Linux-Hardware Digest #363, Volume #14 Sat, 17 Feb 01 21:13:05 EST
Contents:
Re: Should I abandon SCSI? (Eric P. McCoy)
Re: ALS 200 sound support (Jim Cochrane)
Re: Should I abandon SCSI? ("Ron Reaugh")
Re: Should I abandon SCSI? ("Ron Reaugh")
Re: Anyone using a Matrox G450 with 2 Monitors? (Mark Andal)
Re: Redirecting Serial port input to a file ("The Spook")
Re: Should I abandon SCSI? ("sad")
new scsi drive woes (Matt Garman)
Re: Promise Ultra100 (Noble Pepper)
Re: Should I abandon SCSI? (Stuffed Crust)
Re: Zip 250 as easy as Zip 100? (Dances With Crows)
Re: FIXED! Re: Update: Install hangs computer ("D. Stimits")
Can't get CD-RW working under RH 7.0 ("Sean Murphy")
Re: Should I abandon SCSI? ("Ron Reaugh")
Re: Should I abandon SCSI? ("Ron Reaugh")
Re: Should I abandon SCSI? ("Ron Reaugh")
Re: Can't get CD-RW working under RH 7.0 ("Rinaldi J. Montessi")
Re: Should I abandon SCSI? (Stuffed Crust)
----------------------------------------------------------------------------
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
From: [EMAIL PROTECTED] (Eric P. McCoy)
Date: 17 Feb 2001 16:13:49 -0500
"Folkert Rienstra" <[EMAIL PROTECTED]> writes:
> : With that kind of reliability raid 0 is out of the question, regardless if
> : IDE or SCSI.
> Why? Raid0 is only marginally less reliable.
No, it is significantly less reliable than even a single disk. Pretty
much exactly half as reliable. The more disks you put in a RAID0
array (technically you can only have two, but lots of people seem to
support more), the less the reliability of the entire array.
> Granted, forgetting the stripe-size can be lethal but so was forgetting CHS
> settings in the old days. And you can write it down, can't you?
So what if a disk fails? That's what RAID is really for, not for
performance. And it's why striping is RAID zero: because it's not
really RAID at all.
--
Eric McCoy <[EMAIL PROTECTED]>
"Knowing that a lot of people across the world with Geocities sites
absolutely despise me is about the only thing that can add a positive
spin to this situation." - Something Awful, 1/11/2001
------------------------------
From: [EMAIL PROTECTED] (Jim Cochrane)
Subject: Re: ALS 200 sound support
Date: 17 Feb 2001 14:15:07 -0700
I'm sure someone will correct me if I'm wrong, but it sounds like the
instructions you quoted imply that you need to configure and recompile your
kernel. You would choose the specified options by running the
configuration script that allows you to choose options for compiling the
kernel. Then you'll need to actually recompile and install the new kernel.
So you need to look for instructions on how to compile a new Linux kernel.
You can start by reading (assuming the path exists):
/usr/src/linux/README.
In article <[EMAIL PROTECTED]>,
Trent Sams <[EMAIL PROTECTED]> wrote:
>(sorry for cross posting, but I received no replies under redhat.misc.)
>
>I've found in messages that the ALS 200 can work under Linux. I'm new
>at this and don't understand the instructions. Can someone please
>explain what steps I need to follow to accomplish the following:
>
>
>> To use an ALS sound card under Linux, enable the following options in
>the
>> sound configuration section of the kernel config:
>> - 100% Sound Blaster compatibles (SB16/32/64, ESS, Jazz16) support
>> - FM synthesizer (YM3812/OPL-3) support
>> Since the ALS-007/100/200 are PnP cards, ISAPnP support should
>probably be
>> compiled in.
>>
>What file & where is it that needs to be modified. Is there a HOWTO
>that would help me?
>
>Thanks,
>Trent
>
--
Jim Cochrane
[EMAIL PROTECTED]
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sat, 17 Feb 2001 21:18:30 GMT
Bodo Mueller wrote in message <96lu8i$lf9cj$[EMAIL PROTECTED]>...
>
>"Ron Reaugh" <[EMAIL PROTECTED]> schrieb im Newsbeitrag
>news:tynj6.10682$[EMAIL PROTECTED]...
