Linux-Hardware Digest #358, Volume #12 Sun, 27 Feb 00 20:13:09 EST
Contents:
Re: Jaton (Trident 9880) videocard problem (steve)
Re: Max modem speed ("Charles Sullivan")
TNT2 problem ("StyX")
Re: Debian Error.: Need Help Quick (Hartmut Figge)
Re: Update on Linux + OS/2 + Win2k system (Olafur Gunnlaugsson)
Re: TNT2 problem ("Sébastien Bouchet")
Re: Asus K7M vs FIC SDll athlon boards?? (Jay Robert Czajkowski)
Re: Disk Striping/Concat... (Andy Harrison)
Compaq DL380; what is a NC3163 NIC? What about its SCSI? (Andrew Donkin)
Re: Funky Zip drive problem, I need help! (Randy Cooper)
Re: Sporadic SCSI Bus Hang (I think). ([EMAIL PROTECTED])
bttv with the ATI TV-Wonder? (Philip Dean)
Re: Funky Zip drive problem, I need help! (Randy Cooper)
scanner setup: Copeye Smartkey 9636P ("Thomas de Roo")
Re: 3com network cards and 440BX (Dances With Crows)
----------------------------------------------------------------------------
From: steve <[EMAIL PROTECTED]>
Subject: Re: Jaton (Trident 9880) videocard problem
Date: Sun, 27 Feb 2000 17:19:26 -0500
Burak Serdar wrote:
> I have a Jaton Video 107 Blad 3D PCI/AGP (Trident 9880). Redhat
> installer attempts to use the SVGA server, but it crashes. I installed
> using the text mode installer, and tried a few trident chipsets, with no
> luck (again, crash). Should I just buy a new card, or is there anybody
> out there who made it work?
>
> Thanks.
It is quite possible you can use your card.
Rod Smith's article might help you.
http://www.rodsbooks.com/presario/
Steve Kang
------------------------------
From: "Charles Sullivan" <[EMAIL PROTECTED]>
Subject: Re: Max modem speed
Date: Sun, 27 Feb 2000 18:14:56 -0500
Douglas Bayne wrote in message <[EMAIL PROTECTED]>...
>
>I am using v.90 external modem (Compaq Microcom 510) on my Linux
>machine. I have absolutely no connection problems but I can't connect
>higher than 33.6K. My ISP supports v.90 and I did the latest firmware
>upgrade. The port is set at 115000bps. Are there any init strings I
>should try?
>
>Thanks,
>Doug Bayne
You may just have a crummy telephone line. Do you always connect
at 33.6? If sometimes yes and sometimes lower, it would seem that
the telephone line is the limiting factor. If both your modem and
your ISP support either 56Kflex or X2, you might try turning off
V.90 and going with one of these. (X2 is USR's pre-V.90 proprietary
protocol, 56Kflex is used by almost all others.) Try adding S32=66
to the end of your setup string.
------------------------------
From: "StyX" <[EMAIL PROTECTED]>
Subject: TNT2 problem
Date: Mon, 28 Feb 2000 00:17:30 +0100
I've just installed RedHat linux 6.1 on my system, and I have a problem with
my TNT2 drivers. When I try to start X, my screen goes black and turns it
self on and off! I have 32 MB of memory on the card, and when I installed
Linux, it counted 4 megs! There is something very wrong here, and can only
run text-based Linux, which I almost don't know at all... If any of you ever
experienced the same problem or have a solution to it, please contact me!
StyX, [EMAIL PROTECTED]
------------------------------
Date: Mon, 28 Feb 2000 00:32:06 +0100
From: Hartmut Figge <[EMAIL PROTECTED]>
Subject: Re: Debian Error.: Need Help Quick
Lucky wrote:
> >>Libc6 now predepends on debianutils, and run the "readlink"
> binary when
> installing. The problem is that debianutils predepends on libc6
> 2.1, and
you know, that potato is unstable yet?
there's a workaround for this bug from the debian-user list, or wait
until the bug is fixed.
