Linux-Hardware Digest #683, Volume #13 Fri, 6 Oct 00 19:13:04 EDT
Contents:
Re: IDE Tape STT220000a and RH 6.2 (John-Paul Stewart)
Modem works on Mandrake 7.0 but not 7.1 ??? (Aaron Hughes)
Installed new processors: now problems... (dsmith)
midi via serial port interface (Portman PC/S) ([EMAIL PROTECTED])
Re: Looking for a killer Dual CPU board for a Linux system.. (John Burton)
Arco-IDE-RAID not in DMA mode :-( (Dick Visser)
test ("pl")
Sound card help ("bobbybean")
Adaptec 1540B problem (Jan)
Re: ATA/100 drives with Linux (Vidar Skjelanger)
CMOS clock? (Sascha Ueberall)
Re: Best Linux server components questions ([EMAIL PROTECTED])
Re: IBM SCSI does not autostart ("Brett I. Holcomb")
Re: Installed new processors: now problems... (Andrey Vlasov)
Re: IBM SCSI does not autostart ("Folkert Rienstra")
Archos pcmcia Zip100 drive? (Ulrich Lauther)
Re: Installed new processors: now problems... (Andrey Vlasov)
Re: easy-to-setup CDR recommendation (Dances With Crows)
----------------------------------------------------------------------------
From: John-Paul Stewart <[EMAIL PROTECTED]>
Subject: Re: IDE Tape STT220000a and RH 6.2
Date: Fri, 06 Oct 2000 19:27:35 GMT
Bob Alberti wrote:
>
> I have a Seagate IDE Tape installed as hdd in RedHat 6.2 rpm + latest patches
> system.
>
> dMesg shows that the tape is being seen at boot time:
> hdd: Seagate STT20000A, ATAPI TAPE drive
>
> But I cannot use it for anything:
> Sep 30 11:37:59 fw kernel: ide-tape: hdd: Unsupported command in request queue
> Sep 30 11:37:59 fw kernel: end_request: I/O error, dev 16:40 (hdd), sector 21498
> Sep 30 11:38:08 fw kernel: ide-tape: ht0: I/O error, pc = 34, key = 3, asc = 30,
>ascq = 0
>
> and of course:
>
> [alberti@fw ~]# mt -f /dev/nht0 status
> /dev/nht0: Input/output error
>
> Any hints? Any help? A bone to a poor dog?
>
> Thanks in advance
>
> Bob Alberti
I have the same drive running under Corel Linux. Using the
kernel from the CD i get the same error messages on boot
up. Compiling the kernel from source (either the included
source or "clean" from kernel.org) clears up the problem.
See the Kernel-HOWTO if you're new to custom kernels.
HTH,
J-P Stewart
------------------------------
From: Aaron Hughes <[EMAIL PROTECTED]>
Subject: Modem works on Mandrake 7.0 but not 7.1 ???
Date: Fri, 06 Oct 2000 19:30:04 -0000
I have an actiontec pci call waiting modem internal
see www.actiontec.com
Unde mandrake 7.0 it worked fine as long as I typed
setserial /dev/ttyS3 port 0xcc00 spd_vhi skip_test auto_irq autoconfig
setserial /dev/ttyS3 uart 16550A
I then used KPPP and got on the web etc etc
The same trick under mandrake 7.1 doesnt work.
Any idea why?
The hardware detector also does not find the modem? HardDrake or DrakConf
whatever it is called?
Thanks for your help
and please respond here
[EMAIL PROTECTED]
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: dsmith <[EMAIL PROTECTED]>
Subject: Installed new processors: now problems...
