Linux-Setup Digest #196, Volume #21 Wed, 9 May 01 17:13:07 EDT
Contents:
Wrong video mode, how can I get back into my system? (Leon Stringer)
Re: r128_do_wait_for_fifo failed in kernel 2.4.2 (J Hayward)
Re: Recompiled kernel nearly twice as fast, but no sound! (Steve Martin)
Re: user mount ("Brian Wildasinn")
Re: Red Hat Linux 7 With RCA (AT&T@Home) Cable Modem (Leonard Evens)
Re: Wrong video mode, how can I get back into my system? ("Brian Wildasinn")
Re: Question: One IP Address w/ Two Servers (Steve Martin)
Re: LILO no boot .. says "LIL-" then just hangs there ("Brian Wildasinn")
Re: kernel 2.2.16->19 nfs/rpc error (Silvio Luis Leite Santana)
help with dual boot, dual drive setup? (Kuenley Chiu)
Re: /boot/message - howto create (David Efflandt)
Re: Adding DNS causes slow telnet (David Efflandt)
Re: bastian FW/Masquerade (Abbadon)
Re: Mandrake 8 freezes during install (NEWBIE) (Richard Kimber)
----------------------------------------------------------------------------
From: Leon Stringer <[EMAIL PROTECTED]>
Subject: Wrong video mode, how can I get back into my system?
Date: Wed, 09 May 2001 21:15:18 +0100
Hi,
I plugged another monitor into my system. It detected the change and
unsuccessfully changed settings. Now I've got a system which just shows
a flickering image (as if the scan rate is out). The problem is I can't
find out how to recover from this. My system loads X by default
(runlevel 5?) and once this has loaded I can't access any consoles (i.e.
Ctrl+Alt+F1-6 does nothing). When the system is loading it displays
correctly but I can't interrupt this (Interactive Startup seems to be no
help). So all I can do with my system is bring it up and press
Ctrl+Alt+Del to bring it down. It's not networked so I can't telnet to
it.
My system is RedHat 6.1 and I've tried re-running the upgrade to no
avail. Presumably a reinstall will kill everything?
I haven't got a rescue disk. Is there any hope for me? I'm just asking
for a way to get a console/text session so I can reconfigure X.
Thanks in advance,
Leon Stringer
------------------------------
From: J Hayward <[EMAIL PROTECTED]>
Subject: Re: r128_do_wait_for_fifo failed in kernel 2.4.2
Crossposted-To: comp.os.linux.misc
Date: Wed, 9 May 2001 13:21:35 -0700
Hello,
Read this thread from the seawolf mailing list, it has your answer:
On Wed, 9 May 2001, John Horne wrote:
>> r128_do_wait_for_idle failed!
>
>> May 7 05:31:22 seawolf kernel: [drm:r128_do_wait_for_fifo] *ERROR*
>> r128_do_wait_for_fifo failed!
>>
>but I do see this one :-) I have an ati rage 128 (16mb), but as you say it
>all seems to be working fine. I use X window pretty much all the time, and
>I've had no problem.
> Both errors do show up. This is an often reported bug. The fix
> includes a kernel patch for DRM, as well as a XFree86 patch for
> the DRI code.
>
> The single unified patch got into 4.0.3-5 just in time, but
> unfortunately the kernel portion got applied to the kernel DRM
> code included in XFree86 instead of the kernel code in our kernel
> RPM. This means that the actual r128 kernel module as released
> does not have the needed kernel side fix.
>
> I just noticed this the other day after wondering why the heck I
> was still getting reports, and to make a long story short, our
> kernel now has the patch applied (internally), and will make it
> into rawhide shortly. If there are any kernel errata released
> (most probably at some point), the errata kernel will also
> contain the fix. If someone wants to manually patch up their own
> kernel to do it, you can obtain the r128 fixup patch from the
> XFree86 src.rpm, edit it and remove the non-kernel portion from
> it, then apply the remainder to the 2.4.x kernel source and
> rebuild. I can't help out with this though, so just wait for our
> next kernel release if its too heavy to attempt.
>
> With the kernel patch applied, r128 3D is rock solid on all my
> tests.
