Linux-Hardware Digest #157, Volume #14 Wed, 10 Jan 01 16:13:06 EST
Contents:
Re: Radeon and XFree86 4.02 (Scott Alfter)
Re: After installing and removing a 2nd NIC, both are failing. ("Deanna Bonds")
Re: Problem with RedHat 6.2 and Promise Ultra100 (ATA100) (Pavlos Parissis)
Re: Problem with RedHat 6.2 and Promise Ultra100 (ATA100) (Pavlos Parissis)
Re: RH 7 doesn't work with 3C905? (Hal Burgiss)
Re: dual xeon problems (ekk)
Re: dual xeon problems (ekk)
Modem connects, but I can't use it. (James)
issues with sendmail working with ISP ([EMAIL PROTECTED])
Diagnostics Test Program for Linux? ([EMAIL PROTECTED])
Re: HPT366 errors (Tim Moore)
Re: L2-Cache of Pentium2 with Linux (Henrik Carlqvist)
Re: dual xeon problems (jwk)
Re: Support for my sound card? (Matt O'Toole)
Re: Partitioning For Install (Tim Moore)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Scott Alfter)
Subject: Re: Radeon and XFree86 4.02
Date: Wed, 10 Jan 2001 18:29:29 -0000
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
In article <93gp1f$ug0$[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote:
>Has anyone correctly configures XFree86 4.02 with an ATI Radeon video
>card? I tried, but I'm by no means an expert and experienced
>difficulties. I'd appreciate any tips.
>
>I have the following:
>Redhat Linux 7
>XFree86 4.02 (FTP'd from xfree86.org -- are RPMs available?)
>ATI Radeon DDR 32Mb
>Sony G400 Monitor
xf86cfg didn't have any Radeon boards listed, so I wasn't able to configure
modelines with it. However, everything else was set up in it, so I let it
create an XF86Config file and stole the modelines that had been generated by
xf86config under XF86 3.3.6 for the Xpert 98 I used previously. I now have
my Radeon (same one as yours) working well at 1280x960. All that remains
now is to get KDE2 working, as the only window manager I have right now is
twm (mine's an LFS system running Linux 2.4.0-pre12 with reiserfs).
_/_
/ v \
(IIGS( Scott Alfter (remove Voyager's hull number for email address)
\_^_/ http://salfter.dyndns.org
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.2 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6XKn3VgTKos01OwkRAuZDAKDKqdx6Td3V3oxIZDCZ9GnHx0D5wACghHYj
T8W8pQ43lkeDT5zrVI49Dbw=
=C1MW
=====END PGP SIGNATURE=====
------------------------------
From: "Deanna Bonds" <[EMAIL PROTECTED]>
Subject: Re: After installing and removing a 2nd NIC, both are failing.
Date: Wed, 10 Jan 2001 13:21:18 -0500
Hey
Was your cable modem internal or external (if it was the 3Com card I guess
it was internal). If you already had a NIC in your machine you could have
gotten the external modem which will connect to your hub and you then you
only need to configure your DNS and such. If it is internal and not
supported there is not a lot you can do other than trying to find some
driver that works with it. If it is not in the kudzu database you will have
to configure it by hand in your /etc/modules.conf file.
Deanna
"EvilPenguin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Well, I recently got a cable modem, so I attempted to install the NIC
> that came with the service. After wondering why it did not work and
> looking for a kernel module, I learned it was not supported.
>
> So I removed the card, and when during the boot proccess, the "searching
> for new hardware" failed, and so it still tried to bring up the
> network. For some reason, now eth0 does not work either. I tried to
> run kudzu manually and it seg faulted (with no other info). I already
> checked for an IRQ conflict, but to no avail. I also cannot make any
> changes for eth0 under linuxconf or the control panel.
>
> Now, it will boot up, but it will still fail at running kudzu, bringing
> up eth0 and eth1 (there is _no_ eth1!!!) and starting sendmail.
>
> Is there any solution short of rm -fr /???
>
> I'm running RedHat 6.1. eth0 is a Intel 100/10 PCI and has worked
> perfectly up until now. The faulty card was a 3Com 3C900B-TPO PCI.