Date: Fri, 06 Oct 2000 15:16:54 -0400
I had a server that was running 2 PIII 500's on a 100MHz bus. I
installed new PIII733's that run on a 133MHz bus. If I boot linux into
an SMP kernel, I will sometimes get errors. When the SCSI module loads,
it hangs and reports that it is waiting for a TID on CPU 0 or something
like that. It runs, and pauses, and runs, and pauses... If I look at
/proc/cpuinfo, it shows both processors, but CPU1 is 0MHZ and 0
BogoMIPS. If it happens, I can usually reboot and it will be FINE. I
have a SIIG Ultra66 controller that worked fine before. Now if I put it
in, the problem arises all the time, every time. If I take it out, it
goes back to normal. If I tell the motherboard to force the processors
back to the 100MHz bus, everthing seems to be fine.
Is it more likely that hardware is causing problems? The new processors
are Coppermines, while the old ones are Katmai's. Is there any reason
the kernel would need to be recompiled for the Coppermines and/or the
new bus speed? I wouldn't think so, but I can't imagine what else could
be the problem. I can't see why the SIIG card would have a problem with
the bus speed. Keep in mind that if I boot a non-SMP kernel, everything
is FINE... Kernel version (both SMP and non) are 2.2.15.
Thanks!
-Dan
------------------------------
From: [EMAIL PROTECTED]
Subject: midi via serial port interface (Portman PC/S)
Date: Fri, 06 Oct 2000 19:25:33 GMT
Has anyone had any luck sending midi data out the serial port to a sound
module? I'm running SuSE 6.4 with a Portman PC/S serial-midi interface
box. So far I have snd-card-serial (as well as snd-card-emu10k1 for my
SBLive) loaded with the following options line in my /etc/modules.conf:
options snd-card-serial snd_io=0x3f8 snd_irq=4
I have the Portman PC/S connected to /dev/ttyS0 and I've got it to send
_something_ to my sound module (a Roland JV-1010) because the LED's are
flashing on the module, but no sound is produced and when I stop sending
the data I get a Roland error code which, according to the manual, means
bad midi input.
I've also tried the option snd_div=3 up thru snd_div=12, and I've tried
(with even less success) the uart16550_midi module.
If anyone has any ideas about sending midi data through a serial port
(with or without a Portman), please let me know.
Thanks for any suggestions! __sonic
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: John Burton <[EMAIL PROTECTED]>
Subject: Re: Looking for a killer Dual CPU board for a Linux system..
Date: Fri, 06 Oct 2000 16:20:19 -0400
"Richard F. Jr." wrote:
> Hi all,
> I will soon be building a new box and would like to make it SMP.
> But am having trouble picking a board, any help ? I have used the
> Supermicro P6DBE, 440BX board before, but I get conflicting reports on
> it's max speed, some say it's a max at 450, others 700 or even 800MHz...
> I love this board and use it in a Dual PII 400MHz NT Workstation, any one
> have a board like this for P2 / P3, with IDE ATA66 / 100 ???
>
> Socket 1 or 370 is fine... I just need a reliable Linux board !
> And I hate built in stuff...! (Sound / Video / NIC)
>
You like the SuperMicro boards? Why not try the SuperMicro 370DLE/ 370DL3.
They have 133Mhz FSB, dual PCI buses (one is a 64bit PCI bus). The DL3 has
an on-board Ultra160 SCSI controller, not sure what the IDE controller speed
is. It uses the ServerWorks LE chipset which several companies (e.g. IBM &
Compaq) are using in their mid range Intel server boxes (for high end, they
use the ServerWorks HE chipset..)
John
--
John Burton, Ph.D.
Senior Associate GATS, Inc.
[EMAIL PROTECTED] 11864 Canon Blvd - Suite 101
[EMAIL PROTECTED] (personal) Newport News, VA 23606
(757) 873-5920 (voice) (757) 873-5924 (fax)
------------------------------
From: [EMAIL PROTECTED] (Dick Visser)
Subject: Arco-IDE-RAID not in DMA mode :-(
Date: 6 Oct 2000 20:24:32 GMT
Hello
I just bought an Asus CUBX, Celeron 600, two Maxtor 20,4 GB (52049H4) drives,
and an IDE-RAID-adapter by Arco (www.arcoide), model DupliDisk ISA.