<workaround>
On Tue, Feb 22, 2000 at 11:37:13AM -0700, john s anderson wrote:
[ snip ]
> However, I'd like to update the installation to at least frozen, and
> probably to unstable -- and that's where the problem comes in. After doing
> the base install, I add the frozen line to sources.list, do `apt-get
> update` -- all of which works as expected.
>=20
> Any attempt to install pretty much anything after that won't work, because
> of some problem between debianutils and libc6. It looks to me like they're
> mutually pre-dependent, or something. Here's the error output from
> `apt-get install libc6`:
>=20
> <output>
> penguin:~# apt-get install libc6
> Reading Package Lists... Done
> Building Dependency Tree... Done
> The following extra packages will be installed:
> g++ libstdc++2.10 cpp gcc debianutils libc6-dev libstdc++2.10-dev
> binutils
> ldso
> The following packages will be REMOVED:
> timezones libstdc++2.9-dev egcc
> The following NEW packages will be installed:
> libstdc++2.10 libstdc++2.10-dev
> 8 packages upgraded, 2 newly installed, 3 to remove and 582 not upgraded.
> Need to get 0B/7390kB of archives. After unpacking 9989kB will be used.
> Do you want to continue? [Y/n]
> E: Internal Error, Couldn't configure a pre-depend
> </output>
I was able to get aorund this, but I DO NOT recommend that anyone use
my method!! If you do, don't tell me about how screwed up your system
got.
Ok, it appears that libc6 predepends on a version of debianutils that
provides "readlink". debianutils predepends on a version of libc6 >=3D
2.1 ... catch-22. However, a *brief* glance at the libc6 pre-inst
indicates that readlink is being used to preserve time zone
configuration info.
I decided that breaking time zones was not a big deal. Of course,
this particular machine is not mission critical, so I must stress
again that this worked for me, it probably won't work for you. if it
does, cool. If not, you've been warned twice :)
I created a file "/usr/bin/readlink" with the following contents:
#! /bin/sh
true
I then installed libc6 using a --force-depends (this is your thrd
warning! This is not a good idea!) I found the libc6 deb in
"/var/cache/apt/archives", so I cd over there and type
dpkg --force-depends -i libc6_2.1.3-4_i386.deb
Directly afterwards I ran apt-get:
apt-get -f install
Luckily, everything still worked, and I was able to proceed to
`apt-get upgrade', etc.
I'm fairly insane so the above is FYI :)
Luck,
--=20
Nathan Norman Network Magician, Eclectic Engineer
</workaround>
cu
--
hafi
------------------------------
From: Olafur Gunnlaugsson <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.setup,comp.os.os2.misc,comp.os.os2.setup.storage,comp.os.os2.setup.video
Subject: Re: Update on Linux + OS/2 + Win2k system
Date: Sun, 27 Feb 2000 23:50:43 +0000
[EMAIL PROTECTED] wrote:
> Network cards! I forgot the cards! 3COM 10/100-base-T. Suggestions?
>
>
good choise, work well in all os's, my favorites are Realteks, but these are
solid
> >"FIC SD11 ATX ATHLON Mainboard with BIOS Upgrade"
> >-or-
> >ASUS K7M
>
Both are very nice, have built systems around both
> >27GB Western Digital Hard Drive 7200rpm Ultra ATA/66
> >Note: This will be the second of the same drive. May elect to put
> >smaller drive in DRIVE 0 and use one of the 27GB in older system as
> >offline backup for new system.
>
No No No, get an IBM drive, faster and much less prone to failiures
> >Matrox G400 32MB
>
the G400 has some issues under OS/2 unless you want to use SDD/2, actually
have not tried it with linux, but it rocks on Wintendo.
>
> >Creative Labs AWE32
>
Where are you going to get one of those new ?
