Linux-Hardware Digest #330, Volume #9 Mon, 1 Feb 99 18:13:29 EST
Contents:
Re: AHA3940U and Linux 2.2.0 (Raymond Doetjes)
Experiences with Apex Outlook switches at high res/fast refresh? (gregK)
HP 2000C + Linux, will it work? ("Ben Goble, Lakewood Colorado")
Re: STB TV PCI tuner & Linux (Tom Herman)
Re: STB PCI TV blue screen or snow. (Henrik Carlqvist)
Help using HP-7200i CD-Writer (Alexander Smondyrev)
Dual Processor Support (Ron Liddle - Tactics)
Re: Adding second IDE controller (Vince Veselosky)
Re: Same Disk RAID and Mirroring (Andy Glew)
Re: ISDN Modems (Michael Cooper)
Re: Adding second IDE controller (Matt Kressel)
Re: 2.2.x PnP sound card (SB 16 / Vibra 16) (Ken Witherow)
Sound and Network conflict ("Aristotle B. Allen")
Exabyte Mammoth and cheap media? (Bryan K. Wright)
Re: 3COM sells crippled modems (Joe Pfeiffer)
Re: Backup recommendations? (Dave Records)
----------------------------------------------------------------------------
From: Raymond Doetjes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.misc,de.comp.os.unix.linux.misc,alt.uu.comp.os.linux.questions
Subject: Re: AHA3940U and Linux 2.2.0
Date: Mon, 01 Feb 1999 20:11:27 +0100
I have the same problem with my AHA2940.
Except that on your system it starts looking on both my computers a
(P166 and a Alpha 21264a 533MHz) it hangs before scanning the SCSI bus.
I think this is a serious bug in the AIC7xxx.
I am glad i'm not the only one ;-)
WHen you get somehelp will you keep me posted. I will keep you posted.
Raymond Doetjes
------------------------------
From: gregK <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.ibm.pc.hardware.video
Subject: Experiences with Apex Outlook switches at high res/fast refresh?
Date: Mon, 01 Feb 1999 16:29:26 GMT
I'm considering the purchase of a four port Apex console switch, but I'm a bit
concerned about what it will do to the quality of the video on my machine. I
picked my display because I like to run at a high refresh rate (around a
hundred Hertz)-will the switch be likely to handle 1152 by 864 at a reasonable
bit depth with that sort of refresh rate without introducing artifacts? Are
there other switches more likely to be able to do this?
Thanks in advance for your thoughts.
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Mon, 01 Feb 1999 12:49:17 -0700
From: "Ben Goble, Lakewood Colorado" <[EMAIL PROTECTED]>
Subject: HP 2000C + Linux, will it work?
I would like to know if anyone is
running Linux with the HP 2000C printer?
TIA
--
Ben Goble Lakewood, Colorado USA
bgoble at uswest dot net bgoble at nyx dot net
A Stranger and a Pilgrim on the Earth
------------------------------
From: Tom Herman <[EMAIL PROTECTED]>
Subject: Re: STB TV PCI tuner & Linux
Reply-To: [EMAIL PROTECTED]
Date: Mon, 1 Feb 1999 17:30:57 GMT
I'm using this handy-dandy little script to load my bttv
modules. This is for a Hauppauge WinTV card. You should
change accordingly for STB. Maybe you are loading
modules in another order?