All seems to work fine. After initialisation of the drives by software on a
floppy the drive are seen as one single IDE device by the motherboard, and I
installed Mandrake 7.1 on it. No problem.
On all the systems I build before, I put used "hdparm -d1" to get my drives on
speed. This generally increases hard drive throughput at least two times
(sometimes even 4-5 times ;-).
But whit this RAID-controller there's no way I can get it to use DMA. It keeps
popping it up messages to me like:
Using DMA: operation not permitted.
I've tried to change to setting in BIOS but it's no use...
Is there anybody out there that also has this RAID-adapter in a linux box and
that has managed to get it working in DMA mode?
I won't be needing the UATA100 stuff but DMA would be nice cause it's doing a
poor 6-7 Mb/sec right now.
Thanks
--
Dick Visser
Amsterdam
The Netherlands
www.tienhuis.nl
------------------------------
From: "pl" <[EMAIL PROTECTED]>
Subject: test
Date: Fri, 06 Oct 2000 20:27:44 GMT
test
------------------------------
Reply-To: "bobbybean" <[EMAIL PROTECTED]>
From: "bobbybean" <[EMAIL PROTECTED]>
Subject: Sound card help
Date: Fri, 06 Oct 2000 20:40:03 GMT
I'm running Caldera 2.4 with a Intel 810 Sound & Vid, download the rpm from
Intel
but still can't get any sound, vid is fine, but only 24 bit.
Thanks
Bobby
------------------------------
Date: Fri, 06 Oct 2000 22:41:23 +0100
From: Jan <[EMAIL PROTECTED]>
Subject: Adaptec 1540B problem
I have an old scsi card that I know once worked
- and maybe still does. When I boot, I see the
banner from the card's BIOS. After a while it
writes something that is immediately overwritten
by the exceptionally stupid full-screen message
from the system BIOS, so I can't see what it
says.
When Linux boots, it says 'scsi: 0 hosts found'.
I have one disk connected at ID 0, the card has
ID 7. If I press ctl-A when the Adaptec BIOS is
on the screen, nothing happens, so I can't get
in there and look.
Are there any command line tools that I can use
to look at the card? And anything alse that I
should try?
/jan
------------------------------
From: Vidar Skjelanger <[EMAIL PROTECTED]>
Subject: Re: ATA/100 drives with Linux
Date: 06 Oct 2000 22:53:45 +0200
"Adam Short" <[EMAIL PROTECTED]> writes:
> The only IDE controllers I know of at the moment that support ATA/100 are
> RAID controllers. I haven't seen any normal ones yet, there may be a reason
> for that but if so I'm not enough of a hardware buff to know it.
The Asus A7V motherboard has two non-RAID ATA/100 controllers.
--
mvh,
Vidar Skjelanger
------------------------------
From: Sascha Ueberall <[EMAIL PROTECTED]>
Subject: CMOS clock?
Date: Fri, 06 Oct 2000 12:12:45 +0100
Reply-To: [EMAIL PROTECTED]
Hi everybody,
I have a problem with installing linux:
The first time it starts up, it hangs when the system tries to set up
the CMOS clock.
Does anybody know how to handle this? Thanks for any hints in advance,
all the best,
Sascha
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Best Linux server components questions
Date: Fri, 06 Oct 2000 20:55:52 GMT
> If you are building a web-server you
> could buy a preconfigured machine from cobalt or netwinder. There is
> also software from NetMax which is said to easily install on RedHat.
Is there software that provides functionality similar to the Cobalt
ManageRaq? For example, is there anything that provides a way to manage
a farm of servers from a single console?
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Brett I. Holcomb" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: IBM SCSI does not autostart
Date: Fri, 6 Oct 2000 16:23:15 -0500
IIRC, the those drives default to LVD mode so unless you have a controller
that handles that you have to put them in SE mode.
--
Brett I. Holcomb
[EMAIL PROTECTED]
Microsoft MVP
AKA Grunt<><
Remove R777 to reply
"David Wilson" <[EMAIL PROTECTED]> wrote in message
news:39dd66e2$[EMAIL PROTECTED]...
> David_C <[EMAIL PROTECTED]> writes:
> >As for the divvsens pin - no I don't implicitly trust it. But this is a
> >standardized feature. Unless proven otherwise, I presume that a drive
> >will comply to the standard.
>
> Well, when we plugged IBM-DDRS-34560D-DC1B drives into Sun SS4 computers
> we had to strap them to SE before the drives would be recognised. I don't
> know if it was the drive or the computer at fault - I was just happy that
> the drive had a SE jumper (admittedly it was a problem finding 10 tiny
jumpers
> at short notice - normal sized ones we had plenty of).
> --
> David Wilson School of IT & CS, Uni of Wollongong, Australia
------------------------------
From: Andrey Vlasov <[EMAIL PROTECTED]>
Subject: Re: Installed new processors: now problems...
Date: Fri, 06 Oct 2000 14:52:17 -0700
==============405BF44DE3A867021A29555A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hi there,
Do you sure that this card can works stable on 133MHz bus? Check
documentation for this card as it can be a problem. I guess that all
new cards should works with 133MHZ FSB but old will have a problems.
Andrey
dsmith wrote:
> I had a server that was running 2 PIII 500's on a 100MHz bus. I
> installed new PIII733's that run on a 133MHz bus. If I boot linux into
> an SMP kernel, I will sometimes get errors. When the SCSI module loads,
> it hangs and reports that it is waiting for a TID on CPU 0 or something
> like that. It runs, and pauses, and runs, and pauses... If I look at
> /proc/cpuinfo, it shows both processors, but CPU1 is 0MHZ and 0
> BogoMIPS. If it happens, I can usually reboot and it will be FINE. I
> have a SIIG Ultra66 controller that worked fine before. Now if I put it
> in, the problem arises all the time, every time. If I take it out, it
> goes back to normal. If I tell the motherboard to force the processors
> back to the 100MHz bus, everthing seems to be fine.
>
> Is it more likely that hardware is causing problems? The new processors
> are Coppermines, while the old ones are Katmai's. Is there any reason
> the kernel would need to be recompiled for the Coppermines and/or the
> new bus speed? I wouldn't think so, but I can't imagine what else could
> be the problem. I can't see why the SIIG card would have a problem with
> the bus speed. Keep in mind that if I boot a non-SMP kernel, everything
> is FINE... Kernel version (both SMP and non) are 2.2.15.
>
> Thanks!
>
> -Dan
==============405BF44DE3A867021A29555A
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi there,
<p>Do you sure that this card can works stable on 133MHz bus? Check
<br>documentation for this card as it can be a problem. I guess that all
<br>new cards should works with 133MHZ FSB but old will have a problems.
<p>Andrey
<p>dsmith wrote:
<blockquote TYPE=CITE>I had a server that was running 2 PIII 500's on a
100MHz bus. I
<br>installed new PIII733's that run on a 133MHz bus. If I boot linux
into
<br>an SMP kernel, I will sometimes get errors. When the SCSI module
loads,
<br>it hangs and reports that it is waiting for a TID on CPU 0 or something
<br>like that. It runs, and pauses, and runs, and pauses...
If I look at
<br>/proc/cpuinfo, it shows both processors, but CPU1 is 0MHZ and 0
<br>BogoMIPS. If it happens, I can usually reboot and it will be
FINE. I
<br>have a SIIG Ultra66 controller that worked fine before. Now if
I put it
<br>in, the problem arises all the time, every time. If I take it
out, it
<br>goes back to normal. If I tell the motherboard to force the processors
<br>back to the 100MHz bus, everthing seems to be fine.
