Linux-Setup Digest #235, Volume #20 Sat, 16 Dec 00 21:13:09 EST
Contents:
Re: 2.2.18 Won't Build??? ("Brian Morrison")
Re: Kernel panic ("Big Toe")
Re: Help with compact flash (David Efflandt)
Re: windows VFAT partitions too fat?! (Gill)
ASUS A7Vm/board, RH7 and IBM ATA100 drive ("Brian Morrison")
Re: how diagnose hardware - Hard lockup then crc error ("Harry-0")
Re: Why does Sndconfig lock up on me? (tom)
Re: Why does Sndconfig lock up on me? (tom)
Keyboard configuration ([EMAIL PROTECTED])
Re: Keyboard configuration (Noble Pepper)
Re: Mandrake 7.2 error (new install) (him self)
Re: X window font server crash: help (=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Re: how diagnose hardware - Hard lockup then crc error
(=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Re: Need insights into Slackware setup.... (Mark Post)
----------------------------------------------------------------------------
From: "Brian Morrison" <[EMAIL PROTECTED]>
Subject: Re: 2.2.18 Won't Build???
Date: Sat, 16 Dec 2000 22:43:32 +0100 (BST)
Reply-To: "Brian Morrison" <[EMAIL PROTECTED]>
On 14 Dec 2000 11:25:17 -0500, Paul Kimoto wrote:
>In article <[EMAIL PROTECTED]>, Fred Love wrote:
>> gcc --version reports 2.96
>
>The 2.2.* kernels do not support the (never-released) 2.96 and 2.97
>compilers. The 2.2.18 release notes say:
>
>: This x86 code is intended to build with gcc 2.7.2 and egcs 1.1.2. Patches
>: for building with gcc 2.95 are merged but less tested than other
>: compilers. Caution is recommended when using gcc 2.95 and feedback is
>: sought.
>
>http://gcc.gnu.org/gcc-2.96.html
>http://www.linux.org.uk/VERSION/relnotes.2218.html
There is a glibc patch available from RH that is supposed to fix this I
believe, it updates glibc from 2.1.xx to 2.2.5, but of course this may
not help you.
You might want to try the 2.4.0-test series kernels, assuming you are
happy to run close to the edge....
--
Brian Morrison [EMAIL PROTECTED]
to reply, change address from 'news' to 'bdm'
...Grim faced, cold as fishwife's fingers, he snatched from the wall
the sickle-sharp boar tusks he used for defacing Readers' Digest....
------------------------------
From: "Big Toe" <[EMAIL PROTECTED]>
Subject: Re: Kernel panic
Date: Sat, 16 Dec 2000 15:33:01 -0800
Some more details on the kernel panic would be helpful. What kind of panic
was it?
"Rolf Bachmann" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello
> I've installed Rocklinux. After I change the lilo.conf file to boot
Windows
> and Linux. But when I boot up linux, the boot process end with a Kernel
> panic. What can I do?
>
> Pleas help me
>
> Rolf
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: Help with compact flash
Date: Sat, 16 Dec 2000 23:59:16 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On 15 Dec 2000 21:43:47 GMT, Mike Haight <[EMAIL PROTECTED]> wrote:
>The card services do function. They will
>'beep away' with the modem or SCSI cards.
>There have been no problems here.
>
>The card and adapter have been slid into a
>Win98 machine and mounts up on its machines
>file system just fine. It was originally
>formatted by a digital camera and as
>side comment, a WinCE machine has no
>problems with it either.
Do you get 2 high beeps when you insert it? What does /var/log/messages
say about it? I have 2 ide channels (4 drives) so my CF was assigned as
/dev/hde, this did the trick: mount -t vfat /dev/hde1 /mnt/flash
(or whatever existing dir you want to mount it on).
If nothing happens maybe you need a newer pcmcia-cs version.
>--
>Thanks in advance,
>Michael
>
>
>Linux 2.2.6
>PCMCIA package 3.0.9
>Distro: Slackware
>Formatted by Canon S100 camera (it is just vfat)
>
>The /etc/pcmcia/ide.opts file (which I think has
>little to do with card recognition because the
>computer acts as if there is no card in the slot).
>
># ATA/IDE drive adapter configuration
>#
># The address format is "scheme,socket,serial_no[,part]".
>#
># For multi-partition devices, first return list of partitions in
># $PARTS. Then, we'll get called for each partition.
