mismatch. I had noticed that this hard drive had 4,700 (approx) tracks... That is a _lot_ of tracks for a 4 gigabyte drive.
The solution (in our case):
Add the "linear" option to the global section of the /etc/lilo.conf.
Under yast, check the "linear" box.
Linear:
Page 112 of the manual refers to linear:
"This option causes all references to sectors to be written as logical instead of physical addresses. This option might be useful if LILO does not recognize the geometry of the hard disk correctly. Still, it does not make the 1024 cylinders limit obsolete. In practice, this is scarcely ever needed."
I was so happy to be among the few, the proud, the scarce. :-)
Anyway, thought people might want to know a way to tackle this annoying problem. I don't guarantee it will work, but hey, it worked for us.
Argentium
-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/
participants (1)
-
scsijon＠net2000.com.au