Linux-Setup Digest #37, Volume #19 Thu, 29 Jun 00 17:13:11 EDT
Contents:
Re: Must have Domain name? ("John D. Goulden")
Re: "screen" problems while using a vt320 (Pete Zaitcev)
Re: Dual Boot - 2 Drives (DeAnn Iwan)
Re: gtk problems in compiling ("ZephyrQ")
A kernel build question WRT SHMMAX. (Steve Emmett)
Requirements for upcoming 2.4 kernel ("Jesus M. Salvo Jr.")
Re: Can't Boot from Compiled Kernel (2.2.13) (John Gluck)
Re: Netscape font is too small ([EMAIL PROTECTED])
Re: Strange Lilo "LI" problems (John Gluck)
Re: DSL question in Arizona... ("Mark VanBogart")
Re: rebuilt kernel doesn't autoload modules ([EMAIL PROTECTED])
new to linux (sylvain hutchison)
Re: DSL question in Arizona... (Tom Williams)
Re: kernel (Tom Williams)
network question (sylvain hutchison)
gcc compiler (sylvain hutchison)
Re: i810 video card driver support in xfree86 4.0 ("Dave Breece")
Re: network question (Jim Harback)
Re: network question (Mikey)
Re: "screen" problems while using a vt320 (Matthew Miller)
----------------------------------------------------------------------------
From: "John D. Goulden" <[EMAIL PROTECTED]>
Subject: Re: Must have Domain name?
Date: Thu, 29 Jun 2000 12:20:09 -0500
OK, let me jump in here and ask if I did this right (linux newbie). My linux
box is on my home ethernet lan; a Win98 machine has the modem and runs as
the proxy server. I haven't done much with networking yet but once I
configured Netscape to use a proxy server it has had no problem getting on
the 'net. The question is, when asked for the domain name I gave it the name
of the Win98 box (say, 'Win98'). netconf shows me that 'Win98' is associated
with the IP address of the Win98 box, not the linux box, so it seems OK.
However, whenever I log in to the linux box my prompt says [username@Win98].
Is this the way things should be? Just a curious newbie...
--
Please reply by email as well as to the group.
John D. Goulden
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Pete Zaitcev)
Crossposted-To: comp.os.linux.questions,comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: "screen" problems while using a vt320
Date: Thu, 29 Jun 2000 18:24:10 GMT
> >Also make sure that you have x-on/x-off set right in the terminal setup.
>
> Hum, the comm. setup with the terminal right now is that xoff is off. I also
> tested the settings with xoff set to 64 and 128. No change.
My is at 128.
> [...] I'm almost desperate enough to recompile screen with debugging
> symbols and sic gdb on it! But, I'm not to that point yet. It is really
> bothersome to get no output from screen or error messages :(
You are an oldtmer too, I see :) Before going the gdb route, try
running screen under strace, something like "strace -o /tmp/xxx screen".
--Pete
------------------------------
From: DeAnn Iwan <[EMAIL PROTECTED]>
Subject: Re: Dual Boot - 2 Drives
Date: Thu, 29 Jun 2000 14:30:21 -0400
There are several multiboot howtos that are part of the Linux
Documentation Project. A version of these is likely included on CDROMs
for your distribution, but the latest updates are on the net.
Michael Garabaldi wrote:
>
> Thanks for the info! Both hard drives are on the primary IDE. I have a
> DVD/CD player as master on the secondary IDE and a Travan Tape Drive
> as a slave on the secondary IDE. Are the Dual Boot gotchas in a FAQ
> somewhere or must I continue to scan this news group? Also are there
> any other news groups that I should be looking at pertaining to this
> topic?
>
> On Thu, 29 Jun 2000 12:05:45 -0400, DeAnn Iwan <[EMAIL PROTECTED]>
> wrote:
>
> > Yes. However, some BIOS will only see the first 2 IDE drives. If you
> >have a cdrom installed as slave on the first controller (hdb), they may
> >not recognize the second hard drive (on hdc) as a bootable drive. Also,
> >Win9x will seek forever if it sees the second drive but does not find
> >something of its own on it (or an empty drive). {will someone PLEASE
> >compile a patch into the kernel and submit it to Bill for inclusion in
> >the main tree!!!!) The solution here is to put a tiny FAT partition on
> >the second drive. Read the dual boot howtos for other gotchas.
