Linux-Hardware Digest #999, Volume #13            Thu, 7 Dec 00 21:13:04 EST

Contents:
  Re: Adding SCSI as non-boot drive ("Steve Wolfe")
  Re: ECC RAM supported by Linux? ("Steve Wolfe")
  Re: Asus A7V (Robert Fox)
  Re: Hardware Monitor for Linux (Peter Riggs)
  SCSI tape problem (David McCracken)
  Test (nodavidmSPAM@&ims.com)
  hey intel warns! I'd better hold off on buying a PC! (jtnews)
  Re: Can't boot from floppy with Award Bios V4.51PG? _ More information (H Dziardziel)
  Re: ECC RAM supported by Linux? (Carlos)
  Installing ISA modem ("Patrick Nagurny")
  write SCSI DAT ... ([EMAIL PROTECTED])
  Re: Adaptec 2930CU - supported? (Ken)
  Is an AHA-152X Bootable? (Wayne Watson)
  Re: ECC RAM supported by Linux? ([EMAIL PROTECTED])
  Re: ECC RAM supported by Linux? (Dances With Crows)
  Re: ECC RAM supported by Linux? (jens)
  What's the best mobo for building a Linux system? ("Neal Lippman")
  Unix/Linux/BSD server question... ("C.W.R.")
  Re: ECC RAM supported by Linux? (David Van Cleef)

----------------------------------------------------------------------------

From: "Steve Wolfe" <[EMAIL PROTECTED]>
Subject: Re: Adding SCSI as non-boot drive
Date: Thu, 7 Dec 2000 14:12:06 -0700

> This may be a naive problem, so I apologize in advance.  I've installed
> an adaptec AHA-2940 into an existing linux - Redhat 6.2 - box with an
> existing EIDE drive which is the boot drive.  I would like to keep the
> EIDE as the boot.
>
> After installing the card and booting, I get as far as the SCSI
> detection utility and the system hangs for about 5 minutes, gives bios
> not found message and then boots from /dev/hda.
>
> If I start the bios config utility immediately upon booting, the EIDE
> drive is set as the boot harddrive.  This seems like it must be a low
> level hardware configuration problem like an incorrect dip-switch
> setting on the SCSI card itself, but I'm having trouble finding
> documentation related to this.
>
> Once I have booted, it seems the scsi card is recognized, but I cannot
> access the drives with fdisk and see not recognition of the SCSI drives
> in the boot log.
>
> Any help would be greatly appreciated.

  Go into the Adaptec card's BIOS (usually ctrl-A), and tell it not to scan
the bus, and not to try and install the boot BIOS.

steve




------------------------------

From: "Steve Wolfe" <[EMAIL PROTECTED]>
Subject: Re: ECC RAM supported by Linux?
Date: Thu, 7 Dec 2000 14:11:18 -0700

> Does ECC RAM have any support from Linux?
> How are detected errors reported?

  The ECC works at a hardware level, not in conjunction with the OS.  The
question is "does your motherboard's memory controller support it, and is
the support enabled in the BIOS?"

steve



------------------------------

From: Robert Fox <[EMAIL PROTECTED]>
Subject: Re: Asus A7V
Date: Thu, 07 Dec 2000 22:50:28 +0100

Check out this link for more!!!

http://www.mandrakeforum.com/article.php3?sid=20001201075945&mode=flat

RFox

Chris Cioffi wrote:
> 
> [EMAIL PROTECTED] (Drew) wrote in <[EMAIL PROTECTED]>:
> 
> >I'm a Linux newbie who just built a new box with an Asus A7V and Duron
> >CPU. My IBM drive is on the ATA100 connector on the mobo and the Linux
> >install can't see it (Mandrake 7.1).  It sees only the drive on my
> >primary IDE channel.  Are there any new Linux distros that will see the
> >ATA100 controller on the A7V right out of the box?  Failing that, are
> >there any step by step procedures for a newbie (who has yet to
> >experience compiling a kernal) that will enable me to get my system
> >loaded with Linux on the ATA100 channel of the motherboard?
> >
> >I have found several bits and pieces of info on several websites, but
> >still have alot of holes in my knowledge.  For instance, one site said
> >put the Linux drive on the primary IDE channel, upgrade it to kernal
> >2.4.x,, configure the kernal for Promise ATA100 controller, make some
> >lilo.conf changes and move the Linux drive back to the ATA100 channel.
> >But I don't know if the Promise driver is  in the 2.4 kernal, or if it
> >is in another RPM I must add.
> >
> >I'm anxious to learn but need help.  Has anyone succeeded with an A7V?
> >
> >
> 
> Hello there,
> 
> A friend of mine had a similar setup and he could not get it working until he
> replaced the drive.  Apparently there is some kind of conflict between the MB
> and the IBM drives.  The fix was to get a different HD.  YMMV.
> 
> We used a Western Digital and it works just fine.
> 
> Chris