>
> I decided to put together the following to make things a bit
> easier (hopefully) for those who feel like playing with it:
>
> ftp://people.redhat.com/mharris/patches/linux-r128-drm.patch.txt
> ftp://people.redhat.com/mharris/patches/linux-r128-drm.patch.gz
>
> Hope someone benefits from it.
>
>
> ----------------------------------------------------------------------
> Mike A. Harris Shipping/mailing address:
> OS Systems Engineer 190 Pittsburgh Ave., Sault Ste. Marie,
> Red Hat Inc. Ontario, Canada, P6C 5B3
> http://www.redhat.com Phone: (705)949-2136
> ----------------------------------------------------------------------
> Latest XFree86 test RPMS: ftp://people.redhat.com/mharris/testing
Regards,
Jim H
------------------------------
From: Steve Martin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Recompiled kernel nearly twice as fast, but no sound!
Date: Wed, 09 May 2001 12:28:00 -0400
[EMAIL PROTECTED] wrote:
>
> Joel <[EMAIL PROTECTED]> wrote:
> > No ISAPnP Sound Blaster found, trying standard IRQ, IO, and DMA.
> > IRQ, IO, DMA required.
> > Those aren't the exact words, but something similar. How can I tell the
> > kernel what irq, etc to use? Will I need to recompile the kernel?
>
> If you have compiled the SB like a module you should be able to use
> modprobe to pass to it the required parameter. See the documentation
> in /usr/src/linux/Documentation.
If the sound card is an ISA-based PNP, you might want to take a look
at "man isapnp" and "man pnpdump". You can set the desired IRQ, DMA,
et cetera on the card with these commands, then build the driver as
a module and put the correct parameters in /etc/modules.conf, and it
should find the card.
------------------------------
From: "Brian Wildasinn" <[EMAIL PROTECTED]>
Subject: Re: user mount
Date: Wed, 9 May 2001 12:56:56 -0700
This /etc/fstab works for me:
/dev/hda1 / ext2 defaults
1 1
/dev/cdrom /mnt/cdrom iso9660 noauto,owner,ro 0 0
/dev/fd0 /mnt/floppy_msdos msdos noauto,owner,rw 0 0
/dev/fd0 /mnt/floppy_unix ext2 noauto,owner,rw 0 0
/proc /proc proc defaults 0 0
Make sure to make a mount point such as `mkdir /mnt/cdrom && chmod ugo=+rwx
/mnt/cdrom` . The directory must be readable by everyone.
Then you should be able to run `mount /mnt/cdrom && cd /mnt/cdrom` to see
the directory contents. If you use Gnome you can add a panel app that looks
like a little floppy drive icon for one-click mounting. It also includes a
"browse" menu.
Hope that helps,
Brian
"Martin Greco" <[EMAIL PROTECTED]> wrote in message
news:9dc4v3$8i1$[EMAIL PROTECTED]...
> Hi.
>
> I tried to set the "user" swith in the fstab file in some devices as said
in
> the mount man pages, but I still can't mount them with user accounts.
>
> any suggestions?
>
> --
> -
> Martin Greco
> Soporte Residencial - int. 2001
> Sinectis Rosario - Rioja 1659
> Tel: 0810-999-0700
>
>
>
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.misc
Subject: Re: Red Hat Linux 7 With RCA (AT&T@Home) Cable Modem
Date: Wed, 09 May 2001 15:04:11 -0500
Wouldnt You Like to Know wrote:
>
> I recently added a Red Hat Linux partition to my Windows 98 2nd
> Edition P.C. For Windows 98, I am using the AT&T @Home Cable Modem
> Service using an RCA digital / broadband cable modem. I am also using
> a Netgear RT314 Router.
>
> My NIC card is a SMC EZ Connect USB.
>
> I have seen several somewhat condensed articles on getting AT&T@Home
> cable modems to run with Red Hat Linux, but nothing has been very
> comprehensive. It also seems the configuration may be different based
> on the geography you are in. I happen to be in the Chicagoland
> (western suburbs) area.
>
> Has anybody seen or can point me to a document / article which
> describes in detail how to set a AT&T@Home Cable Modem with my Red Hat
> Linux 7 system ? If so, please point me the way !! What makes this
> worse is my Windows 98 modem is only Windows compatible and won't work
> on my Linux system.
