[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Installing Debian/Hurd



>> From: Ali Hamisheh-Bahar <Ali_Hamisheh-Bahar@mw.3com.com>
>> 
>> curt@gwis.com <curt@gwis.com>  wrote --
>> 
>> >  I *think* I found the message to which you refer, and no, that bears
>> >  no resemblance to what I got, or get.
>> 
>> I didn't think there would be: you've already installed on IDE. I just
>> hoped that you may have ran into it at some other point.

 Not at all.
 But then, I'm a bit late to the party.

>> >  I can run a trace/capture now, on what I will shortly find out if it's
>> >  the same, or similar.
>> > 
>> >  Would you like that ?
>> 
>> A trace of your scsi problem? I don't think the problems are similar,
>> but I'd love to see what someone else's install looks like.

 Ummm......
 I'm not quite certain of what you'ld like to see ?

>> >  Here, or elsewhere ?
>> 
>> I prefer here, so there'd be some archive of it kept.


>> PS. I know it's unorthodox, but let me just paste your earlier mail
>> here so I can reply to it:

 OK.

>>  | My problems deal with SCSI.
>>  | So....
>>  | I "acquired" another IDE drive. Making 5 work is not easy :-)
>> 
>> 5 disks?! Can your box support that?!!

 Not easily ;-)
 ( meaning no. I opened it up, and mixed and matched. )

>>  |
>>  | Got hurd running on an IDE drive, but it won't mount a scsi drive at
>>  | all.
>>  | GRUB finds, and uses the SCSI under BIOS control just fine, but the
>>  | kernel
>>  | crashes with a cannot find servers.boot file error every time if it's
>> 
>> When I used grub, I didn't go into command-line by pressing 'c'. I
>> went into edit 'e', instead. Then for each cmd, I used TAB to see if
>> cmd-line completion finds the file. 
>> This way, when you are editing the 'kernel' cmdline, you should be
>> able to use TAB to see if it can find files in different parts of the
>> filesystem, 
>> 
>> It's a dumb test, but that's all I can think of right now.

 'Sok. I'll do it, and see.

>>  | Got me past the native-install script, anyway.
>>  | Now getting "device unconfigured" errors on some disks, while others
>> 
>> I'd suggest figuring out the original kernel problem
>> (servers.boot). If the workaround is running into its own problems,
>> it's best to focus on the original problem.

 Agreed.
 
>>  | work.
>>  |
>>  | Will a Linux scsi driver module work with the hurd kernel somehow ?
>> 
>> Not being intimately knowledgable about hurd, I'd still venture to
>> say, 'no'. A driver is tightly bound to a kernel.

 I'm afraid of that.

>>  | ANY pointers appreciated.
>> 
>> 
>> For my own problem, I'll see if the source code of ext2fs.static will
>> give me some hints. This is not the most time-efficient way of
>> tackling the problem, but it may be educational.

 Well, education is priceless ;-)

 As a follow-up to one of my earlier.....
 VMWare emulates a BusLogic SCSI controller.
 That one works, and the hurd does boot scsi just fine.
 That means that I can't trace the AdvanSys errors, without
 the kernel doing its own logging someplace.

 Gonna hafta see about some "other" means of capturing that
 boot sequence....
 ( redirect to file on floppy, perhaps ? ) 

--
Cowboy

Auribus teneo lupum.
	[I hold a wolf by the ears.]
	[Boy, it *sounds* good.  But what does it *mean*?]




Reply to: