Linux-Hardware Digest #472, Volume #10 Sat, 12 Jun 99 11:13:26 EDT
Contents:
Slight problem with bjc600 driver in gs for BJC-4310SP (Leigh Wedding)
Re: TDA98XX BTTV Driver for RH 6.0 (Julius Longauer)
Re: Need help with Diamond SupraExpress V.90 (MG Measures)
Re: Promiscuous mode (leoxx)
Re: Searching: SMP Machines with many CPUS (Marc Mutz)
Re: Kernel says 'heck' on boot-up ([EMAIL PROTECTED])
Re: RAID options (Marc Mutz)
Re: Linux slowness caused by hard drive? (Marc Mutz)
problem about aic7xxx (Yanjun Yu)
Re: Graphic Card : ATI Expert 98 (Charles E Taylor IV)
Re: Flatbed SCSI Scanner hanging kernel ([EMAIL PROTECTED])
Re: URGENT!!! Modem problem (MOM)
Help: trying to compile for ali 15xx chipset? (tom)
Re: How do I compile drivers? ("1stFlight !")
Re: [Q] SCSI-emulating EIDE controller exists? (Johan Kullstam)
Re: CDrom help needed (Carl Fink)
Re: Sybase On Linux Question (Christoph Terhechte)
Re: Does the HP7200i ATAPI CD Writer need SCSI emulation (David Borowski)
----------------------------------------------------------------------------
Crossposted-To: comp.lang.postscript
Subject: Slight problem with bjc600 driver in gs for BJC-4310SP
From: Leigh Wedding <[EMAIL PROTECTED]>
Date: 12 Jun 1999 21:55:23 +1000
I am using the bjc600 printer driver in ghostscript when printing on
Canon Bubblejet BJC-4310SP printer. It works fine most of the time,
can print in colour. However whenever I print a circle, it comes out
distorted. I have made a rough sketch and included it on site shown below
distort.gif. I also tried the bjc800 driver
and the same thing happens.
To see image go to http://www.corplink.com.au/~lwedding/distort.gif.
The actual printed version appears much smoother than this rough drawing.
Setup info:
output from: gs -h
Aladdin Ghostscript 5.50 (1998-9-16)
Copyright (C) 1998 Aladdin Enterprises, Menlo Park, CA. All rights reserved.
Usage: gs [switches] [file1.ps file2.ps ...]
Most frequently used switches: (you can use # in place of =)
-dNOPAUSE no pause after page | -q `quiet', fewer
messages
-g<width>x<height> page size in pixels | -r<res> pixels/inch resolution
-sDEVICE=<devname> select device | -dBATCH exit after last file
-sOutputFile=<file> select output file: - for stdout, |command for pipe,
embed
%d or %ld for page #
Input formats: PostScript PostScriptLevel1 PostScriptLevel2 PDF
Available devices:
x11 x11alpha x11cmyk x11gray2 x11gray4 x11mono deskjet djet500 laserjet
ljetplus ljet2p ljet3 ljet4 cdeskjet cdjcolor cdjmono cdj550 pj pjxl
pjxl300 uniprint bj10e bj200 bjc600 bjc800 faxg3 faxg32d faxg4 pcxmono
pcxgray pcx16 pcx256 pcx24b pcxcmyk pbm pbmraw pgm pgmraw pgnm pgnmraw
pnm pnmraw ppm ppmraw pkm pkmraw tiffcrle tiffg3 tiffg32d tiffg4 tifflzw
tiffpack tiff12nc tiff24nc psmono psgray psrgb bit bitrgb bitcmyk pngmono
pnggray png16 png256 png16m jpeg jpeggray pdfwrite bbox pswrite epswrite
pxlmono pxlcolor nullpage
Search path:
. : /usr/share/ghostscript/5.50 : /usr/share/ghostscript/fonts
For more information, see /usr/share/ghostscript/5.50/doc/Use.htm.
