Pages

Search

Promiscuous BOOTP server? - Forums Linux

Promiscuous BOOTP server? - Forums Linux

Promiscuous BOOTP server?

Posted: 09 Jan 2008 02:26 AM PST

On Wed, 09 Jan 2008, in the Usenet newsgroup comp.os.linux.setup, in article
<sadnet>, Magnate wrote:
 

Well, I don't think the bootfile for a Sun SparcStation5 is going to
work very well on your NCD X terminal, never mind that Intel box.
 

I don't use DHCP, never mind BOOTP, but I don't believe DHCP _needs_ the
MAC address by default - certainly there are enough people asking how to
make it match MAC to IP is such a hint. My understanding is that most
doents do suggest using it, but that's to reduce the horrendous
security hole of handing out IPs to anyone who asks.

Have you run through the DHCP mini-howto included in most Linux installs?

-rw-rw-r-- 1 gferg ldp 33678 Oct 20 2000 DHCP

Old guy

Can't boot Fedora Release 8

Posted: 08 Jan 2008 04:26 AM PST

On Wed, 09 Jan 2008 04:37:14 -0800, sbossert wrote:
 

This doc will be of immense help.

http://www.mjmwired.net/resources/mjm-fedora-f8.html

Stef

dpkg-reconfigure exim4-config: please help re interfaces to listen on

Posted: 08 Jan 2008 04:09 AM PST

"Bill Mar" <net> wrote in message
news:localnet... 

Doh! Of course, that's it. Thanks Bill - I'm now forwarding port X to port
25 on my firewall, and pointing Outlook to port X. It's working perfectly -
thanks again.

CC

Advanced LILO question: moving drive to a different computer

Posted: 06 Jan 2008 09:34 PM PST

rahul.net (Edward A. Falk) wrote:
 

Usually the boot drive has BIOS code 0x80



Florian
--
<http://www.florian-diesch.de/>
-----------------------------------------------------------------------
** Hi! I'm a signature virus! Copy me into your signature, please! **
-----------------------------------------------------------------------

Messed Up Hard Drive

Posted: 06 Jan 2008 03:37 PM PST

LenBum wrote: 


It sounds like the drive died. Having installed thousands of drives over
the last 20+ years I can tell you that some fail almost immediately. If
the BIOS says there is no drive I would return it and get a new one. I
like Hitachi...

-Philip

clock on news server

Posted: 06 Jan 2008 08:04 AM PST

Unruh <ubc.ca> wrote:
 
can 

Sure. Any idea why leafnode is complaining that they are more than 10 mins
adrift?

--
Richard Kimber
http://www.psr.keele.ac.uk/

USB devices disappearing from LVM

Posted: 06 Jan 2008 02:44 AM PST

On 8 Jan, 04:21, Matt Giwer <REMover.rr.com> wrote: 

Ahh. No. I'm replacing specific big chunks of space, such as /var/ftp/
pub and /var/lib/mock. I'm successfully mounting them, and am able to
mount them from /etc/fstab and umount them. That's not difficult.

The difficulty is when I reboot: I start getting errors about not
being able to detect the drives, even if they are marked "noauto", and
they show up as "inactive" in LVM. This may be what I get for putting
LVM partitions instead of direct partitions on there, but there were
reasons to do so.

I suspect that you are not using LVM?

no redhat kernel-source after building

Posted: 04 Jan 2008 04:11 PM PST

On Jan 5, 10:50 pm, The Derfer <com> wrote: 


Am I right about this or is at all bundled in devel now?
Is this step the one that builds the kernel-source:?

rpmbuild --rebuild kernel-2.4.21-52.EL.src.rpm --target noarch

hwclock problem with leapseconds - posix?

Posted: 01 Jan 2008 10:52 AM PST

<de> wrote: 

 

Yes, hwclock --utc --hctosys is ok but hwclock ---utc -systohc is broken.

hwclock --utc --hctosys runs this code in hwclock.c (note: --utc sets
variable universal):

/* We use the C library function mktime(), but since it only works on
local time zone input, we may have to fake it out by temporarily
changing the local time zone to UTC.
*/
zone = getenv("TZ"); /* remember original time zone */
if (universal) {
/* Set timezone to UTC */
setenv("TZ", "", TRUE);
/* Note: tzset() gets called implicitly by the time code, but only the
first time. When changing the environment variable, better call
tzset() explicitly.
*/
tzset();
}

But hwclock ---utc -systohc does not have the same code. It does this
instead:

if (universal)
new_broken_time = *gmtime(&newtime);
else
new_broken_time = *localtime(&newtime);

gmtime() does not use the timezone files at all but just thinks the system
time is counted at utc. Change the four lines above to something similar to
the --hctosys case and it should work ok. So even when using UTC, the
timezone files will be consulted.

