Linux-Setup Digest #299, Volume #20 Thu, 28 Dec 00 18:13:08 EST
Contents:
Problems connecting to ISP(double login required)RH7 (jrobertson)
Re: where to find the detail LILO document. ("Cameron Kerr")
AWE 64 and Esound (esd) problem ("jason.pentlow")
ftape Trakker parport setup problem ([EMAIL PROTECTED])
Re: qmail relay (RC Walker)
No Sound in KDE2 (MrTaboo)
broken glibc & Redhat 7.0? (Peter Bismuti)
Re: How to make linux "sleep"? (Paul Oliver)
Verifying DMI Pool Data .... freeze (David Punsalan)
486-Floppy Disk Image Trouble (Cheese Dog)
Re: printing from netscape (David Efflandt)
Re: Problems connecting to ISP(double login required)RH7 (David Efflandt)
RH6.1/ATI AIW Rage 128/Dell GX1 ([EMAIL PROTECTED])
Solved, finally was Re: ARRG. Can't even see *one* drive, so won't install -- huh?
(Phil Edwards)
486, Linux and X-windows ("Ed Collins")
PPP Problem - need help urgently please!! (jpenner)
3Com 3c905 cards: here's how (Phil Edwards)
----------------------------------------------------------------------------
From: jrobertson <[EMAIL PROTECTED]>
Subject: Problems connecting to ISP(double login required)RH7
Date: Thu, 28 Dec 2000 19:33:10 GMT
I have used the Rh dial up interface to set up my existing ISP account.My
modem dials and hangs up repeatedly. I used the debugger to see what is
going on and I noticed that the isp requires that I log on twice. I
confirmed this with minicom by connecting manually. Wvdial doesn't know
what to do at the second prompt so it tries login as "ppp" and fails as
expected. How can I set up my account to connect automatically? I have
checked the PPP Howtos and faqs but I can't find the solution. Please
help!
------------------------------
From: "Cameron Kerr" <[EMAIL PROTECTED]>
Subject: Re: where to find the detail LILO document.
Date: Fri, 29 Dec 2000 08:50:49 +1300
You can find the lilo documentation in /usr/src/lilo-<version>
You'll need latex to compile the docs.
+-[ Cameron Kerr ]--------------------------+
|cameron.kerr @ paradise.net.nz |
|http://homepages.paradise.net.nz/cameronk/ |
|Computer Science Student / Systems Geek |
+-------------------------------------------+
------------------------------
From: "jason.pentlow" <[EMAIL PROTECTED]>
Subject: AWE 64 and Esound (esd) problem
Date: Thu, 28 Dec 2000 19:55:05 -0000
After setting up my AWE64 with sndconfig the test sounds play ok. When
enabling the sound sever in Gnome my PC locks up after attempting to play a
sound. Running /usr/bin/esd plays an intro tune and then my pc locks up.
I thought the problem could be with irq's or interrupts but having set the
configuration manually using different settings It makes no difference.
Any suggestions (It's now starting to get annoying)
Thanks
Jason
------------------------------
From: [EMAIL PROTECTED]
Subject: ftape Trakker parport setup problem
Date: Thu, 28 Dec 2000 19:47:24 GMT
Colleagues,
I recently attached a Colorado Trakker 250 tape unit to the parallel
port on my old Win95/Linux P130 box. I previously had an Iomega Ditto
2GB running there until it broke. Under Win95, the trakker drive is
seen at 0x378 IRQ 7 and operates correctly. Under Linux (kernel
2.0.twentysomething + ftape 4.04a) the drive is not found during
bootup, with dmesg complaints that the trakker module can't grab
0x378.
I've tried rebuilding the kernel without PRINTER support, with
PRINTER support, and with PRINTER support as a module without
affecting the outcome. I've also played around with modprobe
settings and /etc/conf.modules settings to force the 0x378 and
IRQ 7 directly instead of autoprobing. The latter silences the
dmesg complaints, creating later messages about drive select
failures with a -6 error (ENXIO).
The only thing that I can think of that has changed in my system
configuration since I had the Ditto 2GB parport drive is that
I added a sound card. The /proc information shows the sound
card coming in at a different base address and IRQ than the
printer port, as one would expect, so I am stumped.
Any ideas?
