Linux-Misc Digest #998, Volume #26 Fri, 2 Feb 01 06:13:01 EST
Contents:
User write privilege on /dos partition
Re: Moving boot (Eric)
Re: Moving boot ("Tauno Voipio")
Re: ALD - Assembly Language Debugger - where? ("Tauno Voipio")
CLUSTER SIZE FOR EXT2 filesystem.. ("Radix")
Re: How to access windows partitions from linux? ("Tauno Voipio")
Re: Backup remotely ? (Wong Sai-kee)
Re: Creating CD from ISO image? ("Tauno Voipio")
Modem works in RH6.2, not 7.0 (Jess Canada)
Re: Kernel NFS problems...help ("Peter T. Breuer")
Re: User write privileges to /dos partitions (Eric)
Re: Mounting MS-DOS File System Problems, e2fsck & Miscellaneous Questions
([EMAIL PROTECTED])
Re: crc error - system halted (Eric)
Re: Linux on Alpha station (Martin Gregorie)
Re: implementation of colored man pages (Martin Gregorie)
Re: MS to Enforce Registration - or Else (Ian Davey)
Re: USB printer with kernel 2.2.18... (John Thompson)
Re: Multibooting 5 OSs => Win98, NT4, Linux, Solaris 8 and Unixware 7 - HOW TO???
([EMAIL PROTECTED])
Re: CLUSTER SIZE FOR EXT2 filesystem.. (Eric)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] ()
Subject: User write privilege on /dos partition
Date: Fri, 02 Feb 2001 07:11:08 GMT
I've browsed the how-to's and the FAQ's, trying different flags,
but to no avail.
As root, I can read/write to /dos*. As joe user however, I can
only read.
Can anyone suggest a fix so that "joe user" can also write to the
/dos* partitions?
I'm using slackware 7, if that matters.
Thanks
Here's an extract from my fstab:
/dev/sda1 swap swap defaults 0 0
/dev/sda2 / ext2 defaults 1 1
/dev/sdb4 /zip vfat user,noauto 0 1
/dev/hda1 /dosc vfat defaults 1 0
/dev/hda5 /dosd vfat defaults 1 0
/dev/hdc /cdrom iso9660 ro,user,noauto 0 1
[snip]
------------------------------
From: Eric <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Moving boot
Date: Fri, 02 Feb 2001 09:07:40 +0100
Reply-To: [EMAIL PROTECTED]
> Disk /dev/hdc: 15 heads, 63 sectors, 13328 cylinders
> Units = cylinders of 945 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hdc1 * 1 1300 614218+ 7 HPFS/NTFS
> /dev/hdc2 2439 13328 5145525 5 Extended
> /dev/hdc5 2439 6877 2097396 83 Linux
> /dev/hdc6 6878 11316 2097396 83 Linux
>
>
> image=/vmlinuz-2.2.5-15
> label=linux
> root=/dev/hdc5
> initrd=/initrd.img
> lilo -v
> Warning: BIOS drive 0x82 may not be accessible
> Warning: device 0x1605 exceeds 1024 cylinder limit
> LILO version 21, Copyright 1992-1998 Werner Almesberger
>
> Reading boot sector from /dev/hda
> Merging with /boot/boot.b
> geo_comp_addr: Cylinder number is too big (2679 > 1023)
>
The complaint is about hdc5
Update LILO to a version 21.4.3 or higher and add the lba32 keyword to
lilo.conf.
You discs are all detected very weird. The maximum number of heads is
255, where your
disc all use approx. 15 heads. This means you'll get a high number of
cylinders.
This is due to poor BIOS settings. If you ever plan to reinstall, fix
that.
(If you would do it now, nothing would be bootable)
AND DON'T CROSSPOST TO SO MANY GROUPS.