Get a Aureal PCI 810 based card, cheap and works with all OS's
------------------------------
From: "Sébastien Bouchet" <[EMAIL PROTECTED]>
Subject: Re: TNT2 problem
Date: Sun, 27 Feb 2000 19:02:58 -0500
StyX,
I happen to have the same problem with my Mandrake 6.0. (graphics card =
guillemot maxi gamer xentor 32) I don't care developing with vi, but lynx
really sucks ! Back to ass-suckin' win98, I'm currently downloading the
Nvidia development OpenGL support files from their web site, which are
supposed to bring a fairly good acceleration on Xfree86 3.3.5 and higher.
I'll keep the group informed of any interesting info.
Sebastien
------------------------------
From: Jay Robert Czajkowski <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Asus K7M vs FIC SDll athlon boards??
Date: Mon, 28 Feb 2000 00:02:42 GMT
[EMAIL PROTECTED] wrote:
> In this posting I present the main results of our investigations with an
> Asus K7M and with a Gigabyte 7IX motherboard.
>
> We experimented with several disks and with several operating systems.
>
> These are the disks
> Quantum Fireball 18.2 Gbyte,
> Western Digital 17.6 Gbyte,
> Samsung 20.4 Gbyte
>
> The operating systems we are using are the following:
> Dos, only for start-up
> Win98 and Win98-SE
> Linux 2.2.14 and 2.2.39
>
> Our configuration is the following:
>
> >- MB k7m (one version with SB16 and another - newer - one without)
> >- Proc. Athlon 500
> >- Linux 2.2.13/2.2.14 or 2.3.36 under Debian Potato
> >- 3com 3c905
> >- video V3800 ASUS
> >- Quantum Fireball 18G2 UDMA disk or Western Digital (wbc-ac418000e)
> 17G6 UDMA
>
> We also tried
> >- MB gigabyte GA 71X motherboard
>
> Main findings
> We cannot get a well-functioning file system with two K7M Asus
> motherboards
> with a VIA BusMaster chipset. With both motherboards we experienced the
> same problems. Under Win98 an Win98-SE we could not get sufficient
> performance for any of the drives and for Linux we could not get a
> stable
> file system.
>
> We also experimented with Gigabyte GA 7IX. Instead of a Via chipset is
> has an
> AMD chipset (Viper) for the IDE interface. With this board we could run
> stable under Linux as well as under Win98-SE. They both provided a good
> file system performance.
>
> Our conclusion is the following.
> Until now (16 jan 2000)
>
> 1. We could not get a stable Linux system get running on a K7M with VIA
> chipset: copying big files crashed the filesystem.
>
> 2. We could not get a well performing file system running under
> Win-98 (first edition and SE) on a K7M with VIA chipset.
>
> 3. We could get a stable and good performing file system (transfer
> speed about 20MB/sec depending on the disk) running under Linux on a
> Gigabyte 7IX motherboard with Viper chipset.
>
> Pierre Jansen
>
> In article <[EMAIL PROTECTED]>,
> Ranaur, the Elven Warlock! <[EMAIL PROTECTED]> wrote:
> >
> > > Anyone have opinions on these two boards. Or for that matter, on
> any
> > > Athlon board under linux, and in general?
> > >
> > I bought a K7m and couldn't run a stable Linux, even with the
> latest
> > (2.2.14/2.3.43) kernel. Then I changed to a FIC and got the very same
> > problems.
> > Problem is: Both IDEs have the VIA chipset that is still
> experimental. So
> > you can't use the IDEs. If you have a SCSI, all right. If not, but a
> > Pentium (I'll change my Athlon to a PIII).
> > If you don't mind to wait some months to have a stable Linux,
> get a K7M.
> > First: it an ASUS. Second it has many more featurs (onboard sound,
> that
> > isn't supported by Linux, but OSS), two serials (SD-11 has only one),
> four
> > USB (only two for SD) and AMR (but forget it ... you should never use
> an
> > AMR if you want to be happy) ;-).