Report bugs to [EMAIL PROTECTED], using the form in Bug-form.htm.
output from: echo "currentpagedevice /VersionString get ==" | gs -q
-sDEVICE=bjc600 -
(2.17.00 5/23/96 Yves Arrouye)
output from: ldd `which gs`
libz.so.1 => /usr/lib/libz.so.1 (0x4000b000)
libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x40019000)
libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x40057000)
libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x40061000)
libpng.so.2 => /usr/lib/libpng.so.2 (0x400f3000)
libm.so.5 => /lib/libm.so.5 (0x4011d000)
libc.so.5 => /lib/libc.so.5 (0x40126000)
libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x401e6000)
libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x401ee000)
I am using linux v2.2.9
Any clues as to the problem?
Please email as well as post to this group.
--
Leigh Wedding
[EMAIL PROTECTED]
------------------------------
From: Julius Longauer <[EMAIL PROTECTED]>
Subject: Re: TDA98XX BTTV Driver for RH 6.0
Date: Fri, 11 Jun 1999 15:48:00 +0200
John Connell wrote:
>
> I don't know the answer but sure would like to. Am in the same boat, Chris.
> I'll post any info as I find out, do the same, OK?
> John Connell
>
> Christopher Burke <[EMAIL PROTECTED]> wrote in message
> news:7j2c6v$mi1$[EMAIL PROTECTED]...
> > Wanting anyone that can point me to a compilable version of BTTV that
> > included the TDA98XX module - or to a compilable version of the module or
> > a compiled version of the module for RH6.0.
> >
> > Seems my sound isn't working in bttv/xawtv because my card is BT878 +
> > some other chip. Wanting to give this one a go as the others didn't work.
> >
> > Unless someone else has a better solution for solving my perfect video
> > perfect tuner - no audio problem.
> >
Your should download bttv version 0.6.4 and compile it yourself. There
is
a file 'tda98xx.c' in the source tree.
Julius
------------------------------
From: MG Measures <[EMAIL PROTECTED]>
Subject: Re: Need help with Diamond SupraExpress V.90
Date: Fri, 11 Jun 1999 13:23:23 GMT
Damn !
I've just bought a SupraExpress and I've been trying to configure the
thing all day. I'm not going to give up just yet ...
So far,
1. setserial recognise the correct IRQ and base address of the card's
DOS mode comm port but doesn't seem to know what UART is on the board
(Not really a problem I can set this manually).
2. The PCI card shows up in the /proc/.../pci file but with the windows
IRQ and other stuff.
3. In Kppp the modem is always busy and therefor sending commands to it
always fails.
I'm hoping that by resetting the PnP bits of the card on bootup the card
will be initilise and the comm port will be usuable from linux.
I'll post to the news group if I have any luck,
Mike
On Thu, 10 Jun 1999, Dave Howland wrote:
> The SupraExpress V.90 doesn't work in linux... i spent two months trying
> to get it to... to no avail... i read somewhere that it doesn't emulate a
> com port, but interfaces with the system differently... a differently,
> unfortunately, that requires the winblows driver to work... i'd suggest a
> new modem...
>
> dave
>
> On Thu, 10 Jun 1999, Keith Ackerknecht wrote:
>
> > I have a Diamond SupraExpress V.90 ISA Pnp Modem. I have been trying to
> > install but I getting error messages when saying I/O error with every damn
> > address I try. I got it to dial once but I don't know how I did it. Any help
> > is appreciated.
> >
> > Keith
> >
> >
> >
> >
>
>
>
------------------------------
From: [EMAIL PROTECTED] (leoxx)
Subject: Re: Promiscuous mode
Reply-To: [EMAIL PROTECTED]
Date: Fri, 11 Jun 1999 14:03:57 GMT
on Thu, 10 Jun 1999 15:46:32 +0200, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> hi
> Anyone has an idea why my network card goes into promiscuous mode
> automatically a few times a day ?
> Is it a bug ? Is there a patch ? I don't know.
> i am using redhat 5.2 / kernel 2.0.36 / 2 NIC / ipfwadm
> i appreciate your help.
You may have been rooted (ie: hacked, cracked, watoozie).