>
> Any thoughts / comments appreciated and thanks in advance.
>
> My email address is: [EMAIL PROTECTED]
>
> Steve
AT&T has been using different modems and different network
cards in different areas around Chicago and at different times.
I was an early enrollee, and I have a Motorola Cable Modem
and some standard PCI NIC. It works fine under RH Linux.
For your setup, you will need to use a version of Linux which
supports USB modems. RH 7.1 with the 2.4 kernel should work,
but some later 2.2 kernels may also work. There is a web page
somewhere which explains what to do, but I can't find it.
I think I found it originally by checking on USB for Linux.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: "Brian Wildasinn" <[EMAIL PROTECTED]>
Subject: Re: Wrong video mode, how can I get back into my system?
Date: Wed, 9 May 2001 12:57:32 -0700
If you are using LILO, reboot and press tab. At the lilo prompt type,
linux single ro <enter>.
That should get you to single user mode. You can leave off the ro, if you
have a clean filesystem (else after fscking, remount root such as `mount -o
remount /dev/hd1`).
If you want to use this other monitor you will need to configure a new
XF86Config file, using either /usr/X11R6/bin/Xconfigurator, xf86config, or
XF86Setup, or by hand. Both Xconfigurator and XF86Setup run a diagnostic so
you can get the best view possible for your video hardware.
Hope that helps,
Brian
"Leon Stringer" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hi,
>
> I plugged another monitor into my system. It detected the change and
------------------------------
From: Steve Martin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Question: One IP Address w/ Two Servers
Date: Wed, 09 May 2001 12:32:19 -0400
If what you're wanting to do is to use two computers on a private
network and have them access the Internet using a single IP
address, then yes, this can be done, although it's an outgoing-
only proposition. The process is called NAT, and can be done with
hardware (3Com makes a box that works nicely), or with software
(see the IP-Masquerading HOWTO).
If what you want to do is to assign the same IP address to the
network interfaces on two different computers, this is absolutely
a no-no.
Chuck Coker wrote:
>
> I am getting a single static IP address (I have a dynamic IP address now).
> Having the static IP address will work for a single Linux box ~or~ a single
> Windows NT Server box. What I am wondering is if there is some way to have
> both servers use the single static IP address. The static IP address will go
> to a Cisco router and then to my local network.
>
> For example, what I want to do is have the Linux box handle the domains
> abc.com and def.com: www.abc.com, smtp.abc.com, www.def.com, smtp.def.com,
> etc., and have the WinNT box handle the domain xyz.com: www.xyz.com,
> smtp.xyz.com, etc. What I would ~really~ like would be to have www.abc.com
> on the Linux box and www1.abc.com on the WinNT box.
>
> I know how to create sub-domains and set up services on both systems, but
> I'm not sure how to use both servers simultaneously. Any hints, suggestions,
> directions, or pointers to information would be greatly appreciated.
>
> This is a learn-as-I-go project and I get stuck occasionally. :^(
>
> Thanks,
> Chuck Coker
> [EMAIL PROTECTED]
------------------------------
From: "Brian Wildasinn" <[EMAIL PROTECTED]>
Subject: Re: LILO no boot .. says "LIL-" then just hangs there
Date: Wed, 9 May 2001 13:07:51 -0700
Make a boot floppy first, if you don't have one. Then rerun lilo and reboot.
For example `cd /etc && lilo -v -C lilo.conf && shutdown -r now`. Assuming
you /etc/lilo.conf file hasn't been changed, else you will need to set this
up. In Slackware there's an option in setup to configure both lilo and boot
floppies.
Hope that helps.
Brian
"Counts" <[EMAIL PROTECTED]> wrote in message
news:9dc3ut$[EMAIL PROTECTED]...
> I installed everything successfully (Slackware 4.1), recompiled a kernel,
> rebooted, everything worked as expected. Then, not long afterwards on
> another reboot, LILO just hangs. I have to boot using a floppy now. Any
> suggestions on how to fix this?
>
> Should I uninstall LILO and then re-install?
>
> Thanks for any suggestions.