Sent via Deja.com
http://www.deja.com/
------------------------------
From: RC Walker <[EMAIL PROTECTED]>
Subject: Re: qmail relay
Date: Thu, 28 Dec 2000 12:07:29 -0800
David wrote:
> RC Walker wrote:
> >
> > I need to set up a linux server. The server is acting as a firewall,
> > also masquerading the clients behind it. I need to configure qmail to
> > all these clients to send messages to the outside world. With
> > 'firewall.server.com' as the out going SMTP server, I get an error
> > message stating that, 'the host (whatever account I am sending to), is
> > not found in rcpthosts'. Any hints on how to fix this?
> >
> > Sorry so long,
> > RCW
>
> Edit /var/qmail/control/rcpthosts and add the host name of the systems
> you want to relay mail for.
>
> localhost
> hostname1.domain.com
> hostname2.domain.org
This sounds like what I need. Or- since all the clients are behind the
firewall, wouldn't it work if I edit /var/qmail/control/rcpthosts and add
'.domain.com?
It seems to me that /var/qmail/control/rcpthosts acts as a filter stopping
any hosts not recognized that you are sending mail to. I will try this and
post my results.
Thanks,
RCW
>
>
> Then /etc/rc.d/init.d/qmail restart
>
> --
> Confucius say: He who play in root, eventually kill tree.
> Registered with the Linux Counter. http://counter.li.org
> ID # 123538
> Completed more W/U's than 98.912% of seti users. +/- 0.01%
------------------------------
From: MrTaboo <[EMAIL PROTECTED]>
Subject: No Sound in KDE2
Date: Thu, 28 Dec 2000 20:24:54 GMT
Basics:
OS = RH 7 -- 2.2.16-22 & 2.2.18 kernels
CPU = PII 400
RAM = 96MB
Sound Card = Diamond Sonic S90 (Aureal 3D (Vortex II engine))
I've downloaded and installed the Aureal Linux driver as per the
instructions. The installation was smooth and I now have sound in other
programs such as KNapster and also under Gnome.
The aRts sound server is enabled in the KDE Control Center.
None of the "personalization" sounds will play within KDE2 (e.g. startup,
shutdown). Anybody have any ideas on how to get this to work?
TIA
------------------------------
From: [EMAIL PROTECTED] (Peter Bismuti)
Subject: broken glibc & Redhat 7.0?
Date: 28 Dec 2000 20:19:41 GMT
I'm trying to install Blackdown JDK1.3 on Redhat 7.0 and it says that
the glibc that comes with it - glibc2.1.92 is broken and won't work.
Is it safe to download a new glibc and try to install it?
Also it says that the Java 2 SDK does not support running on SMP kernels.
Do they mean that it won't utilize multiple procecessors, or do they mean
that it won't work at all?!
Thanks
------------------------------
Date: Thu, 28 Dec 2000 14:25:48 -0600
From: Paul Oliver <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: How to make linux "sleep"?
[EMAIL PROTECTED] wrote:
>
> On newer keyboard, there are two buttons called "Sleep" and
> "Wake up" each. In M$Win environment, when I push "Sleep",
> the OS enters the "sleep" mode. (I don't know what the mode
> is, it turns off monitor, hd, all of the fans includes power
> supply.) It comes back immediately as soon as I push "Wake up".
>
> Are there any daemons or programs which could handle this task?
>
> --
>
> Jang-Ying Lee <[EMAIL PROTECTED]>
> Kaohsiung, Taiwan, R.O.C.
>
> Taiwan, a sovereign independent country,
> is NOT part of mainland China.
Yeah, look into apmd. I did a 'man apmd' and I saw some interesting
things:
it looks like there may be some commands: suspend, standby, etc.
If you don't have a man page installed see:
http://stio1.fh-wuerzburg.de/student/i510/man/apmd.html
--
___________________________________________________________________
In the year 2000 . . .
Former members of the musical groups "Sha Na Na" and "Bow Wow Wow"
will unite to form the supergroup, "Crap."
-- Conan O'Brien
------------------------------
From: David Punsalan <[EMAIL PROTECTED]>
Subject: Verifying DMI Pool Data .... freeze
Date: Thu, 28 Dec 2000 14:30:13 -0600
Hi,
I just tried setting up a dual boot system (Win98/RH6.1) on a computer
with a 13GB Western Digital HD. Before doing so... I made sure to
defragment the drive.
After repartitioning the drive, I installed RH6.1, rebooted and started up
linux. Everything was fine. So I reboot the system and decided to boot
up into Win98.
Then ... disaster... complete and total disaster.
The computer froze after saying:
Verifying DMI Pool Data ....
So I tried to boot from floppy - same thing.
I went and got a commercial rescue disk (from Ontrack) and it allowed me
to get a DOS command line. I got rid of LILO with fdisk /mbr and I aslo
did "sys c:". I thought this might fix the problem.