>>
>> Donovan Rebbechi wrote in message ...
>> >
>> >If you're willing to spend an awful lot of money on performance, then
>SCSI
>> >is worth considering. It really depends on how important I/O performance
>is
>> >to you. If you need a lot of reliable storage space, and performance is
>not
>> >so important,
>>
>>
>> Performance with IDE RAID 0 is tops.
>
>Learn to read, Ron
>
>Donovan spoke of " a lot of reliable storage space, and preformance is not
>so important
Nope. Anyone can read the above. The clear impression was that EIDE RAID 0
didn't provide excellent performance. That is false. I corrected it.
>With that kind of reliability raid 0 is out of the question, regardless if
>IDE or SCSI.
There is no "kind of reliability" here. So RAID 0 is being discussed.
>and I now of no ide 64 bit pci raid soluotion , therefore all present ide
>Raids are limitted by the pci bus to about 80 MB/s.
>and the maximum number of IDE Disks ( With internal ide raid solutions) is
>so far limited to 8 disks.
Right but so what.
>With small random IOs the mechanics of the disks
>is the limiting factor, only to use more disks can increase preformance
>here. With SCSI up to 15 disks per channel are possible, controllers have
up
>to 6 U160 Channels ( ICP-Vortex GDT7x63RN),with multichannel Fibere channel
>controllers evn more disks per controllers are necessary.
Anyone knows that for large server arrays SCSI is generally better but there
are some large OEM solutions beginning to appear using EIDE HDs.
>For sequential applications with an bandwidth of up to 40 MB/s sustained
ide
Up to 80MB/sec.
>raid soulutions are a viable solutione ( Please remember, that in most
>applications the data has to move two times over the PCI bus, first from
the
>source ( i.e.video grabber cards) to memory and from memory to disk, or
from
>disk to memory and form memory to VGA Card. Therefore only halve of the PCI
>Bandwidth is usable for storage. This is only a limitation of the 33 MHz 32
>Bit PCI bus,not a limitation of IDE.
So what is the relevance of bringing that up here?
>For exessive preformance like one needs in large data bank aplications SCSI
>ore fibre channel is still the technology of choice.
So far, right.
>> > you will get a lot more disk for your dollar with an IDE RAID.
>> >If performance (especially asynchronous I/O) is important then SCSI may
>be
>> >useful.
>>
>>
>> No, EIDE RAID 0 does very well.
>
>He did not exclude IDE, just stated that under these conditions scsi might
>be the better choice ;-)
I said it might not be so.
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sat, 17 Feb 2001 21:22:35 GMT
Donovan Rebbechi wrote in message ...
>On Sat, 17 Feb 2001 04:59:05 GMT, Ron Reaugh wrote:
>>
>>Donovan Rebbechi wrote in message ...
>
>>>If you're willing to spend an awful lot of money on performance, then
SCSI
>>>is worth considering. It really depends on how important I/O performance
is
>>>to you. If you need a lot of reliable storage space, and performance is
not
>>>so important,
>>
>>Performance with IDE RAID 0 is tops.
>
>RAID 0 is great for high performance desktops, but if you want a
>storage solution that's also *reliable*, it's useless. (BTW, shouldn't
>RAID 0 be "AID 0" ? THe R doesn't really apply)
>
>>> you will get a lot more disk for your dollar with an IDE RAID.
>>>If performance (especially asynchronous I/O) is important then SCSI may
be
>>>useful.
>>
>>No, EIDE RAID 0 does very well.
>
>Performance, yes. Reliability, no.
RAID 0 does not have any redundancy and in fact is less reliable than a
single drive. That however does NOT convert to saying RAID 0 is unreliable.
It is relatively reliable for single user workstation usage. Many use it on
news servers.
------------------------------
From: Mark Andal <[EMAIL PROTECTED]>
Subject: Re: Anyone using a Matrox G450 with 2 Monitors?
Date: Sat, 17 Feb 2001 23:25:06 GMT
I was reading two different magazine article's on your question.
One was the Feb. 2000 issue of Linux Magazine. It has even a
step-bystep process.
The author of the article "Taming the Two Headed Monster was Lou Grinzo
at : [EMAIL PROTECTED] although the article was on a G400 not G450.
The other more recent one was:
I think about Xinerama in either maximum linux or linuxjournal.