> >
> >Michael Garabaldi wrote:
> >>
> >> Hi All,
> >>
> >> I've seen [EMAIL PROTECTED] post this question in this group, but
> >> didn't see a valid answer to it so I figured I would ask. I have a
> >> computer with two hard drives. Drive C is a WD 30 Gig drive and drive
> >> D is a 8.4 Gig IBM DeskStar. Is it possible for me to install the
> >> newest version of RH on drive D (keeping Win 95 on drive C) and still
> >> be able to dual boot? If more information is needed, just ask, I'll be
> >> glad to answer.
------------------------------
From: "ZephyrQ" <[EMAIL PROTECTED]>
Subject: Re: gtk problems in compiling
Date: Thu, 29 Jun 2000 18:50:19 GMT
How do I do this? I *have* dled and installed the corresponding -dev package,
and still get the error.
In article <8jfs6o$eq3$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (Colin
Watson) wrote:
> ZephyrQ <[EMAIL PROTECTED]> wrote:
>> checking for GTK - version >= 1.2.6... *** GTK+ header files
>> (version
>>1.2.7) do not match
>> *** library (version 1.2.8)
>> It mentions that the header files are not matching...is there a
>>way to edit these files?
>
> They're human-readable (as is more or less everything in
> /usr/include), but just changing the version number is a bad idea.
> According to the error message above, you have version 1.2.8 of the
> library itself and version 1.2.7 of the corresponding -dev or -devel
> package; you should bring these into sync properly.
------------------------------
From: Steve Emmett <[EMAIL PROTECTED]>
Subject: A kernel build question WRT SHMMAX.
Date: Thu, 29 Jun 2000 13:50:05 -0500
I've compiled a new 2.2.14 kernel with a couple of changes to shared
memory and semiphore parameters. In particular, I've changed the value
of SHMMAX in /usr/src/linux/include/asm/shmparam.h from the default
value of 0x2000000 to 0x8000000 (this because I have 256M of RAM). I've
also changed the values of SEMMSL and SEMOPM in
/usr/src/linux/include/linux/sem.h.
Prior to installing the new kernel, ipcs -l shows SHMMAX as 16M, as well
as the default values for SEMMSL and SEMOPM.
After booting with the new kernel, ipcs -l shows SHMMAX still as16M but
shows the new values for SEMMSL and SEMOPM.
Have I missed something in the compile process?
--
Steve
=========================================
Steve Emmett
=========================================
"A mind that is stretched to a new idea
never returns to its original dimension"
=========================================
------------------------------
From: "Jesus M. Salvo Jr." <[EMAIL PROTECTED]>
Subject: Requirements for upcoming 2.4 kernel
Date: Thu, 29 Jun 2000 19:08:19 +0000
I would like to prepare for the new 2.4 kernel. What are the
requirements?
glibc release?
libc relase?
ldd release?
etc....
John Salvo
--
Homepage: http://homepages.tig.com.au/~jmsalvo/
------------------------------
From: John Gluck <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.slackware,comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: Can't Boot from Compiled Kernel (2.2.13)
Date: Thu, 29 Jun 2000 15:00:32 -0400
mv wrote:
> I have been battling a wierd boot problem that has me stumped and wondered
> if anyone has a solution for this problem.
>
> The problem is that I can't boot from any kernel that I compile on the PC.
> However, I
> can boot from the IBMMCA.S kernel that came on the Slackware 7.0
> distribution.
>
> The symptoms I am experiencing are that the loader starts displaying the
> following line and then stops at a blank line below this when I select one
> of my
> compiled kernels at boot time:
>
> loading linux................................
>
> If I pick the IBMMCA.S kernel included at installation time, it boots every
> time.