------------------------------

From: Peter Riggs <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Hardware Monitor for Linux
Date: Thu, 07 Dec 2000 21:51:22 GMT

You want to visit the  Lm_sensors  home page at
http://www.netroedge.com/~lm78/ .
There are also various GUI frontends depending on you window
manager/desktop.

Regards,

Peter.


Kevin Brown wrote:

> Is there a hardware monitor for linux that can print out temperatures
> and fan speeds from the LM78 chip on my motherboard.  Something similar
> to Motherboard Monitor for Windows would be nice.
>
> Thanks,
> Kevin
> [EMAIL PROTECTED]


------------------------------

From: [EMAIL PROTECTED] (David McCracken)
Subject: SCSI tape problem
Date: 7 Dec 2000 22:03:10 GMT
Reply-To: [EMAIL PROTECTED]

Hello,
I have 2 questions/problems with an Exabyte 8505XL 8mm tape drive. It is
mounted internally, and is 1 of 7 devices (full load) on an Adaptec 2940U

All devices work fully in Win98 and Win2000. I leave the power turned
off the tape drive and 2 external devices except when I want to use them.

I recently added the 7th device, an IBM 9GB hard drive (internal) and
installed SuSE V7.0 on it, kernel 2.2.16.  Works very well, except...!!

I expected to be able to use tar with the tape drive to transport large
files to and from a Sun machine.

--##--
1.  The tape drive IS visible during the early stage of boot when the
    2940 is found.

In Linux, the tape drive is NOT present as seen using the SCSI info in
"kdm" (I think) or even using "cat /proc/scsi/scsi"

If I boot the machine with _either_ (or both) of the 2 external devices
turned on, (a scanner and a CD burner) the tape is recognized when I
boot into Linux and works OK with "mt /dev/st0 status" and tar.

Any ideas or suggestions on what the heck I can do to get normal SCSI
function without having to remember to turn on an unneeded device???


--##--
2.  After the tape is recognized by Linux, I can create and extract tape
    without a wimper, AND read & extract the same tape on the Solaris box.

However, I can't extract (same version of gnu tar on both boxes) from a
tape made on the Solaris box to my Linux machine.

mt /dev/st0 status shows "block size 1024" for a Solaris tape on the
Linux box, and tar tvf /dev/st0  bombs out with an I/O error.

I found that "mt" has additional functions in Linux (more than Solaris
2.7) which I learned about on the Exabyte web site.

I learned that if I manually set the Linux block size to "variable"
mode, (ie 0), then it can read/extract Solaris tapes.

mt /dev/st0 setblk 0

So, can I set a parameter to set this, (change default) during boot?

Thanks,
David 

------------------------------

From: nodavidmSPAM@&ims.com
Subject: Test
Date: 7 Dec 2000 22:00:06 GMT
Reply-To: nodavidmSPAM@&ims.com

555

-- 
========================================================================
David 

------------------------------

Date: Thu, 07 Dec 2000 17:12:51 -0500
From: jtnews <[EMAIL PROTECTED]>
Subject: hey intel warns! I'd better hold off on buying a PC!

Intel has warned! I'd better hold off on buying a PC.  More discounting
is
sure to follow!

------------------------------

From: [EMAIL PROTECTED] (H Dziardziel)
Subject: Re: Can't boot from floppy with Award Bios V4.51PG? _ More information
Date: Thu, 07 Dec 2000 22:15:49 GMT

On Thu, 7 Dec 2000 19:31:15 +0100, "Axel" <[EMAIL PROTECTED]> wrote:

>Hi Dziardziel,
>I thought that Linux didn't care about primary and extended partitions ?
>
>Regards Axel
>
Lilo must be on a bootable partition, Linux itself can be anywhere.
The Lilo docs/mini HowTo explains this.  Alternatively loadlin can
be used or a floppy (bootable of course).