(I trimmed the list)
Eric
------------------------------
From: "Tauno Voipio" <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux,comp.os.linux,comp.os.linux.questions,comp.os.linux.redhat,comp.os.linux.setup,linux.redhat.install,linux.redhat.misc
Subject: Re: Moving boot
Date: Fri, 02 Feb 2001 08:09:48 GMT
"Chris Woodhouse" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I have redhat 6.0 installed on a pentium 90 with 4 hard drives. When I
> installed it I had to use a floppy to boot because of the locations of
> the partitions I was using for linux but now I have cleared some
> partitions. In fact except for the first partition on the first hard
> drive I could use any other partition to put /boot on so I tried to move
>
> /boot and change my lilo conf to boot to the new drive but when I run
> lilo -v it tells me that the hard drive has more than 1024 cylinders but
>
> cat /etc/lilo.conf
> boot=/dev/hda
> timeout=100
> message=/boot/message
> prompt
> image=/vmlinuz-2.2.5-15
> label=linux
> root=/dev/hdc5
> initrd=/initrd.img
> image=/vmlinuz-2.2.5-15
> label=rescue
> append="load_ramdisk=2 prompt_ramdisk=1"
> root=/dev/fd0
> initrd=/initrd.img
> other=/dev/hda1
> label=dos
> table=/dev/hda
>
> lilo -v
> Warning: BIOS drive 0x82 may not be accessible
> Warning: device 0x1605 exceeds 1024 cylinder limit
> LILO version 21, Copyright 1992-1998 Werner Almesberger
>
> Reading boot sector from /dev/hda
> Merging with /boot/boot.b
> geo_comp_addr: Cylinder number is too big (2679 > 1023)
>
The cylinder count is not the real problem.
LILO (like other loaders) needs to have the /boot on such disk that the BIOS
is able to access it. Usually, BIOS knows only of two first disks (/dev/hda
= 0x80 and /dev/hdb = 0x81). You are trying to force it to read from the
third disk.
It is *not* enough to have the boot sector on /dev/hda: it is not smart
enough to access the other disks.
The bootstrap procedure is:
1. BIOS reads the boot sector into 0x7c00 and jumps in
2. the boot sector reads /boot/boot.b and jumps in
3. boot.b presents the LILO prompt and reads the kernel in
4. the kernel uncompresses and initialises
5. the kernel starts up in 32 bit mode and own drivers
All data needed till step 4 must be accessible to the BIOS disk driver, as
the Linux drivers come in with the kernel.
If I were you, I'd try to clean up a small partition (8 - 16 MB) at the
start of /dev/hda and put /boot there.
Tauno Voipio
tauno voipio @ iki fi
------------------------------
From: "Tauno Voipio" <[EMAIL PROTECTED]>
Subject: Re: ALD - Assembly Language Debugger - where?
Date: Fri, 02 Feb 2001 08:13:16 GMT
<[EMAIL PROTECTED]> wrote in message
news:95cun2$na1$[EMAIL PROTECTED]...
> Yep. Thanks a lot. gdb is great, but the ALD is (supossed to
> be) an integrated disassembler + debugger. The tools that I
> have seen so far are either a disassembler or a debugger
> (i.e. gdb), with hopes for a reasonable GUI. The ldasm
> is almost there, but falls short of being a debugger.
> Thanks a lot, again.
> -dan
>
There is a GUI interface to GDB. See www.gnu.org for DDD.
Tauno Voipio
tauno voipio @ iki fi
------------------------------
From: "Radix" <[EMAIL PROTECTED]>
Subject: CLUSTER SIZE FOR EXT2 filesystem..
Date: Fri, 2 Feb 2001 04:40:38 -0330
Hi guys:
Does anyone know the cluster size for the ext2 filesystem...??? Does it
depend on the size of the partition that is being used? For instance, 8GB
partitions under FAT32 use 4KB clusters. Anything greater than an 8GB
partition under FAT32 would use 16KB or greater... This all equals a great
drive waste...
Someone said that you could select your own partition size in linux! Is
this true? If so, how???
Thanks,
Trevor...
------------------------------
From: "Tauno Voipio" <[EMAIL PROTECTED]>
Subject: Re: How to access windows partitions from linux?