> >
> > Maybe an Gigabyte can run more smooth, maybe not. I'm tired of
> losing hard
> > disks. I'll go to a Pentium.
> >
> > Good luck,
> > R.
> >
> > --
> > Posted via CNET Help.com
> > http://www.help.com/
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
I just bought a Gigabyte 7IX motherboard I am planning on running an Athlon
850mhz, 256mb of RAM and everything SCSI. I haven't got it finished yet, but
I let you all know how it works out.
------------------------------
From: Andy Harrison <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: linux.redhat.misc,comp.os.linux.misc
Subject: Re: Disk Striping/Concat...
Date: Sun, 27 Feb 2000 19:04:11 -0500
Thanx for the help. That led me to this link, which really made it
pretty painless, especially since I don't need to boot from the raid.
The only thing that was different than what it said, presumably since
I'm not booting from the raid, was that I set my persistent-superblock
to 0.
http://www.ostenfeld.dk/~jakob/Software-RAID.HOWTO/Software-RAID.HOWTO.txt
--
Andy Harrison
ICQ: 1467660 AIM: AndyHarrisonRR
Peter Eddy wrote:
>
> Take a look at:
> http://www.redhat.com/support/docs/tips/raid/RAID-INDEX.html
>
> Peter
>
> Andy Harrison wrote:
> >
> > I've only begun to look, but I'm wondering if anyone knows a way to do
> > this under Linux (running rh6.1 sparc). I'd like multiple hard drives
> > to appear as one drive similar to the way I can in Solaris Disk Suite
> > and Veritas volume manager.
> >
> > Tips or links appreciated. Thanx!
> >
> > --
> > Andy Harrison
> > ICQ: 1467660 AIM: AndyHarrisonRR
------------------------------
Subject: Compaq DL380; what is a NC3163 NIC? What about its SCSI?
From: Andrew Donkin <[EMAIL PROTECTED]>
Date: 28 Feb 2000 13:26:11 +1300
I'm eyeballing a Compaq DL380 (replacement for the 1850) but, as most
Linux users will understand, I'm hesitant about the onboard ethernet
transceiver. It is an "NC3163", and is so new that it isn't mentioned in
any of the usual places.
Is it a rebadge of the old Intel EtherExpresses, or ThunderLANs, found on
older Compaqs? What Linux network driver should *I* use?
Also: any hitches with the SCSI ("smart array") controller?
Thanks in advance!
--
_________________________________________________________________________
Andrew Donkin Waikato University, Hamilton, New Zealand
------------------------------
From: Randy Cooper <[EMAIL PROTECTED]>
Subject: Re: Funky Zip drive problem, I need help!
Date: Sun, 27 Feb 2000 18:30:52 -0600
Reply-To: [EMAIL PROTECTED]
Are you sure you are mounting the Zip drive correctly. If it is not being
mounted to the specified mount point, but you are writing to / reading
from that mount point, then in fact you are accessing the hard drive
containing the mount point.
What commands are you executing?
If /mnt were your mount point, what is the result of ls -l /mnt (it should
be "total 0").
Reply to: [EMAIL PROTECTED] (mail checked week nights and weekends).
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup
Subject: Re: Sporadic SCSI Bus Hang (I think).
Date: Mon, 28 Feb 2000 00:31:57 GMT
In article <[EMAIL PROTECTED]>, Thom Brooke <[EMAIL PROTECTED]>
wrote:
> SUMMARY:
> I'm getting an intermitent system hang/freeze under RH 6.1. It
> occurs after from 10 minutes to 2 hours of up-time. It appears to
> be SCSI related (in that any commands which involve the SCSI bus
> "stall"; everything else continues to run). Everything runs normally
> up to this point.
>
> I DID NOT observe any such hangs under RH 5.2.
>
> SYSTEM:
> Micron P90 Powerstation (circa 1994; no power management).