Hope that helps.
Mark A. X. Andal
John wrote:
>
> Hi,
>
> I'm planning to buy a system next week and I want to get this card. I was
> wondering what had to be done to a RedHat 7.0 system in order to make this
> card work with 2 monitors?
>
> Thanks,
> John
------------------------------
From: "The Spook" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Redirecting Serial port input to a file
Date: Sun, 18 Feb 2001 01:09:35 +0100
John C Bahr wrote ...
-- Cut --
>I have read the Advanced Bash Scripting HOWTO's section on I/O
>Redirection and
>I am sending an ASCII text file (100 lines with the numbers 1 through
>100 on their respective line) from the source computer like so:
>cat sent.txt > /dev/ttyS0
>
>I am attempting, from bash, to redirect the serial port data coming into
>the target to a text file.
-- Cut --
>
>I am running at a mere 9600 baud. Could I be seeing receive buffer full
>overruns?
-- Cut --
You may have trouble with line disciplines (an integral part of ports that
may be used for terminals, like serial ports).
To disable the line disciplines, you should use stty (but be careful to
embed it in a subshell, as changes done to a serial line is lost after the
line has been closed).
The following might work:
(stty raw; cat > received.txt) < /dev/ttyS0
Beware that cat will never end unless the port is closed by hardware
signals, as there is no interpretation of control characters in this setup
(Ctrl-D as end-of-file is normally caught by the line discipline and sent to
the process as an (inherently out-of-band) signal, but in this setup, the
line discipline is effectively disabled).
I hope this can get you somewhere.
/TRY
PS: This will *not* work:
stty raw < /dev/ttyS0; cat /dev/ttyS0 > received.txt
as /dev/ttyS0 is closed after the stty-command (with changes lost) and
reopened by the cat-command (with the original settings).
------------------------------
From: "sad" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sat, 17 Feb 2001 19:11:45 -0500
"Bodo Mueller" <[EMAIL PROTECTED]> wrote in message
news:96lu8i$lf9cj$[EMAIL PROTECTED]...
>
> "Ron Reaugh" <[EMAIL PROTECTED]> schrieb im Newsbeitrag
> news:tynj6.10682$[EMAIL PROTECTED]...
> >
> > Donovan Rebbechi wrote in message ...
> > >
> > >If you're willing to spend an awful lot of money on performance, then
> SCSI
> > >is worth considering.
IDE ATA-100 puts SCSI into the dark ages....
------------------------------
From: [EMAIL PROTECTED] (Matt Garman)
Subject: new scsi drive woes
Reply-To: [EMAIL PROTECTED]
Date: Sun, 18 Feb 2001 00:51:26 GMT
I just installed a new Quantum Atlas 10k II in my system. I have a Tekram
DC-390u2w controller card.
Well, I think I am having some serious geometry problems. Someone on the
comp.periphs.scsi group suggested looking into the "linear" boot option in
LILO. I toyed with that, after what I thought was a successful
installation, and I was getting TONS of SCSI errors dumped to the console.
I wrote down some of these errors, and they were something like this:
sym53c895-0: unexpected disconnect
sym53c895-0: script cmd (some non-human understandable codes here)
sym53c895-0: regdump
I'm pretty sure these are serious errors.
So then I booted from my DOS installation disk and ran fdisk on the new
drive. I created an 8 mb FAT12 partition. Then I rebooted into the
Debian install program to use Linux's fdisk to make my Linux partitions.
The strange thing is that after running DOS's fdisk, Linux's cfdisk
reported my drive as having 255 heads, 63 cylinders. It originally had
different numbers.
Anyway, I made the partitions, then tried to initialize a Linux partition.
After initializing one of my partitions, I got a lot of ext2-fs errors
dumped to the scrren. The errors looked something like this:
EXT2-fs error (device sd(8,6)): ext2_check_blocks_bitmap: Block #246 of
the inode table...
And the rest was overwritten by the installation dialog.
I went ahead with the install just to see what would happen, and sure
enough, when I booted, I got that warning that says you must manually run
fsck in order to boot.
So then I gave up and re-connected my old drive and disconnected the new
one.
This is driving me nuts, and I have no idea what's wrong.
Thanks for any suggestions!