<p>Is it more likely that hardware is causing problems? The new processors
<br>are Coppermines, while the old ones are Katmai's. Is there any
reason
<br>the kernel would need to be recompiled for the Coppermines and/or the
<br>new bus speed? I wouldn't think so, but I can't imagine what
else could
<br>be the problem. I can't see why the SIIG card would have a problem
with
<br>the bus speed. Keep in mind that if I boot a non-SMP kernel,
everything
<br>is FINE... Kernel version (both SMP and non) are 2.2.15.
<p>Thanks!
<p>-Dan</blockquote>
<pre></pre>
</html>
==============405BF44DE3A867021A29555A==
------------------------------
From: "Folkert Rienstra" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: IBM SCSI does not autostart
Date: Sat, 7 Oct 2000 00:08:12 +0200
"David_C" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...
| "Folkert Rienstra" <[EMAIL PROTECTED]> writes:
| >>
| >> Don't trust auto-termination - it sometimes messes up.
| >>
| >> Forcing the hard drive to SE mode shouldn't change anything - it
| >> should auto detect an SE bus and run in that mode on its own.
| >
| > Hmm, you trust the controller to have a grounded diffsens pin but you
| > distrust a drive to have a grounded termination sense pin?
|
| I have personally experienced configurations where auto-termination has
| guessed wrong - making the entire bus useless.
Yes, but there are different ways to do it. I meant the situation where
the controller senses which connectors are in use. Not the ones that measure current.
|
| As for "termination sense pin", where is this defined?
It is not. I think the general agreement is to use pin 22 on the internal
50 pin and its equivalent (50 I believe) on the 68 pin connector.
They are defined as ground and that is what the procedure expects.
That is what AdvanSys and Adaptec use.
As long as those pins are grounded and the segments used as intended that procedure
works without flaw.
|
| There is no standard way for auto-detecting termination. Different
| controllers use different algorithms, and none of them are standardized
| or foolproof.
|
| Please read the SCSI FAQ from your own sig file:
| http://www.scsifaq.org/scsifaq.html#Host_Adapter1
|
| As for the divvsens pin - no I don't implicitly trust it. But this is a
| standardized feature.
But also not defined on a single ended controller.
Pin 16 (mini DB68 based), 11 (mini DB50 based) or 21 (50 internal
header based) is assumed ground there.
| Unless proven otherwise, I presume that a drive will comply to the
standard.
That won't help any if the diffsens line on the SE controller is NC.
|
| This is different from auto-termination, where there is no such
| standard, and I have personally witnessed its failure on busses where
| nothing is broken.
Something has to be, otherwise it should have worked.
Of course it could be the procedure itself that is broken to begin with.
Even the setup firmware can be broken. It was on some AdvanSys SCSI
adapters.
|
| -- David
------------------------------
From: Ulrich Lauther <[EMAIL PROTECTED]>
Subject: Archos pcmcia Zip100 drive?
Date: 6 Oct 2000 22:19:48 GMT
Reply-To: [EMAIL PROTECTED]
Did anyone get the Archos Zip100 pcmcia drive get to work?
If so, which kernel/pcmcia version?
Thanks for any hints.
------------------------------
From: Andrey Vlasov <[EMAIL PROTECTED]>
Subject: Re: Installed new processors: now problems...
Date: Fri, 06 Oct 2000 15:34:41 -0700
Hi again,
I got your message. I still think that you have some problem with some
hardware
which doesn't like 133MHz FSB. Probably I will not able help here as I do
not have
SMP machine to make any tests but I guess that you should post here more
details about your equipment. For most people will rise first question
"which
motherboard do you have?" I read in this newsgroup that SMP kernel was
reporting CPU cache wrong. May be you have something similar to that case.
If it is true you need narrow the problem and direct it to kernel.org.
