Very Computer

Hi,

I've been trying to get my Iomega Zip drive working with RH 6.1, without
success.  I was able to get it to work with RH 5.2 once, but I can't do it
the same way with RH 6.1.

My zip drive is attached to a "Zip Zoom" card, which is refereed to in the
ZIP-Drive mini HOTWO.  In Windows 98 (mine's a dual-boot system), I see
that the IRQ (or is it DMA?) for the zip card is 11, and base address is
0x140.  The redhat 6.1 installation did not recognize this as a SCSI
controller so it did not configure my kernel for SCSI support.

I've tried a lot of things, following the various bits of advice in the
ZIP-Drive mini howto, and others.  I have recompiled the kernel with SCSI
support built in, and have the line:
        append="aha152x,0x140,11,7,1"
in my lilo.conf file (and ran liloconf afterwards), but this didn't help.
I get the following message when I try to mount the zip drive:
[root /]# mount /dev/sda4 /mnt/zip/
mount: the kernel does not recognize /dev/sda4 as a block device
       (maybe `insmod driver'?)

I found that the HOWTO docs on this and SCSI in general seem out of date --
some of the advice doesn't really apply to the newer kernels.  This makes
life much more difficult, of course.

Does anyone know I might be able to get my Zip drive to work?

Thanks,
  Topher Cawlfield

Greetings:

I'll give ya alot of credit :)  Most of the postings I've delt with this
weekend (other groups) have not been as thoughtfully researched as yours :)
  First, let's check to see if you have any support for your SCSI adapter at
all by typing "cat /proc/interrupts" (*should* show "aha152x=11"), or by
typing in "/cat /proc/modules" (*should* show "aha152x:  ##").
  Next, check to see if the problem isn't with the ZIP cartridges
themselves!  Type in "fdisk /dev/sda", but do NOT make any changes!!! This
will tell ya beyond a shadow of a doubt, whether or not you can "see" your
cartridge at all.  It will also show you what partition(s), if any, are
accessable to Linux :)

Let me(us) know what happens,
Daniel W. Erskine

Shouldn't that be: append="aha152x=0x140,11,7,1"

Quote:>in my lilo.conf file (and ran liloconf afterwards), but this didn't help.
>I get the following message when I try to mount the zip drive:
>[root /]# mount /dev/sda4 /mnt/zip/
>mount: the kernel does not recognize /dev/sda4 as a block device
>       (maybe `insmod driver'?)

>I found that the HOWTO docs on this and SCSI in general seem out of date --
>some of the advice doesn't really apply to the newer kernels.  This makes
>life much more difficult, of course.

>Does anyone know I might be able to get my Zip drive to work?

>Thanks,
>  Topher Cawlfield

--
clear skies,            |http://www.erols.com/jringer3/astro1.htm
Joe                     |
                        |Men are born ignorant, not stupid; they are
                        |made stupid by education. --Bertrand Russell

Hi,

Sorry to respond to this so late.  I only had the chance to try out these
suggestions this morning.  Busy schedule these days!

Well, I finally got it to work!  I tried looking at /proc/modules (a neat
trick, thanks!), and didn't see it there.  The fdisk command didn't work,
either, but that's also very nice to know about.  Joe's reply turned out to
nail it -- I had a typo in my lilo.conf file.  According to others, an
entry in lilo.conf shouldn't help or be necessary.  But I think that in my
case it was, because when I re-compiled my kernel earlier, I not only
included SCSI support (and SCSI disk support, and SCSI generic support
which someone else told me I might need), I added built-in support for the
aha152x module ('y' instead of 'm' in the xconfig program).  This was
probably not the smartest thing to do.  But by the time I was preparing to
re-compile the kernel (a scary thing for me, being the first time) I was so
desperate that I went for a little overkill.

I'm guessing that the following is true:
  What I should have done is to re-compile the kernel with SCSI support,
and SCSI disk support, leaving the other SCSI options alone.
  With the aha152x driver as a module, I could put the necessary
information, the base addr and IRQ in /etc/conf.modules.  This is perhaps a
nicer solution than a special lilo configuration.
  With built-in support for aha152x, I have to provide the hardware
information another way, and AFAIK, lilo is the only way.

After fixing this typo in my lilo.conf file, when I rebooted I saw a
message flash by indicating that it found the zip drive, and went into some
kind of "detected new hardware" screen, where I pressed a button to
configure it.  I have no idea what that did, but I was amused -- nice
"autodetect" feature!  After logging in, I was able to mount zip disks and
read them.

BTW, I've been looking for a way to read all those messages that fly across
my screen during bootup.  I thought they were saved to a file, but I can't
seem to find them.  I did see a file, /var/tmp/boot.log which looked very
promising.  But it didn't have the lines I was looking for.  Do you know,
off-hand, how to save these messages?

Thank you for your help!

 - Topher

<snipping out my original post>

Thanks for catching my typo!  That was it.  

 - Topher

They're in /var/log/dmesg, most likely.

--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University

The command "dmesg" will display the most recent 8K (or so) worth of
system messages.  If you run it soon after bootup, you'll see the entire
sequence.

The boot script dumps the output of dmesg to /var/log/dmesg during the
boot sequence.  On my system, the last line dumped is the one where it
turns on the swap file ("Adding Swap: ....")

The entire log file is spooled into /var/log/messages*.  Note that this
has a lot more than just boot sequence text.  It's your entire system
log.

Note that the messages in /var/log/messages* doesn't indicate where you
reboot.  All messages are concatenated together.  To spot where reboots
occur, look for lines like:

        syslog: syslogd startup succeeded

This is whre the logger starts.  Normally, this happens at boot time,
although it could happen at other times.  The syslogd daemon starts up
before all boot messages are generated.  A few lines after this, you
should find lines like:

        Detected 398272042 Hz processor.

Which are the first lines generated to the console during the boot
sequence.  These are also the first lines of dmesg's log.

-- David

When you recompile your kernel, if you select the aha152x SCSI
driver statically (not as a module) then you don't need to use
the append="aha152x,0x140,11,7,1" in the lilo.conf, since the
driver in the kernel should do all the detection work by
itself.

--
  This answer is courtesy of QuestionExchange.com
  http://www.questionexchange.com/servlet3/qx.usenetGuest.showUsenetGue...