>#
>case "$ADDRESS" in
>*,*,*,1)
> DO_FSTAB="y" ; DO_FSCK="y" ; DO_MOUNT="y"
> FSTYPE="vfat"
> OPTS=""
> MOUNTPT="/flash"
> ;;
>*,*,*)
> PARTS="1"
> ;;
>esac
--
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: Gill <[EMAIL PROTECTED]>
Subject: Re: windows VFAT partitions too fat?!
Date: Sun, 17 Dec 2000 02:35:11 +0200
On Wed, 13 Dec 2000, David Efflandt wrote:
> On Tue, 12 Dec 2000 23:59:56 +0200, Gill <[EMAIL PROTECTED]> wrote:
> >
> >Hi all,
> >
> >Since this has turned into an academic debate I would just like
> >to post how i finally managed to get around this problem, in case
> >this might help someone in the future.
> >
> >Here's what finally worked (not that i fully understand what went on)
> >an excerpt from my 'me and my linux' diary i keep under the pillow:
> >
> >
> >1. when creating the extended parttion using linux fdisk
> > it is assigned type 05.
> > this worked fine with the 'old' disk - it had one linux logic
> > and two windows. except the second windows (8GB) was not
> > recognized by linux. and WORSE Svend Olaf Mikkelsen told
> > me this is the wrong type and might damage my system.
> > With or without connection - that hard disk DIED after 3 months!
>
> Not sure what you are saying about the old disk or why it died, but I have
> successfully mixed Linux and FAT32 logical partitions in an extended
> partition.I did use the extended type that Win98 originally called it.
>
not that important. just that it was a new drive, and it had a 05 def for the
extended, and it died prematurely. was curious if this could be part of the
reason.
> >2. Indeed if you let windows create the extended partition it then
> > shows as type 0f. But then it seems that if you place a linux
> > logic there - windows 'recognizes' it and messes with it.
> > (at least when you place it before the windows logic).
>
> Windows should not mess with partitions that are not its type.But
> Windows will step all over BSD if you try to create a FAT partition after
> a BSD slice.I believe that this is because you can only have 1 extended
> partition, and a BSD slice looks like one.
>
What windows should not and what it does not do appear to follow the
footsteps of the company who brought it to life, i guess... ;-)
> The following has been working fine with Mandrake 7.0 on my laptop for
> almost a year now (with 192 MB RAM I don't need swap):
>
Hmm... why is that swap remark? do you say it because you never run processes
that big or because of more technical reasons?
> Disk /dev/hda: 255 heads, 63 sectors, 1099 cylinders
> Units = cylinders of 16065 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hda1 1 653 5245191 b Win95 FAT32
> /dev/hda2 * 654 655 16065 83 Linux /boot LILO
> /dev/hda3 656 1099 3566430 f Win95 Ext'd (LBA)
> /dev/hda5 656 872 1743021 83 Linux /
> /dev/hda6 873 897 200781 83 Linux was swap, now /var
> /dev/hda7 898 1099 1622533+ b Win95 FAT32
>
>
> >3. Eventually the following worked:
> > 1. partitioned the disk using a linux boot into 3 primaries:
> > windows (must be first!), linux, linux swap.
>
> Windows does not have to be first.On hda of my main PC I have (2) /boot
> partitions first and either or both can have LILO.Win98 FAT32 C: is 3rd,
> then an extended partition for everything else.
>
I'll remember that for when i have to do it again (hope not too soon).
Anyway at that point i went for the simplest hoping it would leave me be
(that's why i let go of the linux logical -the sun was beginning to come up..)
> > 2. installed windows on hda1.
> > 3. installed linux on hda2.
> > 4. let *windows* fdisk add the extended, and made several under
> > 4GB logicals within it.
> > 5. added these to the linux /etc/fstab - they work fine.
> > bottom line:
> > a. i gave up on adding a linux logical (ie. have one big fat linux).
> > b. i let windows create and manage the extended.
>
> I agree that any extended partition you intend to use for Windows should
> be created by Windows.And it probably works best to put one big logical
> partition in that first.But once that is done you can remove the logical
> partition with Linux fdisk and split it up like youwant it.As long as
> the Linux partitions are type 82 or 83, Windows should leave them alone.
>
perhaps it is because my setting has three primaries?
anyway ill see what happend when i run out of the 5GB my only linux
partition has...
!! another myth i havent had the energy to check is that linux has difficulties
mounting a dos (logical?) partition that is above 4GB.
is that true?