------------------------------

From: [EMAIL PROTECTED] (Carlos)
Subject: Re: ECC RAM supported by Linux?
Date: 7 Dec 2000 16:50:06 -0600

In article <l62p09.6vn.ln@helix>, Steve Wolfe <[EMAIL PROTECTED]> wrote:
>> Does ECC RAM have any support from Linux?
>> How are detected errors reported?
>
>  The ECC works at a hardware level, not in conjunction with the OS.  The
>question is "does your motherboard's memory controller support it, and is
>the support enabled in the BIOS?"
>
>steve


While this is strictly speaking correct, it would be interesting to have
the memory controller report errors (corrected or detected but not
corrected) to the OS.  This could signal a faulty memory chip, or at the
least warn the administrator or user that the system is becoming
unreliable...   Why isn't there support to read these errors?

Carlos


------------------------------

From: "Patrick Nagurny" <[EMAIL PROTECTED]>
Subject: Installing ISA modem
Date: Thu, 7 Dec 2000 18:06:21 -0500

I'm trying to install a USRobotics/3com 56K Sportster (ISA) plug+play modem.
I tried running wvdialconf and it could not detect my modem.

So I ran pnpdump and it gave me an output file: it recognized my modem.
Now what do I do?

Do I have to do anything with isapnp and setserial?

Thanks,
Patrick



------------------------------

From: [EMAIL PROTECTED]
Subject: write SCSI DAT ...
Date: Thu, 07 Dec 2000 23:41:11 GMT

Salut,

I run redHat 6.2 with an inboard SCSI card (adaptec 160M). 2 disks are
connected on this card and linux boot on one of them (sda). I add
another SCSI card (IDE adaptec 160M) to connect external devices (zip
250 and Dat).

Linux reconize all (see below):

scsi : 2 hosts.
 (scsi0:0:0:0) Synchronous at 160.0 Mbyte/sec, offset 31.
   Vendor: SEAGATE  Model: ST39236LW        Rev: 0004
   Type:  Direct-Access                      ANSI SCSI revision: 03
Detected scsi
 disk sda at scsi0, channel 0, id 0, lun 0
 (scsi0:0:2:0) Synchronous at 160.0 Mbyte/sec, offset 31.
   Vendor: SEAGATE  Model: ST39236LW        Rev: 0004
   Type:  Direct-Access                      ANSI SCSI revision: 03
Detected scsi
 disk sdb at scsi0, channel 0, id 2, lun 0
 (scsi0:0:6:0) Synchronous at 20.0 Mbyte/sec, offset 16.
   Vendor: TOSHIBA  Model: CD-ROM XM-6401TA  Rev: 1001
   Type:  CD-ROM                            ANSI SCSI revision: 02
Detected scsi
 CD-ROM sr0 at scsi0, channel 0, id 6, lun 0
 (scsi1:0:5:0) Synchronous at 40.0 Mbyte/sec, offset 32.
   Vendor: HP        Model: C5683A            Rev: C908
   Type:  Sequential-Access                  ANSI SCSI revision: 02
Vendor:
 IOMEGA    Model: ZIP 250          Rev: H.41
   Type:  Direct-Access                      ANSI SCSI revision: 02
Detected scsi

My problem:

when I want to write on Tape (tar cf /dev/st0 usr) the command respond
"Inpout/output ERROR".
Look at msg:

st0 Error with sense data:[valid=0] Info fld=0x0,current st0 sense key
Aborted Command
espace Kernel: Additional sense indicate Data phase error

I don't understand what the problem come from.

Thanks to help me.

Guyaux


Sent via Deja.com http://www.deja.com/
Before you buy.

------------------------------

From: Ken <[EMAIL PROTECTED]>
Subject: Re: Adaptec 2930CU - supported?
Date: 7 Dec 2000 22:41:41 GMT

Joshua Baker-LePain <[EMAIL PROTECTED]> wrote:
> Ken <[EMAIL PROTECTED]> wrote:

>> Thanks.  I guess the problem is that I can't get it to load the
>> driver.  Nothing I try with the 2.1 CD works, and I tied a Red Hat 6.1
>> CD we have at work and it says somthing like "Loading aic7xxx" and
>> just sits there, I guess locked up.  Ugh.  I'm downloading the Debian

> You can always check that.  During the RedHat install, there are several
> virtual terminals open.  The default, where you spend your time, is
> Ctrl-Alt-F1.  But there are useful messages on Ctrl-Alt-F2, etc.

