Linux-Misc Digest #671, Volume #23 Fri, 25 Feb 00 10:13:03 EST
Contents:
Re: Journaling file system (Ext3 or SGI's) (Peter Cherriman)
Caldera Linux ("Dan Mc")
Please advice: unable to create file/directory (H K Chan)
Re: staroffice or wordperfect?/SO (Ljubomir Josifovski)
Need help: Lilo parameters to override mem=xx ("Peder Godvik")
Re: Major / Minor block device numbers for CDROM (Sitaram Chamarty)
Re: Here's why linux programs are so insecure ! (Chris MacKenzie)
unresolved synbols ("William")
Re: Here's why linux programs are so insecure ! (Robert Grizzard)
Re: Problem with gimp (cll)
Re: LILO patch for past 1024 cylinders? (Steve Gage)
Sending & Receiving Mail --> Local <-- ("SyntaXer")
Re: Boot loader ("Marc")
Re: top in background ([EMAIL PROTECTED])
Re: top in background ([EMAIL PROTECTED])
printing doesn't work, but it does using Samba (Bob Koss)
Apache server mailing list (Subba Rao)
Re: LILO upgrade for the new millenium ([EMAIL PROTECTED])
Re: Here's why linux programs are so insecure ! ([EMAIL PROTECTED])
Re: Boot loader ([EMAIL PROTECTED])
Re: LILO boot problem (CharlieB)
setting my display to 1024x769 (robin)
Re: read -- - PartitionMagic error ( mandrake linux ) ([EMAIL PROTECTED])
Re: Please advice: unable to create file/directory ("Steve Cowles")
kernel panic - please translate/advise ("Chucks")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Peter Cherriman)
Subject: Re: Journaling file system (Ext3 or SGI's)
Date: 25 Feb 2000 10:06:41 GMT
>I have no firsthand experience, nor do I know how far along SGI's project
>is, but it's XFS, not JFS. I'm aware of FOUR journaling filesystems that
>are available in early stages or have been announced for Linux:
>
>1) ext3fs -- An extension of ext2fs.
>2) Reiserfs -- A new open source filesystem.
>3) XFS -- SGI's filesystem, now or soon to be released as open source.
>4) JFS -- IBM's filesystem, now or soon to be released as open source.
>
Can someone tell me what a journaling filesystem is, and what use is it for.
--
Dr. Peter Cherriman EMAIL : [EMAIL PROTECTED]
Communications Group WWW : http://www-mobile.ecs.soton.ac.uk/peter
Dept. of Electronics and Computer Science
University of Southampton, United Kingdom
------------------------------
From: "Dan Mc" <[EMAIL PROTECTED]>
Subject: Caldera Linux
Date: Fri, 25 Feb 2000 10:26:15 -0000
Will Caldera OpenLinux 2.3 run on my P90 with 32mb RAM? i have been told
that this is true, will it run smoother than windows?
Thanks
------------------------------
From: H K Chan <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.hardware,comp.os.linux.setup
Subject: Please advice: unable to create file/directory
Date: Fri, 25 Feb 2000 18:44:34 +0800
Hi,
I am running RedHat Linux release 6.1, I have encountered a wierd
problem that is I, as root, unable to create file or directory in a
partition which has plentiful of disk space. Below is the situation:
[root@it /home1]# touch temp
touch: temp: No space left on device
[root@it /home1]# mkdir temp
mkdir: cannot make directory `temp': No space left on device
[root@it /home1]# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/sda1 5447436 913304 4257416 18% /
/dev/sdb1 6048500 513904 5227348 9% /home
/dev/sda5 3099260 300176 2767600 10% /home1
[root@it /home1]#
Thanks in advance.