Matt
--
Matt Garman, [EMAIL PROTECTED]
"I must not fear. Fear is the mind-killer."
from _Dune_ by Frank Herbert
------------------------------
From: Noble Pepper <[EMAIL PROTECTED]>
Subject: Re: Promise Ultra100
Date: Sat, 17 Feb 2001 18:55:18 -0600
Marc Ulrich wrote:
> Where can I find a Redhat 7.0 linux driver that will support the Promise
> Ultra100 ATA IDE controller? I am unable to install linux without this.
> Someone told me to go to the promise.com website & get it there, but
> promise only has a fasttrak driver -- which will not work.
>
> Thanks,
> Marc
>
>
>
Here's some sources of info
http://sweb.uky.edu/~mabran3/linux/hpt/
http://www.linuxnewbie.org/nhf/intel/hardware/udma66.html
http://www.linuxdoc.org/HOWTO/mini/Ultra-DMA.html
http://www.csie.ntu.edu.tw/~b6506063/hpt366/
http://www.linux-ide.org/ultra100.html
http://www-scf.usc.edu/~vibber/linux/
------------------------------
From: Stuffed Crust <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: 18 Feb 2001 01:00:25 GMT
In comp.periphs.scsi sad <[EMAIL PROTECTED]> wrote:
> IDE ATA-100 puts SCSI into the dark ages....
I think a Quantum Atlas 10K III with a Ultra320 interface would differ
with your opinion.
That's three times the bus bandwidth.
What was that about the dark ages again?
- Pizza
--
Solomon Peachy pizzaATfucktheusers.org
I ain't broke, but I'm badly bent.
Patience comes to those who wait.
...It's not "Beanbag Love", it's a "Transanimate Relationship"...
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Zip 250 as easy as Zip 100?
Date: 18 Feb 2001 01:04:33 GMT
Reply-To: [EMAIL PROTECTED]
On Sat, 17 Feb 2001 13:09:13 -0600, Jorge Alvarez staggered into the
Black Sun and said:
>I know how to use my current Iomega Zip drive (100 MB) on my Linux
>system. I have plans to upgrade to a newer Zip 250 in the near future.
>Can the later be used in such an easy way as the Zip 100?
If your ZIP 250 is the same type (IDE/ATAPI, SCSI, parallel-port) as
your old ZIP 100, then all you'll need to do is physically swap the
drives. If this is a USB ZIP 250, things might be a little more
difficult, but those work too. Check http://linux-usb.org/ for more
info on USB devices.
--
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....
------------------------------
Date: Sat, 17 Feb 2001 18:13:44 -0700
From: "D. Stimits" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup
Subject: Re: FIXED! Re: Update: Install hangs computer
Marc Ulrich wrote:
>
> This is posted in the hardware newsgroup as well. The fix was to simply
> update the bios (from my current version of 225 to version 245) for the
> Intel OR840 motherboard that I have.
Something else you may want to know about. The i840 chipset has a
defective IO-APIC that can hang up under heavy i/o or sometimes
mount/umount. It sends an irq 217 to handler that doesn't exist. Should
you run into hangs, you'll probably want to boot with kernel option
"noapic". This means only the first cpu will handle hardware irq's, and
thus become less responsive under load, but both cpu's will be used
still for applications that can use them normally. If you search through
/var/log/messages, you'll see notes about unexpected IO-APIC.
>
> Marc
>
> "David Cecere" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > I had a similar problem installing RH7 on my Micron TREK 2 notebook. I
> > solved it by telling the Phoenix BIOS that my installed OS was "Other"
> > instead of "Windows" or "DOS".
> >
> >
> > On Mon, 12 Feb 2001 15:32:22 -0500, Marc Ulrich
> > <[EMAIL PROTECTED]> wrote:
> >
> > >Nope, it isn't the PCI hardware causing the trouble. I've removed all
> > >PCI cards, and put the HDD on the Motherboard's IDE controller. The only
> > >expansion card in the system is a Matrox G400 AGP video card.
> > >
> > >Does anyone know what causes this?
> > >Marc
> > >
> > >Marc Ulrich wrote:
> > >>
> > >> I have a new computer that will not install linux RedHat 7.0. Here's
> the
> > >> deal:
> > >>
> > >> When the boot (from CDRom) reaches the spot: "PCI: probing PCI
> > >> hardware", the system freezes. Not even Crtl-Alt-Del to shutdown &
> > >> reboot will work. I tried using the expert mode because it said that
> > >> will bypass the PCI autoprobing. It does exactly the same thing.