> 96M RAM
> BusLogic BT-946C, PCI SCSI host adapter
> ID 0: Connor CFP-1060s (sda)
> ID 1: Micropolis 4221-09 (sdb)
> ID 6: Plextor PX-20TS
> Diamond Alpine Speedstar 64 PCI w/2M
> MediaVision Pro Audio Spectrum Basic
> Motorola ModemSurfr 28.8
> 3Com 3c509B
>
> This is a Linux-only machine.
>
> Custom RedHat 6.1 Install (to select some server packages).
>
> Neither the sound card nor the modem are "configured". (Although 6.1
> install thinks it sees a second SCSI adapter on the sound card,
> listed as "alias host_adapter1 unknown" in /etc/conf.modules. But
> the Pro Audio Spectrum BASIC isn't supposed to have a scsi adatper).
>
> X11R6 is configured. Boots to run level 5.
>
> SYMPTOMS:
> Initially, everything seems to run fine. All SCSI targets are
> identified
> at boot time, and they all "work". I can mount/unmount the cdrom;
> all disk partitions are available.
>
> At some point, usually between 10 min and 2 hours, some command or
> application will "hang". No more disk activity. Linux continues to
> run, and I can do other things, as long as they don't involve direct
> disk access. For example, under GNOME I can open a new terminal
> window and do simple things. TCP stack still up (ping from remote
> machine
> answered, but ftp or telnet hangs -- needs disk access for logon).
> I have kept a terminal window open in which I look at
> /proc/scsi/BusLogic/0
> and /var/log/messages. These files are still available (through cache,
> presumably).
>
> The last lines in /var/log/messages are (all listed at same <time>):
> <time> foo kernel: scsi0 channel 0 : resetting for second half of
> retries.
> <time> foo kernel: SCSI bus is being reset for host 0 channel 0.
> <time> foo kernel: scsi0: Sending Bus Device Reset CCB #39677 to
> Target 6
> The "CCB #" varies from crash to crash. It usually resets Target 6
> (cdrom) at time of crash. I have observed one reset of Target 1
> which did NOT end in a crash.
>
> Contents of /proc/scsi/BusLogic/0 indicate a great deal of Data
> Transfer activity for Target 6 (cdrom), but no actual data read (cd
> not mounted). Data transfer attempted/completed by target:
> 0 5503 / 5503
> 1 2567 / 2567
> 6 14339 / 14339
>
> Target 6 also shows one Bus Device Reset (Requested, Attempted, and
> Completed). No Bus Device Resets for targets 0 or 1. No host
> adapter resets. No external resets.
>
>
> ACTIONS TRIED SO FAR:
> Reinstalled RH 6.1 twice. Tried KDE instead of GNOME.
>
> Ran "memtest86", lots. Six different tests, 18 passes each (over two
> sessions). Total of about 24 hours of memory testing. No errors
> detected.
>
> Checked SCSI termination (looked fine). Replaced SCSI cable, just in
> case. Host adapter termination enabled. Terminators installed on sda
> (Connor; ID 0). SCSI bus configuration looks like this:
> SCSI ID: 7 6 - 1 0
> Device: BT946C----CDROM----<stub>----sdb----sda
>
> Installed almost all RedHat 6.1 errata (i.e., all updates except
> "bind" and "pygnome" -- not running a DNS server, so bind shouldn't
> be applicable, should it?).
>
> Installed "mkinitrd" from 6.2beta (per RedHat suggestion).
>
> Reinstalled RH 5.2. PROBLEM WAS NOT OBSERVED UNDER 5.2.
>
> Reinstalled RH 6.1. Problem is back.
>
> ----------------------
>
> Anyone have any pointers on how to debug this? Does it sound like
> I'm even in the right ballpark? I can provide log files, version
> info, etc., if that would help.
>
> -- Thom Brooke.
> -- cut out "cut_out_" to de-spamify my address.
I would recommend that you head over to www.dandelion.com, which is the
site of the author of the buslogic SCSI driver. I had a similar
problem a few years ago between my buslogic 948 and my conner hard
drive. I applied a flash update to the buslogic's bios, which seemed
to solve the problem. The imformation, and the bios upgrade should
still be available at that site.