Gerald
------------------------------
From: Ljubomir Josifovski <[EMAIL PROTECTED]>
Subject: Re: staroffice or wordperfect?/SO
Date: Fri, 25 Feb 2000 11:09:29 +0000
Thanks to all who replied - I went for SO, and it seems it does the job.
ta,
Ljubomir Josifovski
--
Ljubomir Josifovski, SPandH, DCS, University of Sheffield
Regent Court, 211 Portobello St, Sheffield S1 4DP, UK
tel/fax: +44-(0)114-222-1878/222-1810
mailto:[EMAIL PROTECTED], http://www.dcs.shef.ac.uk/~ljupco
------------------------------
From: "Peder Godvik" <[EMAIL PROTECTED]>
Subject: Need help: Lilo parameters to override mem=xx
Date: Fri, 25 Feb 2000 12:10:11 +0100
I have RH 6.1 installed on a 486 EISA machine with SCSI disks.
I put a line in lilo.conf setting the memory size with
append="mem=57M", and after that I get kernel panic and the
machine will not start.
I've tried to get around it with the "single", "emergency" and "resuce"
parameters for Lilo, but it does not help.
Is there any way I can override the mem=xx... option in lilo at
boot time ?
I do not have a floppy drive around right here, so I would like to
solve this without booting from floppy or CD.
Hoping for help.
Best regards,
Peder Godvik
------------------------------
From: [EMAIL PROTECTED] (Sitaram Chamarty)
Crossposted-To: linux.redhat.misc,comp.os.linux.redhat
Subject: Re: Major / Minor block device numbers for CDROM
Date: Fri, 25 Feb 2000 11:37:47 GMT
On Tue, 22 Feb 2000 16:05:46 -0500, Kendal Montgomery <[EMAIL PROTECTED]>
wrote:
>Detected scsi CD-ROM sr0 at scsi0, channel 0, id 5, lun 0
>
>now I need to make a block device /dev/sr0 to support this cdrom now, how do
>I know the major and minor numbers to give it? Acutally, what the hell are
>the major and minor numbers? What do they mean?
cd /dev
./MAKEDEV sr0
# if that fails...
./MAKEDEV scd0
# then
ln -s /dev/sr0 /dev/cdrom
# (or /dev/scd0 - depending on which one succeeded)
That's the short answer. For the long answer, install the kernel
source tree and look at
/usr/src/linux/Documentation/devices.txt
where all your questions will be answered.
Even if you don't read that, I must tell you that /dev/sr0 is
often called /dev/scd0 on many systems. They both have the same
number (major 11 minor 0).
>A response by email would be appriciated at [EMAIL PROTECTED]
Sorry.
------------------------------
From: Chris MacKenzie <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.os.linux
Subject: Re: Here's why linux programs are so insecure !
Date: Fri, 25 Feb 2000 22:02:36 +1100
[EMAIL PROTECTED] wrote:
> Because you dickheads offer the source code of your programs (!!!)
> There is not better present to potential hackers than to offer your source
> code !
Security by Obscurity is no security at all - something you should
ponder.
> Hackers will exploit the bugs to the bone until someone reposts them !!!
<troll URL removed>
--
Rgds,
Chris MacKenzie
[EMAIL PROTECTED]
"I hear that if you play the NT 4.0 CD backwards, you get a satanic
message"
"--That's nothing. If you play it forward, it installs NT 4.0!"
------------------------------
From: "William" <[EMAIL PROTECTED]>
Subject: unresolved synbols
Date: 25 Feb 2000 12:21:51 GMT
Hi there,
I installed a add-on board to my Linux 2.2.12 environment.
There are so many unresolved symbols error reported when i try to
load the driver module of the board as follows.
/lib/modules/2.2.12-4/misc/xxx.o: unresolved symbol schedule
/lib/modules/2.2.12-4/misc/xxx.o: unresolved symbol securebits
/lib/modules/2.2.12-4/misc/xxx.o: unresolved symbol __wake_up
/lib/modules/2.2.12-4/misc/xxx.o: unresolved symbol jiffies
/lib/modules/2.2.12-4/misc/xxx.o: unresolved symbol printk
.
.
.
Please help me.
Thank you in advance.
------------------------------
From: [EMAIL PROTECTED] (Robert Grizzard)
Crossposted-To: alt.linux,alt.os.linux
Subject: Re: Here's why linux programs are so insecure !