> You could also enter expert mode and tell it what modules to load.

Well, RH6.1 hangs on the boot before I can do anything.  I'd rather
have Debian though.

>> 2.2 ISO right now.  Hopefully it will make a difference.  Any
>> suggestions on getting it to load the driver?  Even a boot floppy
>> would be okay at this point.  I hate the feeling that someone doesn't
>> want me to load Linux.... :)

> Well, I've installed many RH versions (including 6.1) onto machines with
> HDs hanging off of Adaptec cards.  THe first thing I would check is
> your SCSI chain.  Do you have proper termination?  Are your cables good?
> How about cable length?  Did you use an appropriate goat to bless the chain?

> After that, I would get a recent distro (I'm running RH6.2 on all our
> machines here), and try that.  It should Just Work(TM).

My chain is fine AFAIK.  I have loaded DOS 6.22, Win95, and NT 4.0 and
all work fine.

Thanks again.


-Ken

------------------------------

From: Wayne Watson <[EMAIL PROTECTED]>
Subject: Is an AHA-152X Bootable?
Date: Thu, 07 Dec 2000 16:28:05 -0800
Reply-To: [EMAIL PROTECTED]

Yes, that's the question. Is that adaptek SCSI device bootable. I am unable to boot at 
start up from
the RH 6.2 CD-ROM regardless of how I set the device choices. I can boot up on another 
machine with
an IDE drive.

If it requires replacing this old SCSI adapter, what's a good replacement. I have two 
SCSI-2 devices
on my machine. How does on go about replacing it? I would imagine that I would install 
a new card,
then remove the old card and moving the necessary SCSI device cables to the new card.

--
                         "You guys line up alphabetically by height."
                          -- Bill Peterson, a Florida State football coach

                                            Wayne T. Watson
                       Web Page: http://www.sirius.com/~mtn_view (Updated 6/21/2000)
                       Imaginarium Science Museum:
                       http://www.sirius.com/~mtn_view/imaginarium.html



------------------------------

From: [EMAIL PROTECTED]
Subject: Re: ECC RAM supported by Linux?
Date: 07 Dec 2000 18:43:21 -0500

Carlos wrote:
>While this is strictly speaking correct, it would be interesting to have
>the memory controller report errors (corrected or detected but not
>corrected) to the OS.  This could signal a faulty memory chip, or at the
>least warn the administrator or user that the system is becoming
>unreliable...   Why isn't there support to read these errors?

This is an excellent question!

I have seen such reporting on Sun and SGI machines.
I got a parity error on both types of machine.

So, to those who say memory errors never happen... you better look out.

  Richard

------------------------------

From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: ECC RAM supported by Linux?
Date: 8 Dec 2000 01:11:07 GMT
Reply-To: [EMAIL PROTECTED]

On 7 Dec 2000 16:50:06 -0600, Carlos staggered into the Black Sun and
said:
>In article <l62p09.6vn.ln@helix>, Steve Wolfe <[EMAIL PROTECTED]>
>wrote:
>>> Does ECC RAM have any support from Linux?  How are detected errors
>>> reported?
>>
>>  The ECC works at a hardware level, not in conjunction with the OS.  The
>>question is "does your motherboard's memory controller support it, and is
>>the support enabled in the BIOS?"
>
>While this is strictly speaking correct, it would be interesting to have
>the memory controller report errors (corrected or detected but not
>corrected) to the OS.  This could signal a faulty memory chip, or at the
>least warn the administrator or user that the system is becoming
>unreliable...   Why isn't there support to read these errors?

Suppose the memory that holds the kernel segment that responds to memory
errors goes bad?  ("Kernel panic:  Error in error handler!")  When you
set up a system to report faults, it is best to make sure the faults
being detected can't bugger the reporting system--which means in this
case, putting the ECC error handler in ROM.  

I've heard of ECC controllers being able to store failure info in the
motherboard's NVRAM, which the OS could certainly read.  Don't know the
specifics though.

-- 
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin /  Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com     /   Workin' in a code mine, whoops!
=============================/    I hit a seg fault....