HTH
Tom
--
#!/bin/sh
# handy functions for rmmod/insmod
function xrmmod () {
grep -qe "^$1" /proc/modules || return
echo rmmod $1
rmmod $1 || exit 1
}
function xinsmod () {
echo insmod $*
insmod -f $* || exit 1
}
# Source function library.
. /etc/rc.d/init.d/functions
# Source configuration options
#. /etc/sysconfig/bttv
I2C_VERBOSE=1
I2C_SCAN=1
I2C_DEBUG=0
TUNER_DEBUG=0
# 0: Temic PAL tuner
# 1: Philips PAL_I tuner
# 2: Philips NTSC tuner
# 3: Philips SECAM tuner
# 4: no tuner
# 5: Philips PAL tuner
# 6: Temic NTSC tuner
# 7: Temic PAL tuner
TUNER_TYPE=2
BTTV_RADIO=1
#BTTV_VIDMEM=0xa80
BTTV_VIDMEM=0xa0
# 0: Auto-Detect
# 1: Miro
# 2: Hauppauge (bt848)
# 3: STB
# 4: Intel
# 5: Diamond
# 6: AVerMedia
# 7: MATRIX Vision MV-Delta
# 8: FlyVideo
# 9: TurboTV
# 10: Hauppauge (bt878)
# 11: MIRO PCTV pro
#BTTV_CARD=2
BTTV_CARD=10
# prepare for crashing the box -- flush dirty buffers
sync; sleep 1; sync
export MODPATH=/lib/modules/current/misc
case "$1" in
start)
$0 stop
echo "Initializing TV and/or radio modules..."
xinsmod videodev
xinsmod i2c verbose=$I2C_VERBOSE scan=$I2C_SCAN i2c_debug=$I2C_DEBUG
test -f $MODPATH/i2c_chardev.o && xinsmod i2c_chardev
xinsmod tuner debug=$TUNER_DEBUG type=$TUNER_TYPE
xinsmod msp3400
xinsmod bttv radio=$BTTV_RADIO vidmem=$BTTV_VIDMEM card=$BTTV_CARD
if [ ! -f $MODPATH/tvee.h ]; then $MODPATH/readee > $MODPATH/tvee.h;
fi
exit 0
;;
stop)
echo "Deactivating TV and/or radio modules..."
xrmmod bttv
xrmmod msp3400
xrmmod tuner
xrmmod i2c_chardev
xrmmod i2c-dev
xrmmod algo-bit
xrmmod i2c
xrmmod videodev
exit 0
;;
restart|reload)
$0 stop
$0 start
exit 0
;;
*)
# do not advertise unreasonable commands that there is no reason
# to use with this device
echo "Usage: bttv {start|stop|status|restart|reload}"
exit 1
esac
--
The views expressed are the author's and do not necessarily
reflect the official position of GTE or any of its subsidiaries
------------------------------
From: Henrik Carlqvist <[EMAIL PROTECTED]>
Subject: Re: STB PCI TV blue screen or snow.
Date: Sun, 31 Jan 1999 18:18:55 +0100
Jens Sorensen wrote:
> I downloaded bttv-0.5.22.tar.gz and compiled the stuff. I then
> launched xawtv but all im getting is a blue screen or snow.
Did you select the right kind of card and the right kind of tuner in the
Makefile? Did you try to tune in a channel?
regards Henrik
--
spammer strikeback:
root@localhost [EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: Alexander Smondyrev <[EMAIL PROTECTED]>
Subject: Help using HP-7200i CD-Writer
Date: Mon, 01 Feb 1999 11:36:02 -0500
I am trying to write a CD using HP-7200i CD-Writer. I have IDE/ATAPI
interface and reconfigured the kernel to include SCSI emulation
as described in manual pages. I can now mount regular CD on CD-Writer
and read. Then I configured xcdroast and tried to write a few files
on CD disk (I was using CD-RW media). Everything seems to go well
until I tried to verify the results. I am mounting CD-Writer on
/mnt/cdw and it appears that /mnt/cdw is a file of 0 Mb size, rather
than a directory. Does anyone know what is happening? I will greately
appreciate any piece of advise.
- Alex.
------------------------------
From: Ron Liddle - Tactics <[EMAIL PROTECTED]>
Subject: Dual Processor Support
Date: Mon, 01 Feb 1999 12:48:59 -0500
Can anyone tell me how well Linux scales with dual processors.
Specifically, I'm looking at
Dual PII 400
Supermicro P6DGS (GX chipset)
Dual channel Adaptec UWSCSI (7895)
256 MB 100mhz RAM
8MB Matrox Millenium G200 video
I plan on using this to serve an Oracle application.
Any suggestions to a different configuration would be welcome.
TIA
--
===========================================================
Ron Liddle Tactics Inc.
[EMAIL PROTECTED] (803)278-5657
===========================================================
------------------------------
From: Vince Veselosky <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions
Subject: Re: Adding second IDE controller
Date: Mon, 01 Feb 1999 16:30:41 GMT
Jeff Taylor wrote:
> I have Slackware 3.6 (linux kernel 2.0.35) and am trying to get the
> second IDE controller working. Is this on ide1? Any help would be
> appreciated.