Date: Fri, 25 Feb 2000 05:49:22 -0600
[EMAIL PROTECTED] wrote:
[same drivel as last time]
I didn't realize Microsoft had a sales office in France. Learn something
new every day.
------------------------------
From: cll <[EMAIL PROTECTED]>
Subject: Re: Problem with gimp
Date: Sat, 26 Feb 2000 01:54:43 +1300
The Gimp Manual mentions Guashe, but I have never seen it.
I use the Visual Schnauzer window of xv; or the Thumbnail view of kfm
instead.
cll
Reinhard Asmus wrote:
>
> Hallo everybody,
>
> I work under SuSE 6.3 with gimp. But my gimp don't have the function
> guash.
>
> What can I do ?
>
> Thanks Reinhard
------------------------------
From: Steve Gage <[EMAIL PROTECTED]>
Subject: Re: LILO patch for past 1024 cylinders?
Date: Fri, 25 Feb 2000 13:11:39 GMT
"Cameron L. Spitzer" wrote:
>
> In article <[EMAIL PROTECTED]>, Thomas Hommel wrote:
> >So, its not LILO who�s "buggy", its the BIOS. I�ve never seen any OS
> >booting from the area beyond cylinder 1024, except on a SCSI disk.
>
> Microsoft can do it. Windoze-98 can install with a pre-"int13 extensions"
> BIOS motherboard on a hard drive with a single partition of several
> thousand cylinders.
>
> I suspect they have the equivalent of an ioctl() that says,
> "See this file I've just opened for writing? Make sure all its blocks
> are written in the first 1023 cylinders."
> This would require a hack to the Linux IDE driver *and* Lilo
> *and* the ext2 FS.
> I'm surprised nobody's done it yet. Surely Red Hat could afford to
> hire someone to do it.
> There is already an ioctl(), that Lilo has always used, to tell you
> whether a particular block is in the first 1023 cylinders.
>
> >John Brock schrieb:
> >>
> >> In article <[EMAIL PROTECTED]>,
> >> Cameron L. Spitzer <[EMAIL PROTECTED]> wrote:
> >> [...]
> >>
> >> >Considering the size of modern disk drives, the fact is that Lilo
> >> >is no longer being maintained. The fact that it can't work
> >> >around the 1024 cylinder limitatation of the interface between
> >> >BIOS and the OS has become a bug. The workaround is to install
> >> >LILO someplace special, either on a floppy or on a tiny
> >> >/boot partition on the first cylinder of the drive.
> >> >The LILO mini-HOWTO shows how to make a Lilo floppy.
> >> >However, the fast version of the floppy still requires that the
> >> >bootimage reside below cylinder 1024.
> >>
> >> This makes me kind of nervous. LILO is the official Linux way of
> >> booting Linux, which makes it a key bit of software. It's essentially
> >> the doorknob that opens the door into the world of Linux. And it's "no
> >> longer being maintained?" This is bad PR, if nothing else.
>
> It sure is. It's the biggest obstacle most Linux newbies have to
> overcome. XF86Config is perhaps #2, or maybe sound.
Before you get all worried... I just downloaded and installed LILO v22
(dated Nov 99) that handles bootimages past the 1024 cylinder "limit".
As for it not being maintained, when you think about what it does, it
shouldn't require a lot of maintenance. And the source code is there for
anyone to pick up if/when it should be necessary (e.g., to address >1024
cylinders :-).
Check out http://rufus.w3.org/linux/RPM/LByName.html and you'll find
several packages available.