--
JR
------------------------------
Date: Sat, 12 Jun 1999 14:35:44 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: Searching: SMP Machines with many CPUS
Johannes Nix wrote:
>
> If there out-of the box DSP Farms with the Price of one or two SMP
> Systems are available, please let me know.
>
> I definitively do not have resources to develope one of my own.
>
> The solutions (DSP Boards) I know each cost a lot more than one Alpha
> Dual SMP Machine.
>
If it's a university project: Are there no e-engineers there? With
today's DSP's coming with 'huge' on-chip RAM and on-chip DRAM interfaces
if you need even more and the intrinsic SMP-ability (e.g. SHARC
architecture: glueless SMP w/ up to 6 DSP's), it is just a matter of
PCB-layout, which can be made quite easily with the right software.
If it's a private project: Check out the varios evaluation board. They
sometimes come with very powerful com-devices and/or ethernet
connectivity. Bying many of them will decrease their price even
further... I saw some EVM's from Motorola for less than $300 (incl.DSP),
though you might want to use other families which provide more computing
power or lesser pricing.
Marc
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Kernel says 'heck' on boot-up
Date: 10 Jun 1999 19:29:22 PDT
Reply-To: [EMAIL PROTECTED]
In <7jmo9h$djv$[EMAIL PROTECTED]>, "Nigel Hawkes" <[EMAIL PROTECTED]> writes:
>Recently I've been tuning the SCSI BIOS settings on my system, trying to
>eliminate the 'data overrun' messages I'm getting on boot-up. I didn't
>succeed, but I did notice several messages like this in /var/log/messages:
>
>Dan kernel: Loaded 8 symbols from 3 modules.
>Dan kernel: heck:
>Dan kernel: sda: sda1 sda2 < sda5 >
>
>I guess it ought to be saying 'Partition check'. >
Ok maybe it's just too well healed to say 'Hell'.
--
Just my $0.02 worth.
Hope this helps,
Gordon
PS:
I do not tollerate spam. Any unsolicited bulk
e-mail will result in a complaint to your ISP.
------------------------------
Date: Sat, 12 Jun 1999 14:49:42 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: RAID options
Stewart McCullough wrote:
>
> Hi, I'm building a web server that will be running redhat 6.0. I want to run
> RAID 1 (mirroring) across 2 drives. I am considering 2 options:
>
> 1. use the software raid support present in the kernel.
>
> 2. buy a Mylex AcceleRAID 150
>
> questions:
>
> 1. how much CPU overhead is eaten up by the software approach? Would a single
> PII-400 be sufficient for handling this extra raid work under heavy conditions?
>
> 2. I know that there is a driver for the AcceleRAID 150, but according to the
> redhat hardware compatibility list, "Currently you cannot install Red Hat
> Linux's root partition onto a raid device." Does this mean that I would need a
> third drive to hold my root partition? Or would I simply install the root
> partition on one of my 2 RAID'd drives and hope that the drive I choose is the
> good one?
>
> 3. can I attach other SCSI devices to the AcceleRAID card (eg a tape drive), or
> is it only for raid drives.
>
> thanks for any wisdom,
>
> Stewart
1.) Always use hardware over software raid if possible. But:
2.) For a web server, don't use mirroring. Buy a huge amount of RAM and
let linux cache most web pages therein (does that automagically). Then
your bottleneck will not be reading the HTML/GIF's (though many people
think it is), but writing to the log file. Yes, really! No joke. If your
web server is somewhat loaded, one bottleneck is writing to the server
log. So use a raid level that increases write erformance, even if it may
reduce read performance. Mirroring does it vice versa: It increases read
performance up to factor n when using n disks, but surely decreases
write performance. I would recommend striping with big chunks (>=64k or
- if you want to optimize even further: keep the log location in an area
where sector number per cylinder is constant and set the chunk size to
the corresponding amount) or - if failsafeness is an issue - mirroring
over striping.
Marc
------------------------------
Date: Sat, 12 Jun 1999 14:53:05 +0200
From: Marc Mutz <[EMAIL PROTECTED]>
Subject: Re: Linux slowness caused by hard drive?
root wrote:
>
> Hey everyone,
> I've got a fresh install on Red Hat 6.0 running on a P-300 with
> 128MB of RAM and a 11GB EIDE hard drive. Linux is the only OS on this drive.