>
> Thanx,
> Jeff T
Jeff,
Here's an article that might help you out:
http://www.ssc.com/lg/issue34/jachim.html
It installs a second IDE controller in order to use if as a CDROM
server.
Vince
--
Control-Escape: Alternative Software
Linux help for beginners to advanced users.
http://www.control-escape.com/
------------------------------
From: Andy Glew <[EMAIL PROTECTED]>
Crossposted-To: comp.arch,comp.arch.storage,alt.os.linux,comp.periphs
Subject: Re: Same Disk RAID and Mirroring
Date: Mon, 01 Feb 1999 12:12:48 -0600
> 2) Assuming 4 data blocks + 1 parity block, regardless of whether you solve (1),
> you will at best need 4 diskOps to read a full "slice", 4 to write 1 block,
> 5 to write the full slice, 6 to write 2 blocks, 8 to write 3.
I believe that your calculations are incorrect (or based on a bad assumption).
Or, rather, they are correct, but don't state the relative cost:
1X (i.e. little or no relative cost) on read performance,
4X to write individual blocks,
3X to write 2 blocks (2.5X if you optimize by writing the 2 new, reading the other 2
old, and parity write),
2.6X to write 3 blocks (1.6X if optimized),
1.2X to write 4 blocks, an entire RAID group,
RAID is based on haviing EDC (error detecting codes) per disk block,
so you don't need to read the parity block except when the EDC indicates an
error.
If each part of the RAID group is an individual disk block, then
a RAID group represents 4 ordinary disk blocks. Conversely,
each individual disk block of a RAID group can be read individually.
Time to randomly read an individual disk block in single disk
RAID = 1 disk read = time to read an ordinary disk block.
Single disk RAID read slowdown = 1X = no slower than an ordinary read.
The locality issues are a concern, of course, but I see no reason
why track locality for reading is not identical to existing devices.
I.e. the disk blocks of a file could be at exactly the same locations,
all in the same track or cylinder.
Writes are more difficult. To write a random block requires a read
and a write of the data block, and a read and a write of the parity block.
That's bad. But then again, 99% of my disk is effectively never modified
- OS, software installation, etc. This is why I mentioned the possibility
of using ordinary writes initially, and then converting to RAID parity
later.
Note, however, that writing an entire RAID group is only 20% worse than
writing 4 ordinary disk blocks, IF it is allowable to put a RAID group in the
same track. If it is necessary to put the RAID data and parity blocks in different
tracks, then there will seldom be an advantage to such aggregate writes,
so the 4X impact for writes on an individual RAID block will stand.
Your note raises to question, however, of cylinder locality: do errors typically
wipe out entire cylinders? Or would it be acceptable to place RAID group members
on different tracks in the same cylinder - and could RPS (rotational position sensing)
help optimize layour here?
> True. However, the disks already do bad block management.Or the software does. Of
>course performance usually suffers,
> which would be prevented here, if all of the above concerns
> have been addressed. Still, the original impetus for RAID
> was to give you time to hot swap the failing disk before a second
> failure occured, because once the parity block starts being used for
> recovery, you're back in the unprotected zone. It's fault-tolerant,
> not fault-proof.
I wonder if you know how often disk bad block management is effective?
I.e. how often can disk error codes detect & correct a bad block, allowing
the data in the block to be recovered as the block is remapped to a spare,
as opposed to just detecting and remapping, but losing whatever
was already there?
Is this related to the access pattern?: detection and correction works, but only
so long as you periodically access the data? If you do not access the data
often enough, the likelihood is high that a second error will occur? My limited
knowledge of magnetic materials suggests that such flaws do grow over time,
especially with temperature cycling.
As I mentioned, my interest is inspired by a spate of block errors
on my storage media - tapes, Jazz disks, etc.
> The data I've seen falls in five categories: 1) single bit or short-bit-string
>transient errors (correctable w/
> retry).
> 2) ditto, but hard (permanently failed). (correctable, if short enough, with
>ECC).
> Must be followed up, relatively soon, with relocation.
This sort of error is the problem for an individual user
with lots of, say, Jazz disks on the shelf as an archive or backup.
Scanning and relocating will not happen any time soon.