> > >>
> > >> Here's what I think the problem is: The harddrive is using a Promise
> > >> Ultra100 harddrive controller (so that the HDD can use ATA100 instead
> of
> > >> ATA66 which is the max available via the motherboard). I would try just
> > >> attaching the harddrive to the motherboard controller, install linux &
> > >> then put it back on the Promise controller except that the PCI probing
> > >> is done everytime linux boots.
> > >>
> > >> Can anyone help me out?
> > >> Thanks,
> > >> Marc
> >
------------------------------
From: "Sean Murphy" <[EMAIL PROTECTED]>
Subject: Can't get CD-RW working under RH 7.0
Date: Sun, 18 Feb 2001 01:13:39 GMT
I'm trying to get my Plextor W8/4/32A working under my RH 7.0 and am not
having much luck. I've read a couple different FAQ's out there on the
web but they don't seem to work.
(http://lists.kando.hu/pipermail/gnometoaster/2000-September/000373.html)
(http://www.linuxgazette.com/issue57/feenberg.html)
First, I'm using the kernel shipped with RH 7.0. I didn't recompile it
or anything and the FAQ's give the impression that I shouldn't have to.
My system has a hard drive on ide 0, master.
The Plextor CD-RW is on ide 1, master.
A Creative Labs CD-ROM is on ide1, slave.
Using the setup described in both FAQ's I've listed above, here is how my
/etc/modules.conf looks:
----
[root@ linux]# more /etc/modules.conf
alias eth0 tulip
alias eth1 3c59x
alias parport_lowlevel parport_pc
alias sound-slot-0 emu10k1
alias usb-controller usb-uhci
# added 2/17/00
options ide-cd ignore=hdc
alias scd0 sr_mod
pre-install sg modprobe ide-scsi
pre-install sr_mod modprobe ide-scsi
pre-install ide-scsi modprobe ide-cd
----
my /etc/rc.d/rc.local has the following line in it:
---
insmod ide-scsi
---
and my /etc/lilo.conf is:
----
boot=/dev/hda5
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
linear
default=linux
message=/boot/message
image=/boot/vmlinuz-2.2.16-22smp
label=cd
read-only
root=/dev/hda5
append="mem=128M, hdc=ide-scsi, hdd=ide-scsi"
image=/boot/vmlinuz-2.2.16-22smp
label=linux
read-only
root=/dev/hda5
append="mem=128M"
image=/boot/vmlinuz-2.2.16-22
label=linux-up
read-only
root=/dev/hda5
append="mem=128M"
----
and I choose the "cd" option when I boot. The pertinent information from
dmesg is:
----
hda: Maxtor 91728D8, ATA DISK drive
hdc: PLEXTOR CD-R PX-W8432T, ATAPI CDROM drive
hdd: CREATIVECD4831E, ATAPI CDROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: Maxtor 91728D8, 16479MB w/512kB Cache, CHS=2100/255/63
----
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
scsi : 1 host.
Vendor: CREATIVE Model: CD4831E TS030808 Rev: 1.00
Type: CD-ROM ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
sr0: scsi3-mmc drive: 16x/32x cd/rw xa/form2 cdda tray
Uniform CD-ROM driver Revision: 3.10
----
and finally, cdrecord -scanbus reports the following:
----
Cdrecord 1.9 (i686-pc-linux-gnu) Copyright (C) 1995-2000 J�rg Schilling
Linux sg driver version: 2.1.38
Using libscg version 'schily-0.1'
scsibus0:
cdrecord: Warning: controller returns wrong size for CD capabilities page.
0,0,0 0) 'CREATIVE' 'CD4831E TS030808' '1.00' Removable CD-ROM
0,1,0 1) *
0,2,0 2) *
0,3,0 3) *
0,4,0 4) *
0,5,0 5) *
0,6,0 6) *
0,7,0 7) *
----
So my system seems to only be seeing my CD-ROM and not the RW drive.
Anyone have any ideas? I'm a little confused by what appears to be
conflicting methods in the different FAQ's, so I don't even understand if
I should be setting both CD drives for ide-scsi, or not.