(i wanted to make a chubby mp3 partition, and let go of the idea..)
thanks,
-Gill
------------------------------
From: "Brian Morrison" <[EMAIL PROTECTED]>
Crossposted-To: uk.comp.os.linux
Subject: ASUS A7Vm/board, RH7 and IBM ATA100 drive
Date: Sun, 17 Dec 2000 00:31:13 +0100 (BST)
Reply-To: "Brian Morrison" <[EMAIL PROTECTED]>
Right
I have installed RH7 on my new system, the install went just fine until
I removed the CD and rebooted the machine.
After the POST, I get to the point where lilo prints:
L
and then nothing.
I'm assuming that this must be due to the RH7 system being unable to
deal with the ATA66 controller in the m/board's KT133 chipset due
either to something missing in the kernel config or to something that2
I haven't done (kernel parameters?). Perhaps I can fix this without
rebuilding the kernel, but if not then the 2.4.0-test kernel series
will sort it out, but is there a fairly painless way to do it using a
2.2.x kernel? I'm unsure whether it is going to work with the gcc 2.96
version that RH7 has installed.
I have looked at the output of lspci -v, and I see the following for
the IDE controllers:
00:04.1 IDE interface: VIA Technologies, Inc. 82C586 IDE [Apollo] (rev
10) (prog-if 8a [Master SecP PriP])
Flags: busmaster, medium devsel, latency 32
I/O ports at d800
Capabilities: [c0] Power Management version 2
Further down there is an unrecognised controller (Promise PDC20265)
that at present has no drives on it.
A simple solution would be useful, I don't really want to use my boot
floppy all the time ;-)
--
Brian Morrison [EMAIL PROTECTED]
to reply, change address from 'news' to 'bdm'
...Grim faced, cold as fishwife's fingers, he snatched from the wall
the sickle-sharp boar tusks he used for defacing Readers' Digest....
------------------------------
From: "Harry-0" <[EMAIL PROTECTED]>
Subject: Re: how diagnose hardware - Hard lockup then crc error
Date: Sat, 16 Dec 2000 17:04:27 -0800
Crossposted-To: comp.os.linux.hardware,alt.os.linux,alt.os.linux.mandrake
This may well be because your computer vents are clogged with dust or the
fan that cools the CPU is also blocked with the stuff or worn out
(bad bearings?) RAM chips usually don't go bad slowly, but cooling does,
causing the CPU to become very error-prone before complete lockups. Take
the cover off the case and see if the CPU fan is clean and turns fast. See
if it still locks. If it does, gently touch the CPU or the heat sink to
find out if it is very hot, or just warm (if you burn your finger, it's
too hot.) If just warm, look elsewhere for the cause, but I'll be that's
it.
In article <[EMAIL PROTECTED]>, "Stan Towianski"
<[EMAIL PROTECTED]> pounded on his/her keyboard until this came out:
> Hi,
>
> My system; AMD-K6 233Mhz 64mb ram has been locking hard at pretty
> frequently lately. It seems to be getting worse. I did not know if it
> is with each Mandrake upgrade ..,7,7.1, now 7.2 or what. I wonder if
> the new KDE 2.0 is buggy.
>
> But I am questioning hardware and I don't know how to figure that out. I
> found memtest and ran that. It blew out once. Then I decided to boot
> from cd and mount my / with usr on it to run the memtest binary I
> created.
>
> Twice it blew out at the apparently last test. test16: Walking zeroes:
> setting ...XSegmentation fault.
>
> Anyone know if memtest works?
>
> Also, after a hard lockup I have to reset or turn it off and on.
> Usually it does not come back up. I usually get a 'CRC Error....' and
> have to reset a few times or turn the system off and wait a few minutes
> (it won't work right away). Then it will run again.
>
> The last time I got this error on rebooting:
> .
> .
> .
> Partition check:
> hda: hda1 hda2 hda3 < hda5 hda6 hda7 hda8 >
> RamDisk: compressed image found at block 0 crc error<6>autodetecting
> RAID arrays autorun ...
> ... autorun DONE.
> VFS: Cannot open boot device 08:06 kernel panic: VFS: Unable to mount
> root fs on 08:06
>
> I usually think of CRC errors with floppies. Could this be a hard drive
> error? or memory? or what?
>
> What programs are there for linux to find hardware problems??
>
> Stan Towianski [EMAIL PROTECTED]
------------------------------
From: tom <[EMAIL PROTECTED]>
Subject: Re: Why does Sndconfig lock up on me?