------------------------------
Date: Sun, 27 Feb 2000 04:17:12 -0600
From: Philip Dean <[EMAIL PROTECTED]>
Subject: bttv with the ATI TV-Wonder?
I have an ATI TV-Wonder tuner. From what I can tell, this card uses the
878 chip and bttv should support this. Does anyone have, or know of a
working TV-Wonder card? Also, I have not been able to get bttv to
compile. It complains that I do not have loadable module support
compiled into my kernel. Well, I checked the make xconfig, and it is
selected, and I have modules loaded. Does anyone have any suggestions?
Thanks for your time.
PhilD
------------------------------
From: Randy Cooper <[EMAIL PROTECTED]>
Subject: Re: Funky Zip drive problem, I need help!
Date: Sun, 27 Feb 2000 18:40:27 -0600
Reply-To: [EMAIL PROTECTED]
Are you sure you are mounting the Zip drive correctly. If it is not being
mounted to the specified mount point, but you are writing to / reading
from that mount point, then in fact you are accessing the hard drive
containing the mount point.
What commands are you executing?
What is returned when you enter the mount command?
If /mnt were your mount point, what is the result of ls -l /mnt (it should
be "total 0").
Reply to: [EMAIL PROTECTED] (mail checked week nights and weekends).
On Sun, 27 Feb 2000, Martin wrote:
> I have an internal ATAPI 100Mb Zip drive. It is installed as my second
> primary ide drive hdc. It seems to work just peachy, I can mount it and
> read and write to it.. The problem comes when I unmount, switch disk,
> and remount it--The contents of the first zip disk will automagically be
> copied to the next inserted one... This is very ungood and I have no
> idea what is going on.. Can someone please help me?
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
------------------------------
From: "Thomas de Roo" <[EMAIL PROTECTED]>
Subject: scanner setup: Copeye Smartkey 9636P
Date: Mon, 28 Feb 2000 01:48:22 +0100
Can someone help me with suggestions for how to setup my
Copeye Smartkey 9636P - scanner? This is a non-SCSI-scanner, but connects to
the parallel-port...
Thanks!
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: 3com network cards and 440BX
Date: 27 Feb 2000 20:04:04 EST
Reply-To: [EMAIL PROTECTED]
On Sun, 27 Feb 2000 22:25:02 +0000, Joel Phillips
<<[EMAIL PROTECTED]>>
shouted forth into the ether:
>I've been trying for a few days to get a P2 440BX chipset to work with 2
>different 3com network cards, one 3c509 and one 3c90x (both used as
>10BaseT). I've tried with RedHat 6.0 and Mandrake 6.1, both of which
>detect the card, but neither of which are able to use it - I can't ping
>other computers on the network etc.
Boot the machine, log in as root, and do an "ifconfig"... see what that
says. If it comes up blank, then the card hasn't been ifconfigged yet and
there's no way things will work. If that's the case, then go into
Linuxconf and edit the networking settings.
Also make sure that the modules are loaded or the drivers are compiled
into the kernel. The 509 uses 3c509.o and the 90x uses 3c59x.o .
Also check the routing table with "netstat -rn" and see if the default
gateway's OK. (I've been burned by that one several times.)
>but I'm not sure whether my BIOS is in some way conflicting with Linux.
>I've disabled the 'PnP capable OS' option, but nothing else.
Hmm. What does "cat /proc/interrupts" say after you've booted that
machine? Conflicts should stand out. I did once see a 440BX machine that
had onboard Ethernet and video both using IRQ 9, but that was probably a
fluke.
--
Matt G / Dances With Crows \ In the MS-DOStrix,
There is no Darkness in Eternity \----\ there is no fork().
But only Light too dim for us to see \
===== Usenet: ceci n'est pas une guerre des flammes =====
------------------------------
** 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
******************************