But things just got worse. The computer still froze at the DMI pool data
message.
Then I tried to boot into linux with the RH6.1 boot disk. It's just the
install disk actually - but I thought I might use it to re-install LILO
and some other rpm's. I've done this dozens of times before when
acidentally getting rid of LILO and not having a boot disk.
But even the linux install disk wouldn't boot.
Don't ask me what I was planning to do in linux to fix this problem. I
dunno.
I tried all the boot disks (win and lin) on other computers and they were
fine. Lastly - I got my hands on a disk utility from Western Digital and
it said that there is an unrecoverable CRC error.
I think this HD is toast. And furthermore - I think the repartitioning is
what did it. (fyi, I partitioned it at cylinder ~ 1100 - 7.7GB/4.3GB)
Oh yeah - I used fips20 that's available on the RH6.1 cdrom to do the
partition.
After doing an exhaustive search on the internet on this problem, I found
that others ran into this DMI problem after repartitioning their drives
yet noone knows why. There are a lot of tricks people suggest with the
BIOS, but none of them worked for me.
Any ideas, comments, suggestions? Are other partition utilities
(e.g Partition Magic) more robust agaist this type of problem?
As a linux tinkerer, until I hear convincing reasons otherwise - I think
dual booting is good for trying out linux as a newbie - but not good
practice in general. I mean ... I'd rather get an older computer to run
linux (because it can) and then run winXX on my more current machine -
rather than having to reboot one computer every other day. I mean the
entire concept of dual boot is kind of cool at first - but after a while -
it's kind of lame. And from this "DMI problem" - it has certainly left a
bad taste in my mouth.
- David
------------------------------
From: Cheese Dog <[EMAIL PROTECTED]>
Subject: 486-Floppy Disk Image Trouble
Date: Thu, 28 Dec 2000 12:53:25 -0800
I've got a room full of Macs networked together, but I have very little
experience with PCs, MSDOS, Windows, or Linux.
I was given a 486 with 8 mb RAM and a 330 mb HD.
It has a 3.5 inch floppy drive, and no CD-ROM drive.
Windows is not installed.
I wish to install Linux on the thing, but I have a minor problem.
I need to get a file called bootnet.img to it.
I've managed to get rawrite.exe there, but no matter what I try, the
file bootnet.img is larger than the available space on a floppy, even
though that was the point of the disk image.
At first I thought that the Macintosh's finder was putting a few items
on the disk, but I've formatted it in Windows 2000 via Virtual PC, and
in MSDOS with the 486 itself.
Is there a way to split the file?
Am I going to have to find a floppy with Windows 95 on it or something?
It's weird- I've got machines with Zip Drives, CD-RWs, and even DVD-RAM
Drives, but I'm being stopped for lack of an ol' floppy.
Any suggestions would be greatly appreciated.
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: printing from netscape
Date: Thu, 28 Dec 2000 21:09:42 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Thu, 28 Dec 2000, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>How do I print a page from netscape. When I hit the print button,
>adialog box comes up with the command 'lpr' on the top. When I hit O.K.
>the printer prints html code.
That is strange. I thought by default netscape prints postscript, and if
you have a postscript printer or print filter set up for that it should
print fine. To use any printer in your /etc/printcap other than the
default 'lp', you have to specify it with a -P switch (see 'man lpr').
--
David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Problems connecting to ISP(double login required)RH7
Date: Thu, 28 Dec 2000 21:13:28 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Thu, 28 Dec 2000 19:33:10 GMT, jrobertson <[EMAIL PROTECTED]> wrote:
>I have used the Rh dial up interface to set up my existing ISP account.My
>modem dials and hangs up repeatedly. I used the debugger to see what is
>going on and I noticed that the isp requires that I log on twice. I
>confirmed this with minicom by connecting manually. Wvdial doesn't know
>what to do at the second prompt so it tries login as "ppp" and fails as
>expected. How can I set up my account to connect automatically? I have
>checked the PPP Howtos and faqs but I can't find the solution. Please
>help!
If you cannot do a shell login even with minicom, maybe your ISP does not
allow a shell login, so it returns to the password prompt. Either set up
a connection using PAP authentication if your distro has a too for that,
or set up a connection that ends the chat script after CONNECT and put
your username and password in /etc/ppp/pap-secrets.