>
> Any help would be greatly appreciated (really, I will give you god
> status if you can figure this out).
>
>
------------------------------
From: Pavlos Parissis <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.help,comp.os.linux.setup,comp.ibm.pc.hardware,comp.sys.ibm.pc.hardware.storage
Subject: Re: Problem with RedHat 6.2 and Promise Ultra100 (ATA100)
Date: Wed, 10 Jan 2001 20:41:37 +0200
Hello dan,
Look my post "Re: asus a7v promise controller prob booting" on 9/01 in
comp.os.linux.hardware
Pavlos
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I love having the feeling of being in control
while i have the sensation of speed
The surfer of life
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
------------------------------
From: Pavlos Parissis <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.help,comp.os.linux.setup,comp.ibm.pc.hardware,comp.sys.ibm.pc.hardware.storage
Subject: Re: Problem with RedHat 6.2 and Promise Ultra100 (ATA100)
Date: Wed, 10 Jan 2001 20:42:00 +0200
Hello dan,
Look my post "Re: asus a7v promise controller prob booting" on 9/01 in
comp.os.linux.hardware
Pavlos
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I love having the feeling of being in control
while i have the sensation of speed
The surfer of life
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
------------------------------
From: [EMAIL PROTECTED] (Hal Burgiss)
Subject: Re: RH 7 doesn't work with 3C905?
Reply-To: Hal Burgiss <[EMAIL PROTECTED]>
Date: 10 Jan 2001 13:46:10 -0500
On 10 Jan 2001 12:54:11 -0500, Hal Burgiss <[EMAIL PROTECTED]> wrote:
>
>You probably need to specify exact model, and module. I have several
>3C905Bs, and they all have always worked without problems with 3c59x
>module. Does IRQ, etc look OK?
PS -- Also, RH7.
--
Hal B
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
--
------------------------------
From: ekk <[EMAIL PROTECTED]>
Subject: Re: dual xeon problems
Date: Wed, 10 Jan 2001 13:38:40 -0500
Hello,
Thank you again. Actually, it seems to do okay with only one module. I haven't tested
the other one yet, I'm still pushing the first one hard to see if it will crash.
A while ago, you had asked if the RAM was ECC or not - I just found out that it is.
What ramifications does that have?
Another odd thing I noticed when I took out one of the memory modules - when both of
the memory modules were in there, the BIOS would check the memory 3 times, but with
only one modules it only checks it once. What does that mean.
Thank you
Ken
jwk wrote:
> On Mon, 08 Jan 2001 13:59:11 -0500, ekk
> <[EMAIL PROTECTED]> wrote:
> >Thanks for your help-
> >Memtest doesn't seem to crash, although the machine does hang before some of the
> >programs finish running - just like it always does when I have intensive processes
> >running. But, memtest itself does not crash.
> >Ken
> >jwk wrote:
> >
> >> On Sun, 07 Jan 2001 14:51:17 -0500, ekk
> >> <[EMAIL PROTECTED]> wrote:
> >> >I apologize, but is there a universal way to determine if the memory is ECC?
> >> >I know that there are 2 512 MB SDRAM Toshiba PC100-222-620R modules, but I can't
> >> >find anything on the web about this memory being ECC. I am trying memtest right
> >> >now.
>
> will it run with one module? then you should test them one after the
> other. Otherwise, is returning it to the shop an option? To each his own
> :-)
>
> Jurriaan
>
> --
> "Bother!" said Pooh as Q destroyed the universe.
> GNU/Linux 2.2.19pre6 SMP 2x1402 bogomips load av: 0.05 0.04 0.02
------------------------------
From: ekk <[EMAIL PROTECTED]>
Subject: Re: dual xeon problems
Date: Wed, 10 Jan 2001 13:41:26 -0500
Ah, I take it back - it just crashed, almost as soon as my last message was sent. I
will test the other module now-
Ken
jwk wrote:
> On Mon, 08 Jan 2001 13:59:11 -0500, ekk
> <[EMAIL PROTECTED]> wrote:
> >Thanks for your help-
> >Memtest doesn't seem to crash, although the machine does hang before some of the
> >programs finish running - just like it always does when I have intensive processes
> >running. But, memtest itself does not crash.