> I'm having a strange problem that happens quite often (every 10 seconds or so.)
> While working, all input form the keyboard or mouse will pause for 1-2 seconds.
> I then hear a click, or actually a "chunk" come from the hard drive, and then
> everything's ok until it happens again a few seconds later.
> This is a strange problem that has me stumped. I've used this hard
> drive before on another box runnig Linux, and it was ok. I just recently
> moved it over to this new box. Any help or suggestions would be appreciated.
>
> Thanks,
> -Josh-
use 'top' to check that there is no netscape or such running wildly in
the background. Last time I wondered that my box was slow it was due to
a netscape running with nice=0 and consuming 95% CPU time and 70M of
RAM...
Marc
------------------------------
From: [EMAIL PROTECTED] (Yanjun Yu)
Subject: problem about aic7xxx
Date: Mon, 07 Jun 1999 06:15:47 GMT
Hi, all
I'm trying to install Linux(TurboLinux 3.0.2) to an old HP net server
equipped with aic7770 SCSI card on board and 4 SCSI hds, but after the
installation is done, I can not boot from hd but got a error message
as below:
Kernel Panic - Unable to Mount Root FS...
I try to generate a boot disk from cdrom:/images/sysboot.img and try
to boot from floppy, it works this time, and everything work very well
then. But I do want to boot from hd directly, so would anybody give me
some suggestion for that?
Any help will be appreciate.
Yanjun
------------------------------
From: [EMAIL PROTECTED] (Charles E Taylor IV)
Crossposted-To: comp.os.linux.setup
Subject: Re: Graphic Card : ATI Expert 98
Date: Sat, 12 Jun 1999 08:33:56 -0400
In article <[EMAIL PROTECTED]>,
Chuck Snively <[EMAIL PROTECTED]> writes:
> Hi,
>
> When you run Xconfigurator, you don't have to chose a curtain video card. You can
> choose "Unsupported Card", go to the next screen and choose the SVGA server. I
> have done this. This works, I have the same card.
Why would you do this, when Xconfigurator works with this card as is?
Well, as is in the latest release of X they made available on their
errata pages for Redhat 5.2 ...
--
========================================================
Charles E Taylor IV <[EMAIL PROTECTED]>
========================================================
Visit me on the web!
http://orangesherbert.ces.clemson.edu
========================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Flatbed SCSI Scanner hanging kernel
Date: Fri, 11 Jun 1999 13:37:42 GMT
Your situation is different than mine, however,
the errors are the same. This being said I'll tell
you what I did and let you figure out if its of
help.
I'm running Redhat 6.0 using an adaptec 2940UW
SCSI host adaptor. I've been trying to get a old
SCSI hard drive to work but I was getting the same
messages as you during boot up.
I thought all my problems dealt with SCSI
termination. I was wrong.
I ended up going into the SCSI BIOS by pressing
ctrl-A at the boot up prompt.
After hours of troubleshooting I began resorted to
going into the BIOS scsi configuration and
changing individual flag settings and then
rebooting. It it didn't work I set the flag back
and changed another one. Scientific, I know.
I went into SCSI configuration and set the "Wide
negotiation" flag for the scsi ID corresponding to
my hard drive from "yes" to "no". This turned out
to be the only setting I needed to get my drive to
work.
-dan
In article <[EMAIL PROTECTED]>,
Jordi <[EMAIL PROTECTED]> wrote:
> Hello!
>
> I have a problem with my SCSI card & modem.
> Using kernel 2.2.9, I try to boot linux with the
scanner switched on.