Date: Fri, 02 Feb 2001 08:22:26 GMT
"Chris Divine" <[EMAIL PROTECTED]> wrote in message
news:95celv$[EMAIL PROTECTED]...
>
> Another caution is to make sure you know which partitions to mount. I
> created directories under /mnt called /mnt/win98 and /mnt/win2k then tried
> to mount my two windows partitions by saying "mount -t vfat /dev/hda1
> /mnt/win98" and "mount -t vfat /dev/hda2 /mnt/win2k". The first one was
> successful, the second one wasn't. It turns out that the second partition
on
> my primary IDE drive is actually seen by Linux as /dev/hda5. So after I
> mounted the second partition as "mount -t vfat /dev/hda5 /mnt/win2k" it
> worked and now I can see all the files on both of my windows partitions.
>
That is due to the fact that the 'second partition' is indeed the first
logical disk. Windows shows the fact only in fdisk. The naming practice
helps reducing confusion.
The primary partitions are numbered 1 to 4, the logical disks starting from
5.
Tauno Voipio
tauno voipio @ iki fi
------------------------------
From: Wong Sai-kee <[EMAIL PROTECTED]>
Subject: Re: Backup remotely ?
Date: 2 Feb 2001 08:17:05 GMT
LFessen106 <[EMAIL PROTECTED]> wrote:
: Actually, it does, and I do it every day. This is, in fact, my prefered
: ...
: tar czvf - / --exclude /proc --exclude /dev --exclude /mnt | rsh theDDSmachine
: 'dd of=/dev/st0'
Thank you. The reason I said fail was I hadn't think of using dd.
I intended to use also tar in the remote side but the tar syntax doesn't
support.
But, I don't understand why you need to exclude /dev ? And, is it simpler
to do:
tar czvlf - / | rsh theDDSmachine 'dd of=/dev/st0'
SK
------------------------------
From: "Tauno Voipio" <[EMAIL PROTECTED]>
Subject: Re: Creating CD from ISO image?
Date: Fri, 02 Feb 2001 08:26:57 GMT
"Rob Peacock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hey all...
>
> I searched this newsgroup for an answer before I posted so please
> don't hammer me for asking.
>
> I am trying to burn a copy of RH7 from the ISO images downloaded from
> ftp.freesoftware.com using my Mac but am having some issues.
> (The only good use of a PC is with Linux installed on it.)
>
> I have a a Mac 8500, 132Mhz, 304Meg RAM, all SCSI drives, and a SCSI
> Smart &Friendly SAF759 burner. I'm using Adaptec Toast 3.5.7 and
> haven't had an issue making Mac CD's in the past.
>
> This is the first time I have tried to make an ISO CD and am not sure
> how to go about it. I did RTFM but it focused more on how to build
> hybrid Mac/ISO9660 cd's, and only then, building them from scratch. It
> also appears that when they talk about building a CD from an image,
> they are talking about making it from a Mac disk image.
>
> When I tried to follow what I thought they were saying, the progress
> bar just froze. There didn't appear to be any CD activity although the
> hard drive light stayed on for a while. When I checked the CD, it was
> still unburned and had no content.
>
I'm not familiar with Mac's. In Windows, the Adaptec software have options
to burn the CD from an ISO image on disk (instead of Create CD, Write CD
from Image File).
There is the possibility that the download of the ISO image has not happened
in binary mode and the software notices that the structure of the image does
not make sense.
Tauno Voipio
tauno voipio @ iki fi
------------------------------
Subject: Modem works in RH6.2, not 7.0
From: [EMAIL PROTECTED] (Jess Canada)
Date: Fri, 02 Feb 2001 08:37:30 GMT
I have an ActionTec Call Waiting 56k PCI modem that worked under Red Hat
6.2 for me very well. I have since installed RH7.0 (as a fresh install,
not an upgrade) and cannot get the modem to work at all.