> >Ken
> >jwk wrote:
> >
> >> On Sun, 07 Jan 2001 14:51:17 -0500, ekk
> >> <[EMAIL PROTECTED]> wrote:
> >> >I apologize, but is there a universal way to determine if the memory is ECC?
> >> >I know that there are 2 512 MB SDRAM Toshiba PC100-222-620R modules, but I can't
> >> >find anything on the web about this memory being ECC. I am trying memtest right
> >> >now.
>
> will it run with one module? then you should test them one after the
> other. Otherwise, is returning it to the shop an option? To each his own
> :-)
>
> Jurriaan
>
> --
> "Bother!" said Pooh as Q destroyed the universe.
> GNU/Linux 2.2.19pre6 SMP 2x1402 bogomips load av: 0.05 0.04 0.02
------------------------------
From: James <[EMAIL PROTECTED]>
Subject: Modem connects, but I can't use it.
Date: Wed, 10 Jan 2001 19:30:07 -0000
Hi,
I've just installed SuSE 6.4 and have been trying to get my modem to
work, finally I've got to the stage where it dials and connects to my ISP
perfectly.
Unfortunatly if I then open netscape I immed iatly get the error
message;
"Netscape is unable to locate the server "Server". Please check
the server name and try again."
Also if I try to use the shell to telnet anywhere I get a similar
message.
Linux doesn't seem to be able to find or communicate with my modem, but
it is somehow able to dial-up.
What could be causing this?
Thankyou.
JL
My computer
cyrix 266
64MB RAM
SuSE 6.4
Sportster 28,800 external modem
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: [EMAIL PROTECTED]
Subject: issues with sendmail working with ISP
Date: Wed, 10 Jan 2001 19:43:18 GMT
issues with sendmail working with ISP
I have redhat 7.0 and I use install-sendmail to setup my sendmail
server but I have a issues. My ISP will not let me sendmail without
POPing the pop server first??
How can I do this?? I need to do a POP so the ISP can get my john
name/password and IP address
Please email me at [EMAIL PROTECTED]
Sent via Deja.com
http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED]
Subject: Diagnostics Test Program for Linux?
Date: Wed, 10 Jan 2001 19:50:18 GMT
Hi,
I am looking a a diagnostics Test program that run under Linux? Can
you help me on this?
kevinn
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Tim Moore <[EMAIL PROTECTED]>
Subject: Re: HPT366 errors
Date: Wed, 10 Jan 2001 20:17:42 GMT
http://www.<xx>.kernel.org/pub/linux/kernel/people/hedrick/
where <xx> is your country code.
http://www.csie.ntu.edu.tw/~b6506063/hpt366/
> After a while with kernel 2.4.0 I've got the following error message:
>
> hde: timeout waiting for DMA
> ide_dmaproc: chipset supported ide_dmatimeout func only: 14
>
> and the kernel locks.
>
> The chip is HPT366 and the disk is IBM Deskstar 45GXP.
--
timothymoore
bigfoot
com
------------------------------
From: Henrik Carlqvist <[EMAIL PROTECTED]>
Subject: Re: L2-Cache of Pentium2 with Linux
Date: Tue, 09 Jan 2001 21:20:48 +0100
Ralf Render wrote:
> Yes, my bios is set up to use the L2 Cache.
> I wrote a small program that runs through an array (512k) and do some
> calculations. A second version does the same with a loop of 100
> times.
Your cache is only 512k and some of it is probably also needed for
instructions. Also remember that Linux is a multitasking OS, a process
might not have all the cache for itself during the whole loop.
Try to make a new testprogram which uses an array which is much smaller
but still bigger than your L1 cache. What kind of result do you get
then?
regards Henrik
--
spammer strikeback:
root@localhost [EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (jwk)
Subject: Re: dual xeon problems
Date: 10 Jan 2001 20:38:59 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 10 Jan 2001 13:38:40 -0500, ekk
<[EMAIL PROTECTED]> wrote:
>Hello,
>Thank you again. Actually, it seems to do okay with only one module. I haven't
>tested
>the other one yet, I'm still pushing the first one hard to see if it will crash.