- Steve
------------------------------
From: "SyntaXer" <[EMAIL PROTECTED]>
Subject: Sending & Receiving Mail --> Local <--
Date: Fri, 25 Feb 2000 23:19:01 +1000
Hi,
I always get in trouble for not posting my specs, so I will get that out of
the way first ;)
Distro: Mandrake 7.0 Air
Hardware:
Chip: AMD K6-400 3D Now
Ram: 128
Video: AOpen AGP 3D Artist PA50 8mb
Sound: Sound Blaster PCI64
Network: AcerLAN ALN328 10/100 Base-TX Fast Ethernet
Modem: Banksia Wave SP56 V90
Additional Card: Voodoo2
I am trying to set up mail for several users on one box. The mail seems to
send fine and is in then placed in the sent folder, if I try to check for
incoming mail I receive the error message 'cannot open
file://var/spool/mail/username'
Which is understood when you check that folder it is empty. However the mail
folder in the users home directory has files in it. I have selected sendmail
and indicated its path as /usr/sbin/sendmail. (which is default)
When I connect to the net to retrieve mail however I select SMTP and when
checking my mail it just scrolls through several different settings (i
think) until it finds one it likes and downloads the mail. Can someone
please help me with this supposed simple process, Do I select sendmail or
SMTP? Do I have to create the files in the /var/spool/mail directory? Do I
need to use fetmailconfig? Do I point to the inbox file?
Thankyou
Scott
[EMAIL PROTECTED]
--
Microsoft gives you Windoze,
Linux gives you the whole friggin house!
--
Microsoft gives you Windoze,
Linux gives you the whole friggin house!
------------------------------
From: "Marc" <[EMAIL PROTECTED]>
Subject: Re: Boot loader
Date: Fri, 25 Feb 2000 14:26:31 +0100
Hi,
I'm not sure if I've understood your problem correctly but perhaps it helps.
I also boot into Linux using the NT boot loader and lilo is installed in the
root partition. The boot sector has to be copied to a partition NT can read
from, e.g. a fat partition using /bin/dd ... .
But as far as I know there is a limitation to lilo since the necessary files
for booting (in /boot) have to be on the first 1024 cylinders of your
hard-drive). If e.g. your linux root-partition starts after 5GB lilo can't
read them.
BR,
Marc
[EMAIL PROTECTED] schrieb in Nachricht
<8956sp$am8$[EMAIL PROTECTED]>...
>Re: Redhat 6.1
>
>Hello --
>
>I have been looking at ways to boot my machine in Linux from the NT boot
>loader. I have decided to install LILO on my root Linux partition rather
>than the master boot loader. And as it stands now, I can only boot in
>Linux if I boot from a bootable diskette. I have used bootpart on a DELL
>machine with a 4Gb disk before and it worked fine but when I tried it on
>my IBM 10Gb disk it failed saying it can't read my harddisk. I've heard
>somewhere else that I can use partition magic (PM) to do the same thing.
>Has anyone had any experience in using PM to do this? If so, could you
>please pass on any hints?
>
>-- Cheers, Sudhir
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: top in background
Date: Fri, 25 Feb 2000 13:31:54 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
>Try putting it in a script:
>
> while true; do
> top -i -b -n 1 >> top.out
> done
Thanks, but unfortunately that gives me an infinite number of 'top'
processes, all stopped but still taking up memory... (Old tops never
die, they just move to the rest home.)
The problem seems to be that when run in the background, top never
exits. Perhaps it has s.t. to do with not getting any input??
--
Mike Maxwell
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: top in background
Date: Fri, 25 Feb 2000 13:35:16 GMT
>The linux version might try to evaluate stdin which will
>stop the system if you run a command in the background. Try
>
> top -i -n 1 -b >file </dev/null &
>
>instead.
Thanks--unfortunately, that gives me the error msg
top: tcgetattr() failed: Inappropriate ioctl for device
I guess that's a clue to why it won't go anywhere when backgrounded:
it's trying to open input, and can't until it's foregrounded.
Unfortunately, I don't know enough about tcgetattr() to fix it (nor do
I really want to--I shouldn't have to, should I?).
>( while true; do date >>myfile ; ps --options >>myfile ; sleep 1 ;
done ) &
>
>Should work. No shell-script required to my knowledge - and it fits on
one
>line ;-)
Yes, I think I can do it that way (or a modification thereof, so it
doesn't go forever)--thanks!