This is how I set it up: I do "cat /proc/pci" and get the I/O port and the
IRQ, and put those in the setserial command: "setserial /dev/modem uart
16550A port 0x1800 irq 9" and that worked fine in RH6.2. But now, with
RH7.0, that same procedure doesn't work, although the port and irq are the
same. I did notice that the USB driver that 7.0 has uses IRQ 9, like my
modem, so I didn't know if that could cause a conflict, but they should be
able to share. If not I'd be happy not to have the USB driver anyway, I
don't need it.
If anyone knows how I can fix this so I can use my modem, I would really
appreciate it. Sorry, I'm a beginner. Also, email any responses to me at
[EMAIL PROTECTED] if you can, since I might not be able to check the
newsgroup. Thanks,
Jess Canada
[EMAIL PROTECTED]
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.networking,linux.dev.kernel
Subject: Re: Kernel NFS problems...help
Date: Fri, 2 Feb 2001 09:21:57 +0100
In comp.os.linux.help Bob Sully <[EMAIL PROTECTED]> wrote:
> In 2.2.17 and earlier, everything seems to work fine. It's in 2.2.18
> and the 2.4 series that VFAT (and apparently iso9660) directories are
> not mountable via NFS.
> By the way (Peter), I tried a recent version of unfsd (2.2-47beta); no
> luck under 2.4.1; it does work OK under 2.2.17.
OK, so it must be a VFAT change .. if you wanted to locate the bug
you would take the 2.2.17 msdos/vfat dir and plug it into 2.2.18
to start pinning down where it is. Or just eyeball at the vfat patch
slice from 2.2.17 to 2.2.18.
You'd also want to try exporting a umsdos directory to see if that
fixes it (mount the fs as umsdos instead of vfat).
I'll see if I can verify it.
Alan Cox needs to be notified about this one.
Peter
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: User write privileges to /dos partitions
Date: Fri, 02 Feb 2001 10:34:53 +0100
Reply-To: [EMAIL PROTECTED]
Gregory Allen wrote:
>
> I've browsed the how-to's and the FAQ's, trying different flags,
> but to no avail.
>
> As root, I can read/write to /dos*. As joe user however, I can
> only read.
>
> Can anyone suggest a fix so that "joe user" can also write to
> the /dos* partitions?
>
> I'm using slackware 7, if that matters.
>
> Thanks
>
> Here's an extract from my fstab:
> /dev/sda1 swap swap defaults 0 0
> /dev/sda2 / ext2 defaults 1 1
> /dev/sdb4 /zip vfat user,noauto 0 1
> /dev/hda1 /dosc vfat defaults 1 0
> /dev/hda5 /dosd vfat defaults 1 0
> /dev/hdc /cdrom iso9660 ro,user,noauto 0 1
Add a umask option (it's probably in `man fstab`)
/dev/hda5 /dosd vfat defaults,umask=000 1 0
Eric
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Mounting MS-DOS File System Problems, e2fsck & Miscellaneous Questions
Date: 2 Feb 2001 10:46:45 +0100
[EMAIL PROTECTED] wrote:
> David wrote:
>>
>> Have you tried to mount it with something like this?
>>
>> mount -t fat /dev/fd0 /floppy
>>
>> Or mount -t fat /dev/fd0h720 /floppy
>>
> I tried those, but it said
> mount: fs type fat not supported by kernel
The filesystem you want to use is either msdos (classical 8.3 filenames) or
vfat (with long filenames as in Win9x).
--
Alain Borel
<[EMAIL PROTECTED]>
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: crc error - system halted
Date: Fri, 02 Feb 2001 10:47:13 +0100
Reply-To: [EMAIL PROTECTED]
Chris Divine wrote:
>
> I am frequently running into the error that says:
>
> Uncompressing Linux...
>
> crc error
>
> -- system halted
>
> I have gotten several explanations from people as to what may be causing
> this, from bad RAM to a bad hard disk to bad bus speed on the mobo to . . .
And have you investgated these suggestions?
IIRC there's a program called memtest86 that you can use
to test your RAM if you're on an x86(intel) machine.
Use dd to test your HDD.
(dd if=/dev/hda of=/dev/null)
Check the cabling too.