>A while ago, you had asked if the RAM was ECC or not - I just found out that it is.
>What ramifications does that have?
Basically, ECC allows you to survive one-bit errors and detect (by
crashing deliberately) two-bit errors. Normal memory may or may not
crash, depending on what memory locations randomly change. So by using
ECC you should be more sure of what happens.
>Another odd thing I noticed when I took out one of the memory modules - when both of
>the memory modules were in there, the BIOS would check the memory 3 times, but with
>only one modules it only checks it once. What does that mean.
No idea. I'm just an amateur here, remember :-)
since I read below that it also crashed with one module, it's starting
to look more and more (if it crashes with the other one too) like you
have a 'monday-morning' product with random glitches - I do hope you
have warranty....
Good luck,
Jurriaan
--
BOFH excuse #437:
crop circles in the corn shell
GNU/Linux 2.2.19pre7 SMP 2x1402 bogomips load av: 0.04 0.02 0.00
------------------------------
From: Matt O'Toole <[EMAIL PROTECTED]>
Subject: Re: Support for my sound card?
Reply-To: [EMAIL PROTECTED]
Date: Wed, 10 Jan 2001 20:53:49 GMT
Cubic Meter wrote:
> You should install SNDCONFIG from the Mandrake CDs. I have an ESS 1688 and
> it wasn't recognized by my box. I installed all the ALSA drivers and all
> that, and still nothing. Then I installed SNDCONFIG, ran it in a shell,
> and told it all it wanted to know, then blam-o! I had sound. It will ask
> you
> for IRQ, DMA and address settings so be prepared.
I'm way ahead of you. I've been using sndconfig for years. My card is
recognized OK, but there are no *free* drivers for it yet, from ALSA, OSS
or whatever. It appears the OSS commerical driver might work.
Matt O.
------------------------------
From: Tim Moore <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,alt.os.linux.caldera,comp.os.linux.setup
Subject: Re: Partitioning For Install
Date: Wed, 10 Jan 2001 21:06:04 GMT
Way too big. Try this for a start:
/ 25-50MB (bigger if >3 kernels)
swap 1-3 x memory (bigger if small memory)
/boot n/a (part of /)
/usr 2GB
/var 500MB
/opt 1GB
/tmp 500MB
/home 1-2GB
/spare 4GB
- Big partitions are more effort to manage than smaller partitions.
- Backups, file system checks, directory listings, file deletes and
restores from backup take much longer.
- Partitions with smaller cylinder numbers are faster than larger
cylinder numbers. The difference between the outside (low) and inside
(high) is about 50%:
# fdisk -l /dev/hdc
Device Boot Start End Blocks Id System
/dev/hdc5 902 1027 1012095 83 Linux
/dev/hdc7 14 652 5132736 83 Linux
# hdparm -tT /dev/hdc5
Timing buffer-cache reads: 128 MB in 1.20 seconds =106.67 MB/sec
Timing buffered disk reads: 64 MB in 6.53 seconds = 9.80 MB/sec
# hdparm -tT /dev/hdc7
Timing buffer-cache reads: 128 MB in 1.14 seconds =112.28 MB/sec
Timing buffered disk reads: 64 MB in 4.17 seconds = 15.35 MB/sec
- Use smaller partitions and symbolic links (ln(1)) to rearrange
structure:
# ls -l /opt
lrwxrwxrwx 1 root root 9 Oct 9 19:13 /opt ->
/big2/opt
- No need to partition the entire disk up front.
- Use a 2k or 4k block size for larger partitions (mke2fs(8)).
- Use a pair of smaller drives on separate controllers rather than a
single large drive if possible. System performance for swap,
application launch and data I/O is much better spread over separate
controllers, plus there's less I/O wait time.