A couple comments (OK, gripes :-) ) on these things:
The documentation for 'sleep' says the time arguments are "numbers", but
doesn't say what kind. Apparently they must be (positive) integers,
unlike 'free'. Fortunately, 'usleep' seems to work here, but why isn't
the documentation for 'sleep' (and 'vmstat') more specific?
The documentation for 'ps' implies you can mix and match args, but that
doesn't seem to be entirely true. For example, I can do 'ps -h' or
'ps -Cbash', but if I do both of them (in either order), I get
an "ERROR: Unsupported SysV option." Why? (Fortunately, I can
apparently get the same effect with 'ps --no-heading -Cbash'.) I'm
afraid I can't say that I'm impressed by Linux thus far :-(.
--
Mike Maxwell
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Subject: printing doesn't work, but it does using Samba
From: Bob Koss <[EMAIL PROTECTED]>
Date: Fri, 25 Feb 2000 13:57:31 GMT
This is a Redhat 5.1 (2.0.34) system with a HP LaserJet 4000
attached. The printer works fine if I boot the compter to windows. It
also works fine if I boot into Linux and print through samba from
another windows machine.
If I try to print a netscape page, or if I try to print a file with
lpr filename
all that happens is a blank page comes out.
I just downloaded and installed magic-filter, but that didn't
help. Here's my /etc/printcap:
lp:\
:sd=/var/spool/lpd/lp:\
:mx#0:\
:lp=/dev/lp1:\
:if=/usr/local/bin/ljet4-filter:
I don't know if that's the right filter or not, but I would think
something would be printed.
--
Robert Koss, Ph.D. | Object Mentor, Inc. | Tel: (800) 338-6716
Senior Consultant | 14619 N Somerset Cr | Fax: (847) 918-1023
[EMAIL PROTECTED] | Green Oaks IL 60048 | www.objectmentor.com
------------------------------
Date: Fri, 25 Feb 2000 08:58:42 -0500
From: Subba Rao <[EMAIL PROTECTED]>
Subject: Apache server mailing list
Reply-To: Subba Rao <[EMAIL PROTECTED]>
Here is a mailing list for the Apache server and it's modules.
To subscribe to the list
[EMAIL PROTECTED]
To unsubscribe from the list
[EMAIL PROTECTED]
To post to the list
[EMAIL PROTECTED]
Hopefully, this will provide another channel to discuss Apache related issues
and successes.
Subba Rao
[EMAIL PROTECTED]
http://pws.prserv.net/truemax/
=> Time is relative. Here is a new way to look at time. <=
http://www.smcinnovations.com
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: LILO upgrade for the new millenium
Date: Fri, 25 Feb 2000 13:53:18 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
Thanks, Crows (or should I call you Dances?)
but what you're saying is, I'm screwed if I think that I could bypass
this 'auto: file not found' error and I'd get to point B faster if I
start over and reinstall? Inother words, my existing linux is history?
Or this would be another way to boot into, and finally be able to access
my Mandrake?
>
> Make sure your floppy drive is working OK first.
>
> 1. Find a DOS/Win9x system (not hard, right?)
> 2. Using this system, insert a floppy, then enter "FORMAT A:" and "SYS A:"
> at the DOS prompt.
> 3. Copy the FDISK.EXE program to that floppy. In Win98, it's in
> C:\Windows\Command\
> 4. Go to your system, set the BIOS to boot from the floppy (and ONLY the
> floppy, if that's a choice) and insert that floppy you just made.
> 5. Save changes and exit the BIOS setup.
> 6. When MS-DOS starts up, enter FDISK /MBR. Voil�, no more LILO in the
> MBR.
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.linux,alt.os.linux
Subject: Re: Here's why linux programs are so insecure !
Date: Fri, 25 Feb 2000 13:59:29 GMT
In article <895ika$rm6$[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
> Because you dickheads offer the source code of your programs (!!!)