And check the BIOS settings for wrong bus-speeds
eric
------------------------------
From: [EMAIL PROTECTED] (Martin Gregorie)
Subject: Re: Linux on Alpha station
Date: Fri, 02 Feb 2001 09:54:53 GMT
On Fri, 02 Feb 2001 01:29:13 GMT, [EMAIL PROTECTED] wrote:
>Ah, but if it's not 64 bit clean, _that_ would introduce some
>issues...
>
Didn't seem to. By and large you can ignore 64 bit issues when writing
code for an alpha. The only things I really noticed is that word
alignment issues can have a huge impact - especially for pointers -
and that a long is really long.
Its a nice chip and fast with it; we found that an Alphaserver 1000
(150MHz uniprocessor with 256 MB RAM) was quite adequate as a
development box for a team of 16 programmers.
--
gregorie | Martin Gregorie
@logica | Logica Ltd
com | +44 020 76379111
------------------------------
From: [EMAIL PROTECTED] (Martin Gregorie)
Subject: Re: implementation of colored man pages
Date: Fri, 02 Feb 2001 10:09:49 GMT
On 1 Feb 2001 22:38:50 GMT, Thomas Dickey <[EMAIL PROTECTED]>
wrote:
>It's possible that xterm is doing it (colors can be assigned to bold,
>underline and reverse). Just checking, I see that this is the case for
>Redhat 6.2 (read the XTerm app-defaults file).
>
Same applies to termcap - some implementations use colours, others use
underlining and/pr reverse video.
For instance RH 6.2 colours man pages and directories with the xtem
(PuTTY) and telnet (Teraterm) implementations I've used to access it.
--
gregorie | Martin Gregorie
@logica | Logica Ltd
com | +44 020 76379111
------------------------------
Crossposted-To: comp.os.linux.advocacy
From: [EMAIL PROTECTED] (Ian Davey)
Subject: Re: MS to Enforce Registration - or Else
Date: Fri, 02 Feb 2001 10:28:12 GMT
In article <[EMAIL PROTECTED]>, "Aaron R. Kulkis" <[EMAIL PROTECTED]> wrote:
>Ian Davey wrote:
>>
>> In article <[EMAIL PROTECTED]>, "Aaron R. Kulkis"
> <[EMAIL PROTECTED]> wrote:
>> >> As far as Atheists are concerned, god doesn't exist any more than Santa
> Claus
>> >> or the Easter Bunny.
>> >
>> >Which *IS* a belief in itself.
>>
>> Not true. I read lots of novels and enjoy them, but don't believe any of the
>> content as it's just fiction. There's no belief system wrapped up in it.
>> There's no need to pay any attention to people who elevate stories into a
>> belief system.
>
>You have just expressed a belief.
>
>It might be true, or not...either way, it IS a belief.
You've still not managed to convince me. But perhaps it's just a matter of
semantics, Atheism is a lack of *religious* beliefs (theism), but not a lack
of belief. Does that work better for you? So an Atheist believes something
other than religion.
There is a valid proof that God doesn't exist, but I can't remember who coined
it now: "the existance of God disproves the existance of God, therefore God
does not exist". Seeing as you're mathematically/scientifically minded, you
might be able to figure out the logic of that statement.
Not Christian are you?
ian.
\ /
(@_@) http://www.eclipse.co.uk/sweetdespise/ (dark literature)
/(&)\ http://www.eclipse.co.uk/sweetdespise/libertycaptions/ (art)
| |
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.portable,linux.redhat.misc
Subject: Re: USB printer with kernel 2.2.18...
Date: Thu, 01 Feb 2001 20:11:20 -0600
Linux-Addict wrote:
> upgrade to 2.4 kernel the 2.2 kernal doesn't support to much of anything
> with USB.
Kernel 2.2.18 includes backported USB support from the 2.3
development kernels. I haven't tried it with a printer but it
works fine with my Epson USB scanner.