Date: Sun, 17 Dec 2000 00:55:32 GMT
In article <z6D_5.11863$[EMAIL PROTECTED]>,
"enness" <[EMAIL PROTECTED]> wrote:
> Tom,
> I have the same sound card in my Laptop. But the LM7.2 did not detect
it nor
> does Harddrake. How do you run 'sndconfig'. I did not find sndconfig
> anywhere in my machine. Pl let me know where can I find this program
It's something you're supposed to run from the command line. Get
completely out of Xwindows and just type sndconfig. I can't help you
any more than that because I've never gotten it to run successfully. :(
Tom
Sent via Deja.com
http://www.deja.com/
------------------------------
From: tom <[EMAIL PROTECTED]>
Subject: Re: Why does Sndconfig lock up on me?
Date: Sun, 17 Dec 2000 00:58:00 GMT
In article <[EMAIL PROTECTED]>,
devren bennett <[EMAIL PROTECTED]> wrote:
> sounds like you have a dma or irq problem. you should be able to fix
> it(if its on board) in your bios.
>
It works fine in Win98, so obviously Linux is doing something
differently. If I mess with it in the Bios, how will that affect
Windows' using it?
Any ideas how to do it under Linux?
Tom
Sent via Deja.com
http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED]
Subject: Keyboard configuration
Date: Sun, 17 Dec 2000 01:00:08 GMT
I am a relative newbie. I have a logitech keyboard similar to the
Microsoft natural keyboard
(one of th curved keyboards that's supposed to make you hands happy).
Some of the keys, namely the numberpad and the cursor keys don't work
right. Is there anyway to change/remap the keys?
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Noble Pepper <[EMAIL PROTECTED]>
Subject: Re: Keyboard configuration
Date: Sat, 16 Dec 2000 19:52:53 -0600
[EMAIL PROTECTED] wrote:
> I am a relative newbie. I have a logitech keyboard similar to the
> Microsoft natural keyboard
> (one of th curved keyboards that's supposed to make you hands happy).
> Some of the keys, namely the numberpad and the cursor keys don't work
> right. Is there anyway to change/remap the keys?
>
>
> Sent via Deja.com
> http://www.deja.com/
>
I am using a Logitech wireless model but so far I haven't found any diffs
in the setup from my original compaq.
I am assuming your problems are in a GUI since most people new to linux use
one. If so look at your /etc/XF86Config file. Here't the relevant part of
mine (I left out the comments):
Section "Keyboard"
Protocol "Standard"
LeftAlt Meta
RightAlt Meta
RightCtl Control
ScrollLock Compose
Xkbkeycodes "xfree86"
XkbTypes "default"
XkbCompat "default"
XkbSymbols "us(pc101)"
XkbGeometry "pc"
XkbRules "xfree86"
XkbModel "pc101"
XkbLayout "us"
EndSection
------------------------------
From: [EMAIL PROTECTED] (him self)
Crossposted-To: comp.os.linux.misc,alt.os.linux
Subject: Re: Mandrake 7.2 error (new install)
Date: 17 Dec 2000 01:58:21 GMT
Reply-To: [EMAIL PROTECTED]
On Sat, 16 Dec 2000 17:01:49 GMT,
Don Hinds <[EMAIL PROTECTED]> wrote:
>
> Any ideas? Compaq 7470, Trident Blade 3D (video), AMD K6-3D 533, 20G (5G
> allocated to Linux), 64M
>
You didn't say what version of X you installed.
I would run it up in runlevel 3 and then manually
try Xconfigurator until I found a setting that
worked. It's a good idea to start with a conservative
version of X, eg. 3.3.6 without h/w acceleration.
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,linux.redhat.install
Subject: Re: X window font server crash: help
Date: Sun, 17 Dec 2000 02:45:31 +0100
On Sat, 16 Dec 2000, Gilles Lamoureux wrote:
> People,
>
> I was able to login to /home cia x window. After trying several times, I
> now cannot access X windows. Here are the error messages:
> ////////////////////////////////////////////////////////////////////////////
> ////////////////////////////////////
> FontTransSocket UNIX Connect: can't connect: errno=111
> failed to get default font path 'unix/:-1'
> fatal server error:
> not able to open default font 'fixed'
>
> according to /var/run/gdm.pid, gdm was already running but seems to have
> been murdered mysteriously.
> ////////////////////////////////////////////////////////////////////////////
> /////////////////////////////////
>
> Any ideas. I had been downloading font files to run pdf files and such. I
> will gather exactly what I downloaded. But if the errors here are
> sufficient to help me, please,
> HELP ME. I need to get to my sutdy materials pronto.