>
> Bob
>
>
------------------------------
From: Silvio Luis Leite Santana <[EMAIL PROTECTED]>
Subject: Re: kernel 2.2.16->19 nfs/rpc error
Date: Wed, 09 May 2001 17:48:50 -0300
Hi
Yes, I did.
[silvio@tico silvio]$ rpm -q mount
mount-2.11b-3
[silvio@tico silvio]$ rpm -q nfs-utils
nfs-utils-0.3.1-7
thanks, anyway
Silvio
Rex Dieter wrote:
>
> > I use Red Hat 7.0 and
> > I have just done the upgrade of the kernel,
> > from 2.2.16 to 2.2.19 in one of our machines.
> > Everything went fine, but with the new kernel,
>
> Did you update mount and/or nfs-utils too? (hint: you should).
>
> --
> Rex A. Dieter [EMAIL PROTECTED]
> Computer System Administrator http://www.math.unl.edu/~rdieter/
> Mathematics and Statistics
> University of Nebraska Lincoln
------------------------------
From: [EMAIL PROTECTED] (Kuenley Chiu)
Subject: help with dual boot, dual drive setup?
Date: 9 May 2001 20:15:38 GMT
Hi, can anyone offer some advice about this problem:
I just installed a second hard drive, with linux on it. My first hard drive
was windows 98, and they're both IDE. I set the linux drive to be primary and
the windows drive to be slave.
I was hoping to set up lilo so that I could choose either one to boot from
-- but now, when I do "df" in linux, I only see the linux drive. The windows
boot can only see its drive also (actually it sees a new hard drive, but can't
open it).
Any advice or comments on what I need to do?
Thanks!
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: /boot/message - howto create
Date: Wed, 9 May 2001 20:45:35 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Tue, 8 May 2001 09:08:22 +0200, Eric <[EMAIL PROTECTED]> wrote:
>> When SuSE 7.1 was installed, the /boot/message created is 65688; the lilo
>> version is 21.6.
>>
>> After updating lilo to 21.7.x, the error msg on running lilo is that
>> the /boot/message file is greater than 65535 (as per the man lilo.conf).
>> I now have reverted back to 21.6 (suse's version?) so that the new
>> kernel(s) may be created and "added" to lilo.conf.
>>
>> Now, how do I create/change another "message" and/or what app can I use
>> to "see" the /boot/message file?????
>
> IIRC it's just an .png graphics file (I can't check, I don't have one
> present here)
> Try gimp/xv/imagemagick.
> (There was a gimp-plugin on freshmeat last week I believe for creating these
> files too)
>
> Eric
It is the background image you see when you boot. The colors look strange
in gimp. xv displays it properly and says it is 640x480 PCX, 8-bit mode,
127 color (126 unique).
--
David Efflandt (Reply-To is valid) 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: Adding DNS causes slow telnet
Date: Wed, 9 May 2001 20:51:11 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On 9 May 2001 12:52:13 -0500, charadeur <[EMAIL PROTECTED]> wrote:
> When I add enteries for my DNS servers I get a slow connection to a
> telent session into the box. It even slows down pop3 and smtp and
> times out tring to connect. If I remove the DNS server entries it is
> fast. The OS is caldera. Any ideas?
Caching nameserver with forward and reverse zones added per DNS-HOWTO, and
point your boxes at its LAN IP for DNS (except nameserver 127.0.0.1 on DNS
box itself). I noticed this when I deactivated DNS on a box that is no
longer connected directly to the internet. Just make sure you tell it to
only 'listen-on' private IP ranges, to minimize exposure to bind cracks.
--
David Efflandt (Reply-To is valid) 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/
------------------------------
Crossposted-To: comp.os.linux.security,comp.os.linux.networking
From: [EMAIL PROTECTED] (Abbadon)
Subject: Re: bastian FW/Masquerade
Date: Wed, 09 May 2001 20:58:04 GMT
Trevor <[EMAIL PROTECTED]> wrote:
> O.K. - so I've searched several archives, hit a few books (including
> 'Linux
> Firewalls' and gone through the various HOWTO's. Nobody really
> discusses
> what I'm trying to do, and it seams like it should be a common thing.