By the way "Could you say how performance of you computer changed after
you switched to new CPU?". Recently I upgraded CPU from 450MHz to
700MHz on 100FSB and really I did not remark big improvement as in
Windows (Used for Video editing) as in Linux (all another job). Bryce 4 in
Windows still calculate prety slow but in Linux I able download from
Digital camcorder movie and play it without remarkable delay.
Andrey
dsmith wrote:
> I had a server that was running 2 PIII 500's on a 100MHz bus. I
> installed new PIII733's that run on a 133MHz bus. If I boot linux into
> an SMP kernel, I will sometimes get errors. When the SCSI module loads,
> it hangs and reports that it is waiting for a TID on CPU 0 or something
> like that. It runs, and pauses, and runs, and pauses... If I look at
> /proc/cpuinfo, it shows both processors, but CPU1 is 0MHZ and 0
> BogoMIPS. If it happens, I can usually reboot and it will be FINE. I
> have a SIIG Ultra66 controller that worked fine before. Now if I put it
> in, the problem arises all the time, every time. If I take it out, it
> goes back to normal. If I tell the motherboard to force the processors
> back to the 100MHz bus, everthing seems to be fine.
>
> Is it more likely that hardware is causing problems? The new processors
> are Coppermines, while the old ones are Katmai's. Is there any reason
> the kernel would need to be recompiled for the Coppermines and/or the
> new bus speed? I wouldn't think so, but I can't imagine what else could
> be the problem. I can't see why the SIIG card would have a problem with
> the bus speed. Keep in mind that if I boot a non-SMP kernel, everything
> is FINE... Kernel version (both SMP and non) are 2.2.15.
>
> Thanks!
>
> -Dan
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: easy-to-setup CDR recommendation
Date: 6 Oct 2000 22:59:12 GMT
Reply-To: [EMAIL PROTECTED]
On Thu, 05 Oct 2000 16:10:58 GMT, Henry wrote:
>I'm looking for recommendations for a cd writer that's known to work
>well with linux (preferably easy to configure). I can get ide,
>parallel, or usb. Which of these would you recommended?
SCSI! (Those are, without a doubt, the easiest to set up under Linux,
and SCSI stuff has a bunch of other advantages.)
You should avoid parallel/USB CD-RWs unless you absolutely must have the
easy "pick-up and take anywhere" capabilities they provide. They're
more expensive than IDE, generally perform worse, and setup is more
complex under Linux especially for USB devices.
Most any IDE CD-RW should work well, though if I were you I'd stay away
from Sony products and the Iomega ZIPCD since they don't quite follow
the MMC-3 standard and you can have problems with them under Linux.
There's a comprehensive guide to setting up a CD-RW under
http://linuxdoc.org/HOWTO/CD-Writing-HOWTO.html . It looks complex, but
it's not as hard as it sounds. Basically, you:
0 - Attach IDE device to a free IDE slot
1 - Boot system with "linux hdX=ide-scsi" where X is the place you
attached the CD-RW.
2 - modprobe ide-scsi scsi_mod sr_mod sg
(NOTE: Some distros, like RedHat, have scsi_mod and sr_mod compiled
directly into the kernel. You will almost certainly need to modprobe
ide-scsi and sg no matter what.)
3 - You can read from the CD-RW using /dev/scd0 and write to it using
/dev/sg0.
>I'm assuming there's some good cd writing software for linux. Let me
>know if I'd be better off keeping Windows around for cd burning.
There are only 2 cd-writing programs for Linux; cdrecord and cdrdao.
Both are command-line-only, but there are a lot of graphical frontends
available like Xcdroast, gnome-toaster, and kreatecd. These programs
allow you a lot more control than most WinXX software. However, there
are no stable Linux programs for doing Packet Writing (that is, using a
CD-RW like an enormous floppy disk, where you can delete files you've
written to the CD.) If you need to do this, you may have to keep WinXX
around. However, multi-session support under Linux works well. HTH!
--
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....
------------------------------
** 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
******************************