> When it starts the SCSI driver, I get this
message:
> aec67x_detect:
> ACARD AEC-6666671X PCI Ultra/W SCSI-3 Host
Adapter:0 IO: 6300, IRQ:
> 11
> ID: 2 Color FlatbedScanner_90062
> ID: 7 Host Adapter
> scsi0 : ACARD AEC-6710/6712 PCI Ultra/W SCSI-3
Adapter Driver V1.0
> scsi: 1 host
> scsi: aborting command due to timeout: pid 2,
scsi 0, channel 0, id 2,
> lun 0, Request Sense 0000001000 workingu=1
last_cmd=2 quhdu=2 quendu=2 r
> 0=6 r 1=2c r2=cf r 3=3 r 4=0 r 5=0 r 6=0 r 7=10
r 8=0 r 9=0 ra=0 rb=0
> rc=0 rd=0 re=0 rf=0 r10=36 r11=20 r12=0 r13=0
r14=4 r15=2 r16=80 r1c=a1
> r1f=37 in_snd=0 r20=9 r22=1
> scsi: aborting command due to timeout: pid 2,
scsi 0, channel 0, id 2,
> lun 0, Request Sense 0000001000 workingu=1
last_cmd=2 quhdu=2 quendu=2 r
> 0=6 r 1=2c r2=cf r 3=3 r 4=0 r 5=0 r 6=0 r 7=10
r 8=0 r 9=0 ra=0 rb=0
> rc=0 rd=0 re=0 rf=0 r10=36 r11=20 r12=0 r13=0
r14=4 r15=2 r16=80 r1c=a1
> r1f=37 in_snd=0 r20=9 r22=1
> SCSI host 0 abort (pid 2) timed out - reseting
> SCSI bus is being reset for host 0 channel 0
> SCSI host 0 channel 0 reset (pid 2) timedout -
trying harder
> SCSI bus is being reset for host 0 channel 0
> After this, the kernel hangs.
> The kernel, however, does boot if the scanner is
not on.
>
> Anyone can tell me what can I do with this? I
would greatly appreciate
> any kind of help to make this work.
>
> Thanks in advance,
>
> Jordi
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (MOM)
Subject: Re: URGENT!!! Modem problem
Date: Sat, 12 Jun 1999 12:57:38 GMT
Nop its not a winmodem or can an ISA card modem be a winmodem!?!?!?!?
Thanks anyway...
On 11 Jun 1999 15:50:43 GMT, [EMAIL PROTECTED] wrote:
>
>>I have a USR 56K internal modem and i can�t get him to work fine.
>
>If it's a USR Winmodem then forget it. Winmodems don't work with
>Linux. You will need to get a new modem that is Hardware driven
>rarther thansoftware driven.
>
>Ken
>
>Please remove the DHFJAU before replying by email.
>
>***** Posted via the UK Online online newsreader *****
>
> Go to http://www.ukonline.co.uk to find out
> about other online services we offer our subscribers.
>
------------------------------
From: tom <[EMAIL PROTECTED]>
Subject: Help: trying to compile for ali 15xx chipset?
Date: Fri, 11 Jun 1999 10:00:16 -0500
i have tried patching a 2.2.9 kernel and i have also tried to simply
compile the 2.3.6 kernel so that i can get the ide controller recognized
on this asus p5a or whatever it is with ali chipset and k6.
unfortunately, although the chipset now appears when i do make xconfig,
or make config or make menuconfig, both with the 2.2.9 source and the
2.3.6 source, the option for the ALI 15xx chipset is grayed out. so the
contents of the patch are apparently there, but i am unable to select
it. yes dma is enabled and everything and i have replace hdparm in
/sbin with hdparm v3.5 from the site at
www.dyer.vanderbilt.edu/linux/drivers/udma or whatever it is. has
anyone got this to work and if so how?
much appreciated
tom bergerson
ingenious, inc.
------------------------------
From: "1stFlight !" <[EMAIL PROTECTED]>
Subject: Re: How do I compile drivers?
Date: Sat, 12 Jun 1999 10:00:16 -0400
Could someone just answer the question? Please! Inquiring minds want
to know :)
Darryl
Robert Schapiro wrote:
> How do I compile this kernel?
--
"Though we are not now that strength which in old days moved earth and
heaven, that which we are, we are; one equal temper of heroic hearts
made weak by time and fate but strong in will to strive, to seek, to
find and not to yield"
Tennyson's "Ulysses"
------------------------------
Subject: Re: [Q] SCSI-emulating EIDE controller exists?