[snip]
You have most likely compiled something as a module that need to be compiled in the
kernel.
Mostly likely, filesystem, hard-disk driver, mca support,or some unusual device.
--
John Gluck (Passport Kernel Design Group)
(613) 765-8392 ESN 395-8392
Unless otherwise stated, any opinions expressed here are strictly my own
and do not reflect any official position of Nortel Networks.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Netscape font is too small
Date: Thu, 29 Jun 2000 11:33:04 -0700
Following was received as email. Response redirected to list.
On Thu, Jun 29, 2000 at 09:46:50AM -0400, Fred Nastos wrote:
> Hi,
>
> I'd like to continue this dialogue, but I thought I'd get it
> off the newsgroup, since people may not be interested.
>
> By all
> means, feel free to post this and your reply if you think
> people may benefit from it.
I'm going to bounce this back to the newsgroup as it may be useful to
those who aren't familiar with X resources. This is in general my
preference.
I'd ***FAR*** prefer you'd not modify subject lines or drop usenet
references as it makes it more difficult to find the appropriate thread
and group.
I am among those who consider it bordering on rude to take relevant
Usenet discussion offline. I'm not answering *your* question, I'm
providing a resource (through the groups and indexes) for others with a
similar problem.
> You wrote:
>
> My fix is to modify Netscape resources with the attached resource file.
> It's commented, and fixes a number of bugs with Netscape including the
> shopping button.
>
> I wrote:
>
> Is it possible to place this in a user's home directory instead of
> somewhere global? Would it go in .Xdefaults?
>
> And you responded with:
>
> Yes. You can put it where ever you want, so long as it's included by
> whatever collects and sets your X resources.
> $man xrdb
> ------------------------------------------------------------------
>
> Ok. I'd like to understand better how this thing works, and I
> really want to disable that hideous shop button.
>
> I tried to read the man file and I got the impression that
> xrdb needs to read that file you suggested. Is that correct?
Yes. There are a number of locations that X resources are read from
automajikally, including /etc/X11/Xresources (under Linux) and, IIRC,
~/.app-defaults under most Unices.
You might want to check the CDE documentation. This is tough --
it **ALL** sucks -- I've wrestled with CDE and its HP precursor,
VUE, for a couple of years. At least HP provided some fscking
decent man pages for VUE. There is a book _Configuring CDE: the
common desktop environment_, Charles Fernandez (Prentice Hall --
http://www.amazon.com/exec/obidos/ASIN/0131027247/), which covers some
issues, but it's very much WIMP oriented and gives an extremely inadequate
treatment of configuration files. Amazon shows a couple more upcoming
titles.
The O'Reilly X Windows user and admin guides, along with a decent set of
man pages, are far superior.
IMO CDE is a bletcherous blot, the Linux WMs and desktops are far superior
and will likely supplant CDE within the next year or so.
If you choose a nonstandard location for parking your resources (or want
to refresh resources later on), it helps to know how xrdb works. There
are three options I generally use:
$ xrdb -query # view current resources
$ xrdb -load file # load resources from file
$ xrdb -remove # remove listed resources
Generally, I query resources (usually piping through a pager),
frequently delete all current resources, then load fresh ones from a
file or set of files:
$ xrdb -query | less
$ xrdb -query | xrdb -remove # removes *all* current X resources
$ xrdb -load file # for a single file
$ for file in *; do xrdb -load $file; done # for ++ files
Resources are associated with a specific X display, defined by your
$DISPLAY variable.
> The setup here is that I use an IBM with AIX/CDE but run
> netscape off a linux box through the network (via setenv
> DISPLAY). I would like to implement any changes to my usage
> of netcsape only so I don't want to change any system files,
> just ones in my home directory for now. I ran the file you
> gave, and yes, everything worked magnificently in netcsape.
> Unifortunately, my other terminals were kind of screwed.
I don't understand what you mean by this. Other clients (windows)
on the same display, or other X Window sessions?