I'd especially be interested in hearing from someone with 2 drives
installed and seeing how they have things setup.
Thanks,
Sean
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sun, 18 Feb 2001 01:15:48 GMT
Eric P. McCoy wrote in message <[EMAIL PROTECTED]>...
>"Folkert Rienstra" <[EMAIL PROTECTED]> writes:
>
>> : With that kind of reliability raid 0 is out of the question, regardless
if
>> : IDE or SCSI.
>
>> Why? Raid0 is only marginally less reliable.
>
>No, it is significantly less reliable than even a single disk.
Nope.
> Pretty
>much exactly half as reliable.
Twice small is still small.
> The more disks you put in a RAID0
>array (technically you can only have two,
Nope, you can have as many as are connectable.
> but lots of people seem to
>support more), the less the reliability of the entire array.
Yep, but for small numbers the likelihood of failure is still small.
>> Granted, forgetting the stripe-size can be lethal but so was forgetting
CHS
>> settings in the old days. And you can write it down, can't you?
>
>So what if a disk fails? That's what RAID is really for, not for
>performance.
Not true for RAID 0.
> And it's why striping is RAID zero: because it's not
>really RAID at all.
RAID 0 is not really RAID and a logician could have fun with a red rose is
not really a rose.
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sun, 18 Feb 2001 01:16:18 GMT
sad wrote in message ...
>
>"Bodo Mueller" <[EMAIL PROTECTED]> wrote in message
>news:96lu8i$lf9cj$[EMAIL PROTECTED]...
>>
>> "Ron Reaugh" <[EMAIL PROTECTED]> schrieb im Newsbeitrag
>> news:tynj6.10682$[EMAIL PROTECTED]...
>> >
>> > Donovan Rebbechi wrote in message ...
>> > >
>> > >If you're willing to spend an awful lot of money on performance, then
>> SCSI
>> > >is worth considering.
>
>
>IDE ATA-100 puts SCSI into the dark ages....
Nope.
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: Sun, 18 Feb 2001 01:17:19 GMT
Stuffed Crust wrote in message <96n6r9$jav$[EMAIL PROTECTED]>...
>In comp.periphs.scsi sad <[EMAIL PROTECTED]> wrote:
>> IDE ATA-100 puts SCSI into the dark ages....
>
>I think a Quantum Atlas 10K III with a Ultra320 interface would differ
>with your opinion.
>
>That's three times the bus bandwidth.
And of little significance for all but server farms.
------------------------------
From: "Rinaldi J. Montessi" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Can't get CD-RW working under RH 7.0
Date: Sat, 17 Feb 2001 20:28:12 -0500
Sean Murphy wrote:
>
> I'm trying to get my Plextor W8/4/32A working under my RH 7.0 and am not
> having much luck. I've read a couple different FAQ's out there on the
> web but they don't seem to work.
> (http://lists.kando.hu/pipermail/gnometoaster/2000-September/000373.html)
> (http://www.linuxgazette.com/issue57/feenberg.html)
>
> First, I'm using the kernel shipped with RH 7.0. I didn't recompile it
> or anything and the FAQ's give the impression that I shouldn't have to.
>
> My system has a hard drive on ide 0, master.
> The Plextor CD-RW is on ide 1, master.
> A Creative Labs CD-ROM is on ide1, slave.