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: 12 Jun 1999 10:26:48 -0400
Abdullah Ramazanoglu <[EMAIL PROTECTED]> writes:
> Skaya wrote:
> >
> > Abdullah Ramazanoglu wrote:
> > >
> > > Hello,
> > > Think of a PCI or off-case disk controller card that that emulates some
> > > popular SCSI controller, and drives 7 or 15 EIDE disks (possibly UDMA)
> > > with only one IRQ line, DMA, and I/O range. I think that such a card,
> > > coupled with RAID-5 support in Linux, and some additional RAM (fairly
> > > cheap nowadays) to cover performance issues of EIDE and RAID-5, would be
> > > very cost effective in building reliable, fast and large arrays.
> > > Is there such a thing? Or am I asking (or dreaming) too much?
> > > Thanks for putting up :)
> > > --
> >
> > that would cost you a lot of money, for sure.
> >
> > if you really *damn fast* I/O, go for UW disks and use raid (striping).
> > they leave IDE disks far behind.
> >
> Well, I am after a good cost/benefit balance. UDMA EIDE disks have 1/3
> price of an average SCSI disk.
> If you compare 7200rpm (and up) disk with EIDE, or average SCSI disks
> with PIO mode IDE disks, there may be a huge performance difference. But
> I don't "believe" that there would be much performance difference
> between an average SCSI disk and an UDMA EIDE disk.
there isn't. so long as you are on a *single* drive. two drives off
the same IDE controller will conflict with each other. two SCSI
drives off the same SCSI adapter will hardly know the other one is
there (the SCSI bus is usually faster than several drives).
> > real life example : while copying 200 megs from IDE to IDE, our server
> > feels quite slow. you don't even notice while copying from SCSI to SCSI,
> > and it also runs quicker.
> >
> Driving them in PIO mode directly by CPU would eat up CPU cycles. But If
> the controller emulates a SCSI card, then so far as Linux and CPU
> concerned, it is a SCSI card, and disks are SCSI disks. So it would eat
> up no more CPU than a real SCSI controller with real SCSI drives. Of
> course, the "translation" between SCSI and IDE protocol is the
> controller's job.
>
> > IDE may be cheap, but for the purist who wants raw power, get SCSI.
> >
> > also, problem of raid 5 is that it's very cpu greedy. without a
> > specialized controller, don't think about exceeding 10 to 20 mbytes/sec.
> > correct me if I'm wrong about this last point.
>
> I agree. It is the main drawback for this theory of mine. And finding a
> "SCSI-emulating EIDE controller with hardware RAID5 and battery backed
> cache" is not making sense. (Such a controller being a "server class"
> thing, it conflicts with the compromises made by using IDE. Also it's
> price would lessen the savings made by using IDE.) On the other hand, I
> would prefer using a dedicated "RAID5 server" than to use a hardware
> RAID card and SCSI disks. It would be much cheaper for me. I can build a
> Celeron-400 machine with 128MB ram for as low as $500 which is probably
> cheaper than the SCSI controller alone.
this is false. a SCSI adapter can be had for between $50 and $200.
> There are many niches for this approach. I, for one, plan to start
> up a home brew web hosting facility for free (or minimal fee to
> cover expenses). I just want to be "in" and get some experience,
> barely covering my expenses. So I need huge disk area with
> "acceptable" performance. In a standard PC I can install as much as
> 8 EIDE disks (putting CD and backup drive and everything to another
> machine). Assuming 15GB disks I can expand no more than
> 120GB. Also, 8 disks would eat up 4 irq lines. For me, such a
> controller would be invaluable.
you orignal questions come up. does this IDE mega controller exist?
how much would it cost? would the price of such a IDE mega controller
card be higher or lower than a SCSI adapter?
if the IDE doesn't exist at all, or if it exists but at an
unreasonable price, your whole argument is moot.
--
J o h a n K u l l s t a m
[[EMAIL PROTECTED]]
Don't Fear the Penguin!