The Netscape resources given are all fully defined (starting with
"Netscape*..."), and should *not* affect other client resources.
> I don't have the manual that's refferred to (unless it's
> online somewhere). Are there any other suggestions/pointers
> you can give me? How can I get the defaults beck? Thanks
> kindly,
See the xrdb references above.
> Fred Nastos
> Dept of Physics
> Queen's University
> Kingston Ontario
--
Karsten M. Self <[EMAIL PROTECTED]> http://www.netcom.com/~kmself
Evangelist, Opensales, Inc. http://www.opensales.org
What part of "Gestalt" don't you understand? Debian GNU/Linux rocks!
http://gestalt-system.sourceforge.net/ K5: http://www.kuro5hin.org
GPG fingerprint: F932 8B25 5FDD 2528 D595 DC61 3847 889F 55F2 B9B0
------------------------------
From: John Gluck <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.questions,comp.os.linux.redhat
Subject: Re: Strange Lilo "LI" problems
Date: Thu, 29 Jun 2000 15:18:08 -0400
"Douglas W. Martin" wrote:
> I've seen the problem before too, although I won't
> pretend to understand the root cause as well as the
> others in this group. In my case, I had installed R.H.
> on a drive and subsequently tried to clone another drive
> to this one. My cloning operation was less than successful
> at clearing the contents of the MBR. Try "fdisk /mbr"
> using a dos bootable floppy and then giving the drive a
> fresh install.
> Hope this helps!
[snip]
OK, I didn't see the original post but here's the explanation.
Each letter printed by lilo means something was done. Because lilo is on the
MBR there's very little space.
This makes it impossible to put nice messages like "I looked for the kernel but
I can't find it"
The first L means that the first stage loader was found and started executing.
The I means that the second stage loader was found by the first stage. It tried
to find the kernel and couldn't.
The LI problem means, in most cases, that you either made a new kernel and
forgot to run lilo or that the kernel is located somewhere above cylinder 1024
on your disk.
When you use a large capacity hard disk and have one monolithic partition for
linux the kernel will often be above cyl 1024.
The way around this is to make a small partition at the start of the disk
(about 20 Megs) and call it "/boot" then make the / partition.
All the kernels and other stuff required to boot goes in this partition.
There is also a version of lilo (21.4 if memory serves) that does not have the
1024 cyl limitation.
--
John Gluck (Passport Kernel Design Group)
(613) 765-8392 ESN 395-8392
Unless otherwise stated, any opinions expressed here are strictly my own
and do not reflect any official position of Nortel Networks.
------------------------------
From: "Mark VanBogart" <[EMAIL PROTECTED]>
Subject: Re: DSL question in Arizona...
Date: Thu, 29 Jun 2000 14:37:19 -0500
I would recommend going with the external because you can treat the
connection as ethernet and not have any special PPPOE drivers as you
normally would with the internal card. If you later decide to switch to
cable modem later you should not have to make any special changes other than
IP address information. You can telnet or use Windows Hyper Terminal to
open up ports on the external DSL box if you need a server on the internet.
I currently use US West DSL on Minnesota and it works a lot better.
"Bryan Roseberry" <[EMAIL PROTECTED]> wrote in message
news:FyN65.196$[EMAIL PROTECTED]...
> I have Linux and I want to get DSL from USWest in Arizona. What hardware
> should I get? External or internal modem? Any suggestions about
> preparing for an install?
>
> Change the ! in the email address to a w to respond...
>
> Thanks!
>
> Bryan
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: rebuilt kernel doesn't autoload modules
Date: Thu, 29 Jun 2000 19:43:16 GMT
In article <[EMAIL PROTECTED]>,
"David .." <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] wrote:
> >
> [snip]
>
> Did you build loadable module support in the kernel when you
configured it?
Yep, I got that part right. I know the kernel is able to load modules
and I know I have the right module compiled because I can load it
manually later.
> Did you add the "CONFIG_NET_EISA and CONFIG_DEC_ELCP" modules while
> configuring the kernel?
Nope, I didn't include these things. I don't have an EISA motherboard
though. What are they for?