>
> Using the setup described in both FAQ's I've listed above, here is how my
> /etc/modules.conf looks:
> ----
> [root@ linux]# more /etc/modules.conf
> alias eth0 tulip
> alias eth1 3c59x
> alias parport_lowlevel parport_pc
> alias sound-slot-0 emu10k1
> alias usb-controller usb-uhci
>
> # added 2/17/00
> options ide-cd ignore=hdc
> alias scd0 sr_mod
> pre-install sg modprobe ide-scsi
> pre-install sr_mod modprobe ide-scsi
> pre-install ide-scsi modprobe ide-cd
> ----
> my /etc/rc.d/rc.local has the following line in it:
> ---
> insmod ide-scsi
> ---
> and my /etc/lilo.conf is:
> ----
> boot=/dev/hda5
> map=/boot/map
> install=/boot/boot.b
> prompt
> timeout=50
> linear
> default=linux
> message=/boot/message
>
> image=/boot/vmlinuz-2.2.16-22smp
> label=cd
> read-only
> root=/dev/hda5
> append="mem=128M, hdc=ide-scsi, hdd=ide-scsi"
>
> image=/boot/vmlinuz-2.2.16-22smp
> label=linux
> read-only
> root=/dev/hda5
> append="mem=128M"
>
> image=/boot/vmlinuz-2.2.16-22
> label=linux-up
> read-only
> root=/dev/hda5
> append="mem=128M"
> ----
> and I choose the "cd" option when I boot. The pertinent information from
> dmesg is:
> ----
> hda: Maxtor 91728D8, ATA DISK drive
> hdc: PLEXTOR CD-R PX-W8432T, ATAPI CDROM drive
> hdd: CREATIVECD4831E, ATAPI CDROM drive
> ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
> ide1 at 0x170-0x177,0x376 on irq 15
> hda: Maxtor 91728D8, 16479MB w/512kB Cache, CHS=2100/255/63
> ----
> scsi0 : SCSI host adapter emulation for IDE ATAPI devices
> scsi : 1 host.
> Vendor: CREATIVE Model: CD4831E TS030808 Rev: 1.00
> Type: CD-ROM ANSI SCSI revision: 02
> Detected scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
> sr0: scsi3-mmc drive: 16x/32x cd/rw xa/form2 cdda tray
> Uniform CD-ROM driver Revision: 3.10
> ----
> and finally, cdrecord -scanbus reports the following:
> ----
> Cdrecord 1.9 (i686-pc-linux-gnu) Copyright (C) 1995-2000 J�rg Schilling
> Linux sg driver version: 2.1.38
> Using libscg version 'schily-0.1'
> scsibus0:
> cdrecord: Warning: controller returns wrong size for CD capabilities page.
> 0,0,0 0) 'CREATIVE' 'CD4831E TS030808' '1.00' Removable CD-ROM
> 0,1,0 1) *
> 0,2,0 2) *
> 0,3,0 3) *
> 0,4,0 4) *
> 0,5,0 5) *
> 0,6,0 6) *
> 0,7,0 7) *
> ----
>
> So my system seems to only be seeing my CD-ROM and not the RW drive.
> Anyone have any ideas? I'm a little confused by what appears to be
> conflicting methods in the different FAQ's, so I don't even understand if
> I should be setting both CD drives for ide-scsi, or not.
>
> I'd especially be interested in hearing from someone with 2 drives
> installed and seeing how they have things setup.
>
> Thanks,
> Sean
I have a similar set up to yours hardware wise. I did recompile my
kernel for generic scsi support, although I've read it is not absolutely
required. Secondly, I used this link for setting things up. It was
written in part by the fellow who gave us cdrecord, so I trust his
abilities.
http://www.ibiblio.org/pub/Linux/docs/HOWTO/CD-Writing-HOWTO
--
Rinaldi]$
When we remember we are all mad the mysteries disappear and
life stands explained. - Mark Twain
------------------------------
From: Stuffed Crust <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Should I abandon SCSI?
Date: 18 Feb 2001 01:33:06 GMT
In comp.periphs.scsi Ron Reaugh <[EMAIL PROTECTED]> wrote:
>>No, it is significantly less reliable than even a single disk.
> Nope.
>>much exactly half as reliable.
> Twice small is still small.
> Yep, but for small numbers the likelihood of failure is still small.
*shrug*
Raid-0 and reliability:
Instead of loosing everything on your disk when one dies, you loose
everything on all disks when one dies. Twice nothing is still nothing.
Which is why if your data is of any importance at all, you won't trust it
to a RAID0. RAID1 is the minimum, and even then you should make regular
backups onto something physically seperate.
If RELIABILITY is a criterion, then RAID-0 is flat out.
RAID0+1 would work though... but if you're going to throw that many disks
at it, you might as well go to RAID-5.
But anyway.
- Pizza
--
Solomon Peachy pizzaATfucktheusers.org
I ain't broke, but I'm badly bent.
Patience comes to those who wait.
...It's not "Beanbag Love", it's a "Transanimate Relationship"...
------------------------------
** 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 by posting to comp.os.linux.hardware.
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
******************************