This is the ancient debate for "security through obscurity". Only it
doesn't really work that way. E.g., Microsoft doesn't offer their
sources, and there are still hundreds of discovered exploits for all
their programs. Outlook, IIS, SMB, etc, you name it, someone's
discovered bugs in it, and someone exploited them. E.g., whoever wrote
Melissa or BackOrifice probably didn't have the source to Windows or
Word or Outlook, but that didn't stop them, did it?
Or to put it otherwise, "security through obscurity" really boils down
to "oh, our back door is wide open and un-guarded, but that's OK,
because noone knows about it." Only some day, someone stumbles that
way. And if they're not the extremely honest sort, they'll just take
what they want and do what they want. And all the good that obscurity
did is: you won't even know they were there, or that you had a security
problem.
_Real_ security, on the other hand, means that even though someone
might know where your guards are, they STILL can't get inside. It also
means that the user has a right to know exactly what you're doing for
his/her security. And since it's his/her own property being guarded,
chances are they WILL tell you if there's some obvious oversight,
instead of abusing it.
And what makes "security through obscurity" actually be a big problem
is that the vast majority of people in the computer business are
blissfully clueless of what security even means. I wouldn't mind an
obscure implementation, if it were actually anywhere near secure, but
usually it's not. They'll sell you snake oil, and they won't even know
it's snake oil. They'll think rot13, or xor with a fixed sequence, or
their kid's decoder ring, are actually military grade encryption. Or
that having a HTTPS site is automatically secure... even though they
leave the whole passwords and credit card file accessible to anyone who
guesses where to point their browser. (Or what search to do on a search
engine.) Etc.
Without anyone checking their idea of security... how would you know
whether it's actually secure, or whether you bought snake oil?
---
Moraelin -- the proud member of the Idiots' Guild ;)
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Boot loader
Date: Fri, 25 Feb 2000 13:55:04 GMT
In article <8956sp$am8$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Re: Redhat 6.1
>
> Hello --
>
> I have been looking at ways to boot my machine in Linux from the NT
boot
> loader. I have decided to install LILO on my root Linux partition
rather
> than the master boot loader. And as it stands now, I can only boot in
> Linux if I boot from a bootable diskette. I have used bootpart on a
DELL
> machine with a 4Gb disk before and it worked fine but when I tried it
on
> my IBM 10Gb disk it failed saying it can't read my harddisk. I've
heard
> somewhere else that I can use partition magic (PM) to do the same
thing.
> Has anyone had any experience in using PM to do this? If so, could you
> please pass on any hints?
>
> -- Cheers, Sudhir
>
>
Hi,
I use bootmagic that came with PM4 to boot win98, RH linux and another
linux (some tweaking of my own) and it works great. It sees all
bootable partitions on both my drives (maxtor 20,4GB and quantum 6,4GB)
and haven't had any problems on booting. I've also had it running on a
machine at a previous job. This setup was WinNT4 and RH Linux. Had no
problems there either. It also looks 'better' on boot. It's graphical
with mouse support. Although that's not really important for a
bootloader/manager.
gr,
Erik.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: CharlieB <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,linux.redhat.list,linux.redhat.misc
Subject: Re: LILO boot problem
Date: Fri, 25 Feb 2000 08:07:39 -0600
I think 'cat' works o.k. too, but here's what I would try.
1. Start with 'cp bzImage /dev/fd0' and boot up on the floppy to test your new
kernel.
2. Once you're sure it's O.K., 'cp bzImage /bzImage'. Edit /etc/lilo.conf to
reflect the new name and run /sbin/lilo to select the new image. This way you
still have one last bail-out if your not happy. (And a boot floppy to help out).
3. If you're thrilled with everything and feel compelled to use vmlinuz, then 'cp
bzImage /vmlinuz', edit /etc/lilo.conf and run /sbin/lilo to install the new
image.
In any case, even with what your doing, if you replace /vmlinuz with another
/vmlinuz you will at least have to run /sbin/lilo in order to point to the correct
(physical) file location on the hard drive.