------------------------------
From: [EMAIL PROTECTED] (Carl Fink)
Subject: Re: CDrom help needed
Date: 12 Jun 1999 13:09:55 GMT
Reply-To: [EMAIL PROTECTED]
On Sat, 12 Jun 1999 06:07:50 GMT root <[EMAIL PROTECTED]> wrote:
>I can't seem to get any CDs mounted. I have iso9660 compiled in as
>it shows when I cat /proc/filesystems
[details omitted]
I would tend to take the kernel at its word: you may have a hardware
problem. Things I would try:
1)A different CD-ROM. (I'm sure you've tried this.)
2)Another OS, if you can dual-boot. That is, boot
into Windows or DOS and see if they can see the
drive.
3)Remove the cables, blow off any dust on the contacts
(on the cable and the drive and on the controller),
clean the contacts with isopropyl alcohol and let
them dry thoroughly, and reattach everything. In my
computer repair days, about half the problems I saw
were incorrectly attached cables.
--
Carl Fink [EMAIL PROTECTED]
"This fool wishes to reverse the entire science of astronomy."
-Martin Luther on Copernicus' theory that the Earth orbits the sun
------------------------------
From: Christoph Terhechte <[EMAIL PROTECTED]>
Crossposted-To: comp.databases.sybase
Subject: Re: Sybase On Linux Question
Date: 7 Jun 1999 07:19:31 GMT
In comp.databases.sybase Bryan Scott <[EMAIL PROTECTED]> wrote:
> I've got three sybase servers running on devices like /dev/sdb2 and
> /dev/hda1 and so forth and they work fine.
There's yet another possibility to achieve unbuffered disk-i/o. Put
the database files on an ext2-partition that you have mounted with the
"sync" option (the default is async). You can specify the sync option
when mounting the partition interactively, or you can put in
/etc/fstab. This way, you avoid inconsistencies when the system
crashes, as no data will be lost in unwritten buffers.
Database access is still going through the ext2 filesystem, though. It
would be preferable if Linux supported true character raw devices.
--
Christoph Terhechte <[EMAIL PROTECTED]>
"Toto? - I have a feeling we're not in Kansas anymore."
------------------------------
From: [EMAIL PROTECTED] (David Borowski)
Subject: Re: Does the HP7200i ATAPI CD Writer need SCSI emulation
Date: 11 Jun 1999 14:24:45 GMT
Paul John ([EMAIL PROTECTED]) wrote:
: Relevant config:
: SuSE 6.1 kernel 2.25
: KDE
: HP 7200i IDE CD writer
: 8.4GB EIDE hda
: 4.2GB SCSI sda
: PPro 128M
: In order to get my HP 7200i IDE CD writer to work, I enabled SCSI
: emulation in the kernel & recompiled. Everything is fine, and the
: drives are recognised correctly by cdwrite & cdrecord. However,
: whenever I write a CDR, 9 times out of 10 it fails part way through
: due to a SCSI I/O error & the buffer becoming empty. I stepped the
: write speed down to x1 and the result is the same. Oddly enough,
: it does seem to work without error on a rewriteable CD. I have tried
: several brands of CDR's & all fail somewhere between 25 & 60%
: through the write process.
: I can only assume the problem lies with the SCSI emulation. I notice
: there are other cd recording apps about, keasycd for example and I
: wondered if there are any that will work with the CD writer in
: native ATAPI/IDE mode.
: Failing that, are there any special parameters I should be using for
: the SCSI emulation. At the moment, it is recognised by CDToast as a
: generic SCSI CD-Writer.
: Any help appreciated. Thanks.
I use the 7200i on a p133 on 2.26 without trouble.
some things to check: your 7200i should be set to master on the
second ide bus. some other cd readers if set to slave on the same bus
mess it up like mitsumi. 7200i is somewhat prone to dust if it is
in a dusty room and even afew months old, clean it!
Try cdrecord that's what I use. If you're running samba or other odd
network stuff like dhcp try turning it off.
Hope this helps.
------------------------------
** 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
******************************