------------------------------

From: jens <[EMAIL PROTECTED]>
Subject: Re: ECC RAM supported by Linux?
Date: Fri, 08 Dec 2000 01:35:55 GMT

On 07 Dec 2000 18:43:21 -0500, [EMAIL PROTECTED] wrote:

>Carlos wrote:
>>While this is strictly speaking correct, it would be interesting to have
>>the memory controller report errors (corrected or detected but not
>>corrected) to the OS.  This could signal a faulty memory chip, or at the
>>least warn the administrator or user that the system is becoming
>>unreliable...   Why isn't there support to read these errors?
>
>This is an excellent question!
>
>I have seen such reporting on Sun and SGI machines.
>I got a parity error on both types of machine.
>
>So, to those who say memory errors never happen... you better look out.
>
>  Richard

Ok, I got a related question .... how come that the additional ram
chip (going from 8 to 9 in the average module) pretty well doubles the
price tag (this is a wild guess but I certainly remember having to
pick up my lower jaw off the floor when I was asking my dealer for ram
prices on ECC nodules)

Jens

------------------------------

From: "Neal Lippman" <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.hardware.homebuilt
Subject: What's the best mobo for building a Linux system?
Date: Fri, 08 Dec 2000 01:51:49 GMT

My question: What's the best mobo to use for building a new Linux system? I
am planning an Athlon 900 Mhz Socket A, so obviously my question relates to
mb's that support that chip. I plan to use Mandrake 7.2 as my distro (see
below) and an IBM Deskstar 75GXP HD.
I've been bouncing back and forth between the ABIT KT7-RAID, ASUS A7V, and
Microstar KT7-Pro2A. As you can see from my choices, I'm looking to run the
HD at ATA100. Although all three boards are also good to OC, I'm probably
not going to do that, at least at first.

I've heard than Mandrake 7.2 supports the Promise ATA-100 controller out of
the box, which makes the Asus board attractive, but I've also read about a
ton of bios related problems with that board (of course, I am sure just as
many people hae trouble with the others too).

Any thoughts...success stores...warnings?




------------------------------

From: "C.W.R." <[EMAIL PROTECTED]>
Crossposted-To: 
alt.os.linux,alt.os.linux.mandrake,alt.os.linux.suse,comp.unix.bsd.freebsd.misc,comp.unix.bsd.openbsd.misc,openbsd.misc,openbsd.tech
Subject: Unix/Linux/BSD server question...
Date: Fri, 08 Dec 2000 01:40:27 GMT

Hello --

I want to build or purchase (will probably end up building) a system for
around $350 or less for use as a dedicated test machine and server on my
home network. I wish to create a machine with a Pentium I 200 MHz or better.
I am also thinking of bending my budget and going for a Pentium II or maybe
even an AMD Duron. My main question is:

What type of processor and motherboard would you recommend that I use to get
the best bang for my buck? And are there any online retailers that still
sell parts this old. Remember, this is only going to be used so I can learn
about Unix/Linux/BSD, so I don't need anything better than the lowest end
Duron, and I don't even need that. Thanks for your help everybody!

CWR



------------------------------

From: [EMAIL PROTECTED] (David Van Cleef)
Subject: Re: ECC RAM supported by Linux?
Date: Fri, 08 Dec 2000 01:57:09 GMT

In article <90p46u$7bl$[EMAIL PROTECTED]>, Carlos wrote:
>In article <l62p09.6vn.ln@helix>, Steve Wolfe <[EMAIL PROTECTED]> wrote:
>>> Does ECC RAM have any support from Linux?
>>> How are detected errors reported?
>>
>>  The ECC works at a hardware level, not in conjunction with the OS.  The
>>question is "does your motherboard's memory controller support it, and is
>>the support enabled in the BIOS?"
>>
>>steve
>
>
>While this is strictly speaking correct, it would be interesting to have
>the memory controller report errors (corrected or detected but not
>corrected) to the OS.  This could signal a faulty memory chip, or at the
>least warn the administrator or user that the system is becoming
>unreliable...   Why isn't there support to read these errors?
>

Because the capability to do this is totally broken in just about every
x86 chipset that claims to support ECC.

A friend was working on this a while back but basically abandoned the
project when he realized how badly the support was broken.


------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.hardware) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Hardware Digest
******************************

Reply via email to