> 3) single platter "scratch", usually a straight line,
> and nearly radial. (bad blocks). Usually just during manufacture. Any cause
>present during normal operation
> would lead quickly to (4) or (5).
> 4) whole platter failure. (head crash or platter electronics path failure).
> Note that a single stuck bit anywhere on the path takes out the whole
>platter.
> (Or drives it into fulltime ECC, and your days are numbered).
> 5) Whole disk assembly failure.
>
> Of course, as independant events, one can get multiple occurances.
> RAID is really just short-term insurance against a single occurance of
> (4) or (5). The various flavors trade cost/performance/availability.
I think that single disk RAID would also be insurance against (1) and (2)
errors, in situations where the data is not regularly scanned to afford
the opportunity for correction.
This applies to Jaz disks on the shelf.
I think it also applies to how many home users use their PC disks
- 90% idle
------------------------------
From: Michael Cooper <[EMAIL PROTECTED]>
Subject: Re: ISDN Modems
Date: Mon, 01 Feb 1999 17:47:12 GMT
I am using a USR 128K ISDN Adapter and it works fine although I had to work for
a coupls of weeks to get it there. Go to Red Hat's web site I am sure they have
some list of known working isdn's
Good Luck
Michael
Andrew Gregory wrote:
> I am due to get an ISDN line put in this week. I am currently connecting my
> small LAN to the Internet through my Linux box using IP masquerading through
> a Motorola 28.8kbs modem.
>
> I would like to continue doing this using the ISDN modem. I will create two
> ppp-up scripts, one for a 64kbps connection and one for a 128kbps
> connection.
>
> Does anyone know which ISDN modems work well with Linux and should I go for
> an Internal or external modem.
>
> I am running a pentium overdrive on a VLB motherboard with RedHat 5.1 with
> all the current updated packages on it.
>
> Andrew
------------------------------
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions
From: Matt Kressel <[EMAIL PROTECTED]>
Subject: Re: Adding second IDE controller
Date: Mon, 1 Feb 1999 16:46:59 GMT
Jeff Taylor wrote:
>
> I have Slackware 3.6 (linux kernel 2.0.35) and am trying to get the
> second IDE controller working. Is this on ide1? Any help would be
> appreciated.
>
Your second controller should be ide1, but you should not have to
configure this. Linux should autodetect it and make devices for your
drives at /dev/hdc and /dev/hdd for the third and fourth drives. Make
sure your second controller is enabled in the BIOS.
-Matt
--
Matthew O. Kressel | INTERNET: [EMAIL PROTECTED]
+--------- Northrop Grumman Corporation, Bethpage, NY ---------+
+--------- TEL: (516) 346-9101 FAX: (516) 346-9740 ------------+
------------------------------
From: Ken Witherow <[EMAIL PROTECTED]>
Subject: Re: 2.2.x PnP sound card (SB 16 / Vibra 16)
Date: 1 Feb 1999 18:24:07 GMT
[EMAIL PROTECTED] wrote:
>
> I have a Soundblaster 16 PnP clone (detected as Vibra 16). I don't want to
> set up DOS / Win95 to find the settings for it. I thought the 2.2.x (2.2.1 in
> my case) kernel had some kind of extended support for these devices? I know
> that my network card has "stolen" IRQ 5, so I can't use standard settings.
> How can I find the IRQ from Linux? Do I have to set it up in the BIOS?
Get the isapnptools package. It lets you setup all of your plugnpray
devices.
> I have checked the HOWTOs as usual, but I think a lot has changed in the 2.2
> kernel?
The drivers in the kernel were replaced with the OSS drivers.
------------------------------
From: "Aristotle B. Allen" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Sound and Network conflict
Date: Mon, 01 Feb 1999 16:06:28 -0500
Reply-To: [EMAIL PROTECTED]
I just set up my linux box from scratch, a clean install when I set it
up I built an NE2000 ISA card at 0x240 10, it worked fine
after that I ran sndconfig to get sound set up and had some problems
My card is a SB AWE 64
1) it diddn't find the card the first time I ran sndconfig
2) I ran it again and it seems to have found the card and set up
everything properly, awe, opl3, wav etc..