--
-John ([EMAIL PROTECTED])
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: linux.redhat.install,comp.unix.solaris,alt.solaris.x86
Subject: Re: Multibooting 5 OSs => Win98, NT4, Linux, Solaris 8 and Unixware 7 - HOW
TO???
Date: Fri, 02 Feb 2001 10:43:41 GMT
Harold,
Thanks for your help. Your suggestions seem interesting, especially
because they are exactly what I want to accomplish (Win98 on HD1 and
all the rest on HD2). However, it seems that UnixWare 7.1 *needs* to be
on the boot disk and occupying an area below 4GB. From the ton of
usenet articles I've read, I've gathered the following constraints to
my question (opinions vary, which complicates things!):
1. Microsoft OSs should be installed first (W98, then NT).
2. System Commander should be installed after W98.
3. Unixware should be installed on the boot disk, under the 4GB mark.
4. Linux and Solaris could be installed on HD2 (same as NT), but care
should be taken to avoid their clobbering each other's swap partitions.
5. The correct order appears to be:
Win98
Win NT
[UnixWare??]
Solaris
[UnixWare??]
Linux
[UnixWare???]
6. I'm still unsure about the correct order of the UnixWare
installation. Should it be the first Unix to be installed? Does it even
matter?
Thanks again,
Alex
> On HD1, just put Win98 on one big FAT32 partition. Screw the Fat16's.
>
> Install PowerQuest's BootMagic (Free with Partition Magic, and
cheap.) Don't
> know much about System Commander, but I try and avoid Norton except
for
> anti-virus.
>
> On HD2, partition it so you have 1 NTFS partition for NT, 3 native
unix
> filesystems for each unix, and 3 native swap partitions for each
unix. Gotta
> tell you, you'd be better putting Win98 on the 8, and using the 30
for the
> unixes, or buying about 30.
>
> Then, with boot magic, put one option to NT, and each Unix option to
the
> bootable filesystems of the unixes. Install order doesn't matter, but
I'd do
> win98 first, so it doesn't blow away the work you do with Boot Magic.
>
> Any help?
>
> -H
>
> <[EMAIL PROTECTED]> wrote in message
news:95905f$5aa$[EMAIL PROTECTED]...
> > My intended setup:
> > =================
> >
> > HD1 (30GB-IDE)==> Win98 on FAT32 (or divided into 2 Fat16/Fat32
> > partitions if necessary)
> >
> > HD2 (8GB-IDE): Solaris 8, linux Mandrake 7.2, NT4 and Unixware 7.1
> >
> > My (quick) questions:
> > ====================
> > (NOTE: I have access to System Commander 2000 in case I need it)
> >
> > 1. Is this even possible?
> > 2. Can I have Win98 on one disk and all the other OSs on the other?
> > 3. If not, what OSs would I *have* to put on HD1?
> > 4. What is the proper installation order?
> > 5. At what point do I install System commander 2000 to assist me?
> >
> >
> > Well, I'm sure I'll have more questions when I try this out, but
> > clearing these preliminary doubts will sure help...
> >
> > Thanks a lot,
> >
> > Alex Olazabal
> >
> > PD. Please copy your reply to [EMAIL PROTECTED]
> >
> >
> > Sent via Deja.com
> > http://www.deja.com/
>
>
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: CLUSTER SIZE FOR EXT2 filesystem..
Date: Fri, 02 Feb 2001 12:07:41 +0100
Reply-To: [EMAIL PROTECTED]
> Does anyone know the cluster size for the ext2 filesystem...??? Does it
> depend on the size of the partition that is being used? For instance, 8GB
> partitions under FAT32 use 4KB clusters. Anything greater than an 8GB
> partition under FAT32 would use 16KB or greater... This all equals a great
> drive waste...
>
> Someone said that you could select your own partition size in linux! Is
> this true? If so, how???
>
`man mke2fs`
As the ext2 FS is open source, you can, no doubt, dig up all the
information you want about it. Go to a decent search engine and give it
a try.
Eric
------------------------------
** 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.misc.
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-Misc Digest
******************************