> Here's the scoop:
> I'm trying to setup a bastian firewall that masquerades my LAN, but
> still gives access to a web server. The LAN only consists of 3
> windoze boxes which need to be masq'd and the web server needs to
> keep the PUBLIC IP address. -Setup looks somethin' like this:
>
Why does the web server need a public address? Just forward all
queries to port 80 to the internal ip address of the webserver.
I have a bastion firewall that does exactly this.
>
> __________
>
>| Masq'd IP's |
>
> /--------------
> _______ ________ _____/
> INTERNET---| DSL rtr |---| Firewall |------| Hub |
> --------- ---------- ------\
>
> \__________
>
>| Webserver |
>
> -----------
>
>
> The concept seams easy enough but I have a few Questions
>
> 1) Do I need a third NIC on the firewall to segregate the web server
> and take it off the hub?
No, not unless you want to set up a DMZ.
> or,
>
> 2) Can I just take care of it by forwarding packets destined to the
> public IP address while still masquerading packets from the LAN?
> -How do I set up the ipchains? (yes, still on the 2.2.16 kernel)
This is confusing. If you mean what I think you mean, yes.
# This is for forwarding http and https traffic
/sbin/ipchains -A forward -s 192.158.1.14/32 -j MASQ
/usr/sbin/ipmasqadm portfw -a -P tcp -L 24.5.39.134 80 -R 192.168.1.14 80
/usr/sbin/ipmasqadm portfw -a -P tcp -L 24.5.39.134 443 -R 192.168.1.14 443
> 3) I'm having a hard time conceptualizing the gateways. Do I need to
> set the gateway for the LAN and web server to point to eth1
> (internal)
> and then set the eth1 gateway to point to eth0 (external) ??
Yes. Have the machines behind the bastition point to the bastition as
thier gateway. Have the bastition point to whatever your $GATEWAY is
if you only had a single machine.
> Any help would sure be appreciated. (and please don't tell me what a
> dumbass I am for not knowing this stuff, I'm still a bit of a newbie,
> but
> I AM trying to do my homework first. ;-)
>
> Thanks!
>
> ~ Trevor ~
>
> -Oh! One other thing - if I ?route add? a default gateway on the
> firewall to point to the DSL gateway (designated by the phone co.) -
> takes almost 5 minutes to return the full kernel routing table. -Why
> the major delay? ...hmmmm... curious :-p
>
I would look at dns first. Try doing a route -n.
Abbadon
--
Madness and genius are separated only by degrees of success.
-- Sidar Jabari
------------------------------
From: Richard Kimber <[EMAIL PROTECTED]>
Subject: Re: Mandrake 8 freezes during install (NEWBIE)
Date: Wed, 9 May 2001 22:05:48 +0100
Kwan Lowe wrote:
> Cold Fusion <[EMAIL PROTECTED]> wrote:
> > Trying to install Mandrake 8 on a separate, clean drive. The install
> > always
> > locks up at the 30% mark, regardless of which package. Tried graph,
> > low-res graph, AND text install, same thing. HELP!
>
>
> Try looking at the log screens (press ALT + Function Keys). It sounds like
> your installation CD may have problems. I recently had a similar problem
> that was caused by bad memory.
>
> BTW, the DVD ROM itself is not the problem. I install from them regularly.
I also had problems. It stopped just after I had made my package choices
and complained about an 'ldconfig' problem and then said there was no
'hdlist'. It finally suggested that there was memory missing, but I did a
fairly thorough test (> 8 hours) using memtest 2.5 and it found no errors.
Also, Mandrake 7.2 (admittedly on a different disk) and Debian Potato both
install fine. Windows 2000 works OK from the same disk. And Partition
Magic found no errors in any partition on the disk.
I installed from CD iso images that I downloaded, and the checksums were OK.
This suggests a buggy install programme to me. It certainly doesn't
inspire confidence when the onscreen help says you can choose from three
options, but only 'normal' and 'expert' actually appear as options that can
be chosen. My feeling is that 8.0 is too flakey to be used yet.
- Richard
--
Richard Kimber
[EMAIL PROTECTED]
Political Science resources http://www.psr.keele.ac.uk/
British Election 2001 http://www.psr.keele.ac.uk/area/uk/ge01.htm
------------------------------
** 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
******************************