Here's a working system based on 2x8GB UDMA-33 drives. Dual boots linux
and W95, runs Vmware, StarOffice, email, several tools for
mp3/CD-R/graphics manipulation, a robust RH 6.2 installation and several
different kernel trees. The /kits? areas are sources, the /big? areas
are audio processing and Vmware virtual disks:
# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/hda5 46636 35643 8585 81% /
/dev/hda6 2016016 1437020 476584 75% /usr
/dev/hda8 124427 17964 100039 15% /var
/dev/hda9 521748 604 494640 0% /tmp
/dev/hda10 995680 701244 284156 71% /home
/dev/hdc7 5112192 461496 4650696 9% /big
/dev/hdc6 1968372 1818172 150200 92% /big2
/dev/hda13 1009972 724188 285784 72% /kits1
/dev/hdc5 1010004 859944 150060 85% /kits2
# fdisk -l /dev/hd{a,c}
Disk /dev/hda: 255 heads, 63 sectors, 1027 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 * 1 62 497983+ 6 FAT16
/dev/hda2 63 1027 7751362+ 5 Extended
/dev/hda5 * 63 68 48163+ 83 Linux
/dev/hda6 69 323 2048256 83 Linux
/dev/hda7 324 334 88326 82 Linux swap
/dev/hda8 335 350 128488+ 83 Linux
/dev/hda9 351 416 530113+ 83 Linux
/dev/hda10 417 544 1028128+ 83 Linux
/dev/hda11 545 646 819283+ 6 FAT16
/dev/hda12 647 901 2048256 6 FAT16
/dev/hda13 902 1027 1012063+ 83 Linux
Disk /dev/hdc: 255 heads, 63 sectors, 1027 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hdc1 1 13 104391 82 Linux swap
/dev/hdc2 14 1027 8144955 5 Extended
/dev/hdc5 902 1027 1012095 83 Linux
/dev/hdc6 653 901 2000029+ 83 Linux
/dev/hdc7 14 652 5132736 83 Linux
# cat /etc/fstab
/dev/hda5 / ext2 defaults 1 1
/dev/hda6 /usr ext2 defaults 1 2
/dev/hda8 /var ext2 defaults 1 4
/dev/hda9 /tmp ext2 defaults 1 4
/dev/hda10 /home ext2 defaults 1 3
/dev/hdc7 /big ext2 defaults 1 2
/dev/hdc6 /big2 ext2 defaults 1 3
/dev/hda13 /kits1 ext2 defaults 1 5
/dev/hdc5 /kits2 ext2 defaults 1 4
none /proc proc defaults 0 0
/dev/hdc1 swap swap pri=1 0 0
/dev/hda7 swap swap pri=1 0 0
/dev/fd0 /floppy ext2 defaults,rw,user,noauto 0 0
/dev/cdrom /cdrom iso9660 defaults,ro,user,noauto 0 0
/dev/cdr /cdr iso9660 defaults,rw,user,noauto 0 0
/dev/sda1 /zip ext2 noauto,user,suid,defaults,rw 0 0
/dev/fd0 /A vfat noauto,user,defaults,rw 0 0
/dev/sda4 /Z vfat noauto,suid,user,defaults,rw 0 0
/dev/hda1 /C vfat noauto,user,defaults,rw 0 0
/dev/hda11 /E vfat noauto,suid,user,defaults,rw 0 0
/dev/hda12 /F vfat noauto,user,defaults,rw 0 0
> the results I take it there's no definitive answers. Generally, here's what
> was suggested for installing (Caldera) Linux on my 46 GB HD:
>
> /(root) 500MB-10GB
> Swap 128MB-1GB
> /boot 10MB-25MB
> /opt 2-3GB
> /home 2GB-As Big As You Can Make It
> /temp 6GB
> /usr Big
> /var ?
>
> Now another question comes to mind: When I install programs, are they
> going to "know" where these directories/partitions are, or am I gonna have
> to "show" them? The same question holds true for uninstalling programs.
> Where are the majority of programs going to install? Oh yeah, and I take it
They "know". /usr, /opt or /usr/local.
> Oh yeah, and I take it
> that a program only needs to be installed once, and not for each individual
> user? From what I've been reading, this point is somewhat unclear to me....
Correct. In some cases (eg- StarOffice, Netscape) there is a small
per-user component.
--
timothymoore
bigfoot
com
------------------------------
** 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
******************************