> Did you use the new "tulip.c" module?
>
> What brand & model NIC is it?
Ya, there were two: tulip.o and old_tulip.o. Actually either seemed to
work, and I compiled both as modules. When RedHat configured my
system originally, it added:
alias eth0 old_tulip
to /etc/conf.modules, so I would have expected it to load the old_tulip
driver. It's a Kingston 10/100 NIC.
Aaron
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: sylvain hutchison <[EMAIL PROTECTED]>
Subject: new to linux
Date: Thu, 29 Jun 2000 16:10:04 -0700
Hello,
I kind new to linux, and I am trying to get myself connected to the
server, but the IT guys don't know to it with linux!!!!
I was told that I needed to use a couple of commands. Does anybody know
any commands that can help me,
thanks a lot for any information!!!!
Sly
------------------------------
From: Tom Williams <[EMAIL PROTECTED]>
Subject: Re: DSL question in Arizona...
Date: Thu, 29 Jun 2000 19:55:26 GMT
In article <FyN65.196$[EMAIL PROTECTED]>,
Bryan Roseberry <[EMAIL PROTECTED]> wrote:
> I have Linux and I want to get DSL from USWest in Arizona. What
hardware
> should I get? External or internal modem? Any suggestions about
> preparing for an install?
>
> Change the ! in the email address to a w to respond...
>
> Thanks!
>
> Bryan
>
>
We have a DSL connection to the Internet in my office and we plugged the
external modem into our Linux-based gateway box which has two NICs.
I would think an external modem would be best so that you don't have to
shutdown and poweroff the Linux box to do things with the modem.
Peace....
Tom
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Tom Williams <[EMAIL PROTECTED]>
Subject: Re: kernel
Date: Thu, 29 Jun 2000 19:57:18 GMT
In article <B2K65.7303$[EMAIL PROTECTED]>,
"bLACK" <[EMAIL PROTECTED]> wrote:
> How do you config the kernel in mandrake 7.0???
> tried make config, make menuconfig & make xconfig but keep getting a
> parramiters error please help.
>
> --
> -----------------------------------------------------
>
>
I think you should be able to go to /usr/src/linux (as root) and type
"make config" or "make menuconfig" (for a character-mode color
configuration) or "make xconfig" (if you're running X).
What specific errors are you getting?
Peace.....
Tom
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: sylvain hutchison <[EMAIL PROTECTED]>
Subject: network question
Date: Thu, 29 Jun 2000 16:12:29 -0700
Hello,
I kind new to linux, and I am trying to get myself connected to the
server, but the IT guys don't know to do it with linux!!!!
I was told that I needed to use a couple of commands. Does anybody know
any commands that can help me,
thanks a lot for any information!!!!
Sly
------------------------------
From: sylvain hutchison <[EMAIL PROTECTED]>
Subject: gcc compiler
Date: Thu, 29 Jun 2000 16:39:55 -0700
Me again, I am looking for a gcc compiler. I'm thinking of downloading
it off the web, do I have to do anything else but install it on my
pc???? And if anybody knows a site where I can download c++ compliers
from???
Thanks a lot.
Sly.
------------------------------
From: "Dave Breece" <[EMAIL PROTECTED]>
Subject: Re: i810 video card driver support in xfree86 4.0
Date: Thu, 29 Jun 2000 13:47:34 -0700
"D G" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Dave Breece wrote:
> >
> > Supposedly, xfree86 4.0 supports Intel's i810 chipset, including
> > the i810e board that is in my quantex box. Although it's not listed
> > in the x86config card database, you can edit the XF86Config by
> > hand.
> >
> > After getting everything set up, running startx gives me the error:
> >
> > (II) I810(0): AGPIOC_ACQUIRE failed
> >
> > Does anyone know how to fix this, or alternately, does anyone
> > know where I can look this error up?
>
> Do you have the agpgart module loaded? Do you have a /dev/agpgart
> device (or /dev/agpgarti810 (i think) for suse)?