David Efflandt wrote:
> On Thu, 24 Feb 2000 00:15:50 -0800, Eric Marquez <[EMAIL PROTECTED]> wrote:
> >Here we go again. Now I'm not sure why this does not work but this
> >time I made a typescript file. For some reason my kernel just does
> >not want to load when I reboot. My old kernel will work. I have
> >attached the file to this can somebody take a look at it and tell me
> >where I screwed up. I am running redhat 6.1 with 128MB RAM, 10GB HD,
> >ASUS P5A Board
> >When I try to boot the new kernel that I have compiled, the system
> >just lockes and I have to do a hard reboot.
> >
> >
> >Script started on Wed Feb 23 23:54:10 2000
> (snip)
>
> >cat bzImage > /vmlinuz
>
> Isn't 'cat' mostly for text files? I would use 'cp' instead.
>
> --
> David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
> http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
> http://thunder.prohosting.com/~cv-elgin/ http://cgi-help.virtualave.net/
--
Thinking of running your critical apps on NT?
Isn't there enough world suffering?
Remove N O S P A M to reply
------------------------------
From: [EMAIL PROTECTED] (robin)
Subject: setting my display to 1024x769
Date: 25 Feb 2000 14:14:25 GMT
Hi,
I have a Red Hat Linux 5.xx.
My display gives me a 800x600 resolution, but I am sure my monitor and
video card can do better than that. How could I change my display properties
to get a 1024x769 resolution?
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: read -- - PartitionMagic error ( mandrake linux )
Date: Fri, 25 Feb 2000 14:09:46 GMT
In article <894h0t$iuc$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Here is my problem, please take the time to read it :
>
<snip>
> I made a boot disk on my other windows computer ( typing sys a: at
> dos prompt at other computer ). I then put the disk into the computer
> and it booted up into a dos prompt. People told me to type fdisk /mbr,
> which i did at both the a and the c prompt, but neither recognized it.
> I also typed win / win98 / windows, but that didn't work either.
> I can't get back into my computer ( windows ), so please try to
> solve my problem.
> THANKS A MILLION, me
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
You need the dos fdisk program also copied to your boot disk. It's in
c:\windows\command (or do a search for fdisk.exe). Then boot with the
disk and type the fdisk /mbr command at the a: prompt. It should work
then.
gr,
erik.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Steve Cowles" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.hardware,comp.os.linux.setup
Subject: Re: Please advice: unable to create file/directory
Date: Fri, 25 Feb 2000 14:42:48 GMT
"H K Chan" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hi,
>
> I am running RedHat Linux release 6.1, I have encountered a wierd
> problem that is I, as root, unable to create file or directory in a
> partition which has plentiful of disk space. Below is the situation:
>
> [root@it /home1]# touch temp
> touch: temp: No space left on device
> [root@it /home1]# mkdir temp
> mkdir: cannot make directory `temp': No space left on device
> [root@it /home1]# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/sda1 5447436 913304 4257416 18% /
> /dev/sdb1 6048500 513904 5227348 9% /home
> /dev/sda5 3099260 300176 2767600 10% /home1
> [root@it /home1]#
>
> Thanks in advance.
> Gerald
>
hmm... the only time I've ever seen this is when the inode table for the
filesystem is full. e.g. bytes-to-inode ratio. (man mke2fs) I had this
happen when I setup a news server. I got "No space left on device" at 22%.
The news server had created over 500,000+ files. I had to re-create the file
system and change the byte-to-inode ratio so that 5,000,000+ files
(inodes)could be created on the file system.
Also run: dumpe2fs /dev/sdb1
and check for amount of remaining inodes.
Steve Cowles
------------------------------
From: "Chucks" <[EMAIL PROTECTED]>
Subject: kernel panic - please translate/advise
Date: Fri, 25 Feb 2000 09:47:31 -0500
"kernel panic: VFS: unable to mount root fs on 03:01"
this occurred on first reboot of new install w/ OpenLinux 2.3
is this a recoverable condition, without reinstalling?
how?
------------------------------
** 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.misc) 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-Misc Digest
******************************