--
David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To:
linux.redhat.install,comp.os.linux.admin,comp.os.linux.hardware,comp.os.linux.questions
Subject: RH6.1/ATI AIW Rage 128/Dell GX1
Date: Thu, 28 Dec 2000 21:14:14 GMT
I installed RH6.1 on my Dell GX1 with on-board ATI Rage Pro and add-on
ATI AIW-128(Rage 128) under the text mode. I have also downloaded and
installed XFree86 v4.0.2 for glibc21. All of this was done using
text-mode on the ATI AIW-128 card. There is nothing coming out from the
Rage Pro port and I don't want to use it.
Xconfigurator always detect the on-board Rage Pro card(and use the
Mach64 driver) but not the AIW-128. I tried Xconfigurator --card r128
without any success(it reports some error and exit). Could someone
please help?
I am wondering whether the PCI probe at boot time detect the Rage Pro
before the AIW-128 and wrote it somewhere. And then Xconfigurator read
from this file?
TIA
Pok
------------------------------
From: [EMAIL PROTECTED] (Phil Edwards)
Subject: Solved, finally was Re: ARRG. Can't even see *one* drive, so won't install
-- huh?
Date: 28 Dec 2000 17:31:21 -0500
Okay, here's the solution. Everybody watch Deja archive it so that it
will be useful, and then lose it as they screw up the archives again...
Jason Byrne <[EMAIL PROTECTED]> wrote:
+ >
+ > I can walk right through the choices until the "what kind of installation
+ > do you want" screen, and then no matter what I choose (workstation,
+ server,
+ > custom), it prints "no valid devices were found on which to create new
+ > filesystems," and reboots.
+
+ hmmm... it's too bad that the install decides to reboot on its own ;-)
Yah, that was quite pissy. At least they leave a shell open on another
console until then.
(As a side note, RH7's GUI installer has a bug that /will/ crash towards
the end of the installation, period, if you aren't doing the "simple
workstation" install. How the fsck that made it out the door without
being caught is beyond belief. Even after I solved this drive problem,
I had to do it all over again in text mode.)
+ RedHat 'quirks' (polite version) aside... my guess is that you're already
+ passed the magical 1023 cylinder limit on your hard-drive with your Windows
+ installation... so you don't have a suitable place to begin your Linux
+ install. Also... I believe you are limited to booting from the first two
+ physical drives on the primary IDE controller... but this *might* also
+ depend on your hardware (I don't remember exactly)
[...]
+ If I'm right about first two hard-drives on the primary IDE controller
+ (/dev/hda or /dev/hdb) - you might need to move your hard-drive from
+ /dev/hdc -> /dev/hda or /dev/hdb anyway.
This turned out to be part of the problem; or at least similar to the
problem.
Since there wasn't any hard drive on the onboard controller (onboard had
the CDROM and the ZIP drive, and somewhere in there was the floppy),
the RH7 boot kernel died. It doesn't even look for drives on the extra
controller (a "promise" something according to /proc/pci).
(Before y'all flame me for my ignorance of hardware: I have no control over
what hardware we get here. That's okay, I don't really mind. I'm willing
to experience a little pain for a huge drive and a flat-panel monitor.
The only hardware I've had to deal with for years has been SCSI under
Solaris, and frankly, SCSI is way easier. :-) Ah well.)
Somebody saw the "80 gig" part of my post and pointed me here:
http://www.linuxdoc.org/HOWTO/mini/Ultra-DMA.html
That was it. Booted off the CD, paged through /proc/pci output, found
the numbers, did some hexidecimal math, booted again with "ide0=foo,bar"
since ide0 was completely unused (see the howto), and voila.
Typing those numbers was annoying while I was booting off floppy, but once
I taught the NT/Lose2000 boatloader about the Linux bootsector, the LILO
"append" statements solved that for me.
Changing the drives around didn't seem like an option; I have no other spare
drives or even spare cables here. Plus Lose2000 was already working and I
wasn't about to hork that up (in case I cannot convince the admins to send
mail to the Linux box, I'll be required to use Outlost/sExchange for email).
Much thanks to all those who posted or emailed me directly.
Phil
------------------------------
From: "Ed Collins" <[EMAIL PROTECTED]>
Subject: 486, Linux and X-windows
Date: Thu, 28 Dec 2000 22:20:32 GMT
Hello, there:
I'm in the process of installing Linux and X-windows on my 486 with 126K of
cache and 8 MBytes of RAM.
My question is, will Linux and X-windows work well? Should I get more
cache? More memory? Or should I purchase a MB from e-bay that will fit
inside a 486 case?