>
> If not, that's probably why you get the error.
/dev/agpgart is there. Installed it a few days ago, and cleanly. After a
couple of
days of fiddling with the XF86Config, I get the similar, but possibly more
informative error:
(II) I810(0): AGPIOC_ACQUIRE failed
Fatal server error:
AddScreen/ScreenInit failed for driver 0
Additionally, when I ran vgatest (I'm trying everything), it informed that
if found
no VESA bios, which doesn't sound good, but I'm having trouble finding any
info on such a thing.
------------------------------
From: Jim Harback <[EMAIL PROTECTED]>
Subject: Re: network question
Date: Thu, 29 Jun 2000 22:50:56 +0200
Sly,
What OS is on the server and what OS are you running. What distro
and version? Do you have root access to both the server and work
station?
Jim
sylvain hutchison wrote:
> Hello,
>
> I kind new to linux, and I am trying to get myself connected to the
> server, but the IT guys don't know to do it with linux!!!!
>
> I was told that I needed to use a couple of commands. Does anybody know
> any commands that can help me,
>
> thanks a lot for any information!!!!
>
> Sly
------------------------------
From: Mikey <[EMAIL PROTECTED]>
Subject: Re: network question
Date: Fri, 30 Jun 2000 09:58:43 +0000
sylvain hutchison wrote:
> I kind new to linux, and I am trying to get myself connected to the
> server, but the IT guys don't know to do it with linux!!!!
If you're using X-windows (the GUI for Linux), logged in as root, go to
the shell (terminal) and
type 'netconf', or 'netconfig', or 'netcfg'.
You should be able to do anything that you need to set up your network
settings there.
> I was told that I needed to use a couple of commands. Does anybody know
> any commands that can help me,
Logged in as root at the command line, you can type 'ifconfig' and that
will let you know your basic network info.
Let me know if this helps, or if you need more info.
--
Since-beer-leekz,
Mikey
Best comment in a kernel /*Drunk...fix later*/
------------------------------
From: [EMAIL PROTECTED] (Matthew Miller)
Crossposted-To: comp.os.linux.questions,comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: "screen" problems while using a vt320
Reply-To: [EMAIL PROTECTED]
Date: Thu, 29 Jun 2000 21:07:16 GMT
Hi Pete,
On Thu, 29 Jun 2000 18:24:10 GMT, Pete Zaitcev <[EMAIL PROTECTED]> wrote:
>>
>> Hum, the comm. setup with the terminal right now is that xoff is off. I also
>> tested the settings with xoff set to 64 and 128. No change.
>
>My is at 128.
Well, I tried that and 64 but I've had no prior problems with xoff turned off.
>> [...] I'm almost desperate enough to recompile screen with debugging
>> symbols and sic gdb on it! But, I'm not to that point yet. It is really
>> bothersome to get no output from screen or error messages :(
>
>You are an oldtmer too, I see :) Before going the gdb route, try
>running screen under strace, something like "strace -o /tmp/xxx screen".
Thank you, though I'm not an old timer (I'm still a newbie w/ only three
years of unix, linux and netbsd behind me), yet. Me and gdb are only just
close personal friends :)
I've haven't used strace before, my privious slack install was without it,
but I see how useful it is now. Below is the last few lines from the
output:
readlink("/proc/self/fd/0", "/dev/ttyS1", 4095) = 10
stat("/dev/ttyS1", {st_mode=S_ISGID|0101, st_size=0, ...}) = 0
geteuid() = 0
getegid() = 0
open("/dev/ttyS1", O_RDWR) = ? ERESTARTSYS (To be restarted)
--- SIGINT (Interrupt) ---
+++ killed by SIGINT +++
Well it seems that the open blocks on trying to open the serial device
I have the vt320 attached to. This is the last syscall regardless of
how long I wait.
I'm stumpted right now. Any ideas are welcome before ask the maintainer and
look foolish (it's probably something obvious).
Thanks Pete, Matthew
------------------------------
** 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.setup) 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-Setup Digest
******************************