If you have downloaded and installed buggy fonts, the font server will
crash at startup. You will have to remove the buggy fonts before starting
xfs.
Not 100% sure (sitting on a non-X box), but I think the xfs error messages
should be found in /var/log/messages.
Rasmus B. Hansen
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,alt.os.linux,alt.os.linux.mandrake
Subject: Re: how diagnose hardware - Hard lockup then crc error
Date: Sun, 17 Dec 2000 03:00:48 +0100
On Sat, 16 Dec 2000, Stan Towianski wrote:
To me this sounds like a memory problem. Are you running PC66 RAM at
speeds >66MHz?
If you have any chance of pulling out your RAM and inserting some working
RAM, you could try that (e.g. borrow some). Also, if you have more than
one piece of RAM, try pulling one out - then try pulling the other one out
etc.
If you get an error (like the kernel uncompressing CRC-errors) sometimes,
but not others, it is probably not the hard drive.
You could also look at www.bitwizard.nl/sig11/ - it tells something about
memory problems.
Hope this helps.
Rasmus B�g Hansen
> My system; AMD-K6 233Mhz 64mb ram has been locking hard at pretty
> frequently
> lately. It seems to be getting worse. I did not know if it is with
> each
> Mandrake upgrade ..,7,7.1, now 7.2 or what. I wonder if the new KDE 2.0
> is buggy.
>
> But I am questioning hardware and I don't know how to figure that out.
> I found memtest and ran that. It blew out once. Then I decided to boot
> from cd and mount my / with usr on it to run the memtest binary I
> created.
>
> Twice it blew out at the apparently last test.
> test16: Walking zeroes: setting ...XSegmentation fault.
>
> Anyone know if memtest works?
>
> Also, after a hard lockup I have to reset or turn it off and on.
> Usually
> it does not come back up. I usually get a 'CRC Error....' and have to
> reset a few times or turn the system off and wait a few minutes (it
> won't
> work right away). Then it will run again.
>
> The last time I got this error on rebooting:
> .
> .
> .
> Partition check:
> hda: hda1 hda2 hda3 < hda5 hda6 hda7 hda8 >
> RamDisk: compressed image found at block 0
> crc error<6>autodetecting RAID arrays
> autorun ...
> ... autorun DONE.
> VFS: Cannot open boot device 08:06
> kernel panic: VFS: Unable to mount root fs on 08:06
>
> I usually think of CRC errors with floppies. Could this be a hard drive
> error? or memory? or what?
>
> What programs are there for linux to find hardware problems??
>
> Stan Towianski
> [EMAIL PROTECTED]
>
------------------------------
From: [EMAIL PROTECTED] (Mark Post)
Subject: Re: Need insights into Slackware setup....
Date: Sun, 17 Dec 2000 02:08:25 GMT
On 13 Dec 2000 17:01:37 GMT, Bob the old greybeard
<[EMAIL PROTECTED]> wrote:
>I am trying to set up a slackware 7.1 box
Good choice. :)
>and am missing the boat in
>getting LILO set up correctly.
What does your /etc/lilo.conf file look like?
>I mostly use BSD/AIX/Solaris boxes
>and it has been several years since I last installed a Linux box.
>On setting up the partitions, I use the entire disk, starting root
>at cylinder 0. Everything installs fine, but, when I get to the
>point of installing LILO, it errors out with a file system full
>error 139 thing, and won't write the boot properly. Reboot dumps
>registers all over the screen, indicating that it really did not
>write a boot sector. On the install I chose it to write to the mbr,
>and was expecting it to write sector 0, properly.
Assuming this is the master drive on the primary IDE channel, you should
have something on the order of boot = /dev/hda1 in your /etc/lilo.conf file.
>I noticed that
>there was also an option to write to another location or something
>like that rather than the mbr. Which should be used?
Based on what you say you want, write into the MBR.
>What I want
>is no dualboot loader of any kind, but only boot directly into Linux.
>I don't want to have to use a boot floppy to correctly boot it up.
>Do I need to set the root partition to one cylinder in from the first
>cylinder so that LILO has room to write its boot blocks, or what?
That shouldn't be necessary. When you partitioned your drive, the
partitioning software (fdisk, cfdisk) make the allowances necessary for this
to work right.
Mark Post
Postmodern Consulting
Information Technology and Systems Management Consulting
To send me email, replace 'nospam' with 'home'.
------------------------------
** 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
******************************