I would certainly appreciate a discussion on this topic.
ed collins
------------------------------
Date: Thu, 28 Dec 2000 14:46:16 -0800
From: jpenner <[EMAIL PROTECTED]>
Subject: PPP Problem - need help urgently please!!
I am having an annoying problem with PPP in my Mandrake 7.0. As root, I
can connect to my ISP and various other services, but none of my user
accounts are able to. I can start KPPP, and it appears to behave
normally, making the connection, and indicating that it has connected,
but once it is connected, my web browser can not access anything on the
web. I have set suid on my kppp, but to no avail. These are the
permissions currently on my /etc/ppp directory, and as far as I can tell
they seem correct:
-rw------- 1 root daemon 78 Dec 21 1999 chap-secrets
-rw-r--r-- 1 root root 0 Nov 11 1999 ioptions
-rwxr-xr-x 1 root root 265 Sep 16 1997 ip-down*
-rwxr-xr-x 1 root root 406 Sep 6 1999 ip-up*
-rw-r--r-- 1 root daemon 5 Dec 21 1999 options
-rw------- 1 root root 123 Dec 18 21:04 pap-secrets
drw------- 2 root daemon 4096 Dec 21 1999 peers/
-rw-r--r-- 1 root root 47 Dec 18 21:19 resolv.conf
My KPPP is set to: -rwsr-xr-x 1 root root 523380 Dec 18
13:02 /usr/bin/kppp
The PPP files from /usr/sbin seem fine, although I am not sure if they
are actually required in a dial-up modem connection. They are:
-rwx------ 1 root root 150362 Nov 11 1999 /sbin/ipppd*
-rwxr-xr-x 1 root root 18117 Nov 11 1999 /sbin/ipppstats*
-rwxr-xr-x 1 root root 48124 Jan 11 2000 /sbin/ppp-watch*
I have read the "ppp-HOWTO" and have not found the answer to this
problem. If anyone could give me a hand with this, I would be very
grateful. Responses to this NG, or to "[EMAIL PROTECTED]" would be
most appreciated. Thanks in advance, John
------------------------------
From: [EMAIL PROTECTED] (Phil Edwards)
Crossposted-To: comp.os.linux.hardware
Subject: 3Com 3c905 cards: here's how
Date: 28 Dec 2000 17:58:07 -0500
While setting up my new Linux box (thanks again to those of you who tipped
me to the ultra-dma thing), I went a few rounds with the network card.
And I won. Saw a lot of posts from people who were having problems and
gave up. Here's something you might try if you don't mind a little cheap
kernel makefile messing.
Basically, when booting the RH7 kernel, 2.2.16-22, the card worked.
After grabbing the 2.2.18 release and building, it didn't. The 905 series
(in my case, a "Tornado") shows up in menuconfig as the same choice as
the 59x. Unfortunately, no 3c90x.o file was ever being built, hence,
no module support nor builtin support.
You can learn a lot at http://www.scyld.com/network/updates.html and pages
linked from there. As it happens, I never used anything from there except
the explanations on the pages themselves. Maybe a newer driver is there.
Maybe Red Hat added this driver to 2.2.16 for their release. I don't know;
I don't have the pure .16 sources and am not going to download them just
to satisfy my curiosity. :-) Anyhow, with the 2.2.16-22 sources from
RH7 and the pure 2.2.18 sources sitting side-by-side,
- From .16's drivers/net, copy 3c90x.[hc] into drivers/net in the
.18 sources. Maybe they're stright from 3Com, maybe RH wrote them
from scratch; I didn't look and don't care. The header size is
49928 bytes, the source is 170567 bytes, and their m5sum's are
ad376e94415fcc568f7a4a51962f7732 3c90x.c
eae6351d10b480ef004eb8b371fb856c 3c90x.h
so you can see if you have the same files. (Please don't email me
asking for them; I don't read this account often. Surely someone
with access to a web or ftp server can put them up.)
- Edit .18's drivers/net/Makefile and find the section for the
CONFIG_VORTEX. For both L_OBJS and M_OBJS, add the 90x object
file to the list, like
n_OBJS += 3c59x.o 3c90x.o
^^^^^^^ add this
Even if you're not using the module, go ahead and add it to the
M_ list, just in case you change your mind in the future.
- make *config and be sure to add the Vortex support in networking
cards, then "make dep clean bzImage blah blah" you know the deal.
Works like a champ. I have no idea what kind of throughput I'm getting
because I don't recall how to measure that under Linux.
Luck++;
Phil
------------------------------
** 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.setup.
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-Setup Digest
******************************