3) now I cant get out on the network
there are no apparent conflicts with io, irq, or dma, I checked but the
are all set properly
what is wrong and how can I fix it.... I need to use my network... but I
also want sound for some Quake2
Aristotle B Allen
Lucent Technologies
------------------------------
From: [EMAIL PROTECTED] (Bryan K. Wright)
Subject: Exabyte Mammoth and cheap media?
Date: 1 Feb 1999 18:33:18 GMT
Hi folks,
I just bought a shiny new Exabyte Mammoth (8900) tape drive.
I installed it in a rack with a bunch of older 8505s, popped in one of the
Fuji 160M 8mm tapes I've been using, started a backup and walked away. When
I came back, I found that nothing had been written on the tape, because
of I/O errors.
After doing some experiments and reading the fine print in the
manual, I decided that I needed to buy new tapes. The manual says
that the drive will read 8500 or 8200 format tape, but it won't write them.
It also says that it won't write metal particle tape.
Looking around for acceptable tapes, I find that the prices
range from $68 to $99 (!) per cartridge. I've been paying about $11 per
cartridge for the Fuji tapes. Ack! I don't want to spend a factor of
ten more per cartridge. I'll only be using this drive for monthly
backups, but (even at $68 each) that's over $800 per year for media.
At those prices, I'm strongly considering just sending the drive
back and buying a few more 8505s instead.
The bottom line is: does anyone know of either (a) a cheap
source for 8mm 170M AME tapes or (b) some way I can persuade my drive
to write on the cheaper tapes I've been using? (For reference: the
system the drive is attached to is a Linux box with an Adaptec 1542
controller.)
Thanks in advance,
Bryan
--
===============================================================================
Bryan Wright |"If you take cranberries and stew them like
Physics Department | applesauce, they taste much more like prunes
University of Virginia | than rhubarb does." -- Groucho
Charlottesville, VA 22901 |
(804) 924-7218 | [EMAIL PROTECTED]
===============================================================================
------------------------------
From: Joe Pfeiffer <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.os2.comm,comp.os.os2.misc,comp.os.linux.setup,comp.os.os2.setup.misc
Subject: Re: 3COM sells crippled modems
Date: 01 Feb 1999 12:47:14 -0700
[EMAIL PROTECTED] (Alan Boritz) writes:
>
> Maybe you should instead concentrate on educating your household on the
> benefits of NOT picking up the phone while you're using it, rather than
> blaming a manufacturer for YOUR problems?
Well, addressing his actual problem would involve installing a second
phone line (or cable modem, etc)... :)
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
------------------------------
From: Dave Records <[EMAIL PROTECTED]>
Subject: Re: Backup recommendations?
Date: Mon, 01 Feb 1999 15:24:54 -0500
[EMAIL PROTECTED] wrote:
>
> Hi all!
>
> I'm currently investigating what kind of backup solution I should get for
> our machines at home. I've a lot of stuff there I don't want to loose in a
> disk-crash...
>
> I've come to the conclusion that a SCSI-DAT (I've got SCSI-disks) would be
> the best solution, but also quite expensive. So, I was looking at Iomega
> Ditto drives as they're a lot cheaper. The 1/2GB versions seem to work
> very well, but are the Max-versions supported? How about the internal
> version; is it IDE, floppy or SCSI. I'd prefer SCSI.
>
> Any other ideas as to what could be a good solution for backups? We've
> three machines at home, but only two need to be backed up, so the datasize
> is about 2 * 1GB. Speed really doesn't matter, as the backups will be made
> while we're at work.
>
> I'd appreciate any hints as to what could be a good solution.
>
> Regards,
> Chakie
>
> --
> ---------------------+------------------------------------------------------
> Jan 'Chakie' Ekholm | CS at �bo Akademi University, Turku, Finland
> Linux Inside | I'm the blue screen of death, no-one hears you scream
Chakie;
I've got the Conner 4/8GB TR SCSI drive, use it externally on my Mac &
Linux and works fine... but tapes are expensive. I am getting ready to
buy the Seagate STD28000N from Surplus Direct for under $500 and tapes are
less than half the price. Lookup http://www.surplusdirect.com and it's
under Storage.
Regards;
Dave Records
--
Dave Records - Systems Analyst
Thirdware Systems, LTD
616.222.9556
------------------------------
** 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
******************************