Try the patch that follows... (There are also additional sleeps for about
half a second to fix resetting RTC. Without them a 'hwclock --systohc;
hwclock --hctosys' sequence causes an error of about 0.5 s in system time.
The question is whether the RTC time is really set to an exact second or an
exact half second when it is set.)

--- hwclock.c~ 2004-12-15 21:13:48.000000000 +0200
+++ hwclock.c 2008-01-10 13:38:11.000000000 +0200
@@ -455,8 +455,12 @@
*/

if (universal)
- new_broken_time = *gmtime(&newtime);
- else
+ {
+ /* Set timezone to UTC */
+ setenv("TZ", "", TRUE);
+ tzset();
+ }
+
new_broken_time = *localtime(&newtime);

if (debug)
--- cmos.c~ 2004-12-15 22:07:54.000000000 +0200
+++ cmos.c 2008-01-06 21:19:02.000000000 +0200
@@ -568,6 +568,7 @@
static int
set_hardware_clock_cmos(const struct tm *new_broken_time) {

+ usleep((unsigned long)430000);
hclock_set_time(new_broken_time);
return 0;
}
--- util-linux-2.12r/util-linux-2.12r/hwclock/rtc.c~ 2004-12-15 22:21:48.000000000 +0200
+++ util-linux-2.12r/util-linux-2.12r/hwclock/rtc.c 2008-01-06 21:24:34.000000000 +0200
@@ -302,6 +302,8 @@
int rtc_fd;
char *ioctlname;

+ usleep((unsigned long)430000);
+
rtc_fd = open_rtc_or_exit();

#ifdef __sparc__

Linux reboots right after Grug runs initrd

Posted: 30 Dec 2007 11:57 PM PST

Gilles Ganault <com> did eloquently scribble: 
 
 

An alternative would be to see out an antique friendly distribution.
(not all distributions are compiled for i686, some are still compiled for
i386)
--
__________________________________________________ ____________________________
| co.uk | |
|Andrew Halliwell BSc(hons)| "The day Microsoft makes something that doesn't |
| in | is probably the day they start making |
| Computer science | vacuum cleaners" - Ernst Jan Plugge |
------------------------------------------------------------------------------

What package will restore my application help files

Posted: 29 Dec 2007 11:14 PM PST

On Dec 30, 8:58 am, tek <com> wrote: 

By golly I stumbled on the solution. I first had to figure what
command was being executed to start the GNOME Help window. After some
digging I discovered it was /usr/bin/gnome-help which is actually a
link to /usr/bin/yelp. When I ran the command I received errors
pointing to the following missing shared objects.

lrwxrwxrwx 1 root root 43 Dec 30 11:47 libgtkembedmoz.so -> /
usr/lib/firefox-2.0.0.10/libgtkembedmoz.so*
lrwxrwxrwx 1 root root 37 Dec 30 11:49 libxpcom.so -> /usr/lib/
firefox-2.0.0.10/libxpcom.so*
lrwxrwxrwx 1 root root 42 Dec 30 11:49 libxpcom_core.so -> /
usr/lib/firefox-2.0.0.10/libxpcom_core.so*

So after recreating the links in /usr/lib, I ran gnome-help again and
received this error.

Yelp-ERROR **: Could not initialize gecko!
aborting...

Next I updated the yelp package to yelp.i386 0:2.18.1-8.fc7. After the
update everything was back to normal. Happy day! I wish I could
remember what I did to get things so out of whack.

kernel panic: no init found

Posted: 26 Dec 2007 08:33 PM PST

In comp.os.linux.setup, Jiancong
<com>
wrote
on Wed, 26 Dec 2007 20:59:03 -0800 (PST)
<googlegroups.com>: 

That, as it turns out, is the correct fix. The *boot* volume (/boot in
your /etc/fstab) is where the kernel resides, but the *root* is the root
of the entire directory tree. Of course /etc/init doesn't exist on
/dev/hdc7, which is what linux was trying to look for with root=/dev/hdc7.

You might also see

kernel /vmlinuz-2.4.20 root=/dev/ram real_root=/dev/hdc12 ramdisk_size=12000
initrd /initramfs-2.4.20

or some such in your instructions. In this case root=
is referring to a ramdisk, which has scripts to pick up
the real root later. The distinctions are a bit subtle.

--
#191, net
Useless C/C++ Programming Idea #2239120:
void f(char *p) {char *q = p; strcpy(p,q); }

--
Posted via a free Usenet account from http://www.teranews.com

locating and installing packages

Posted: 26 Dec 2007 11:25 AM PST

 

it should show up in your uname -a. Unless you are using more than 16g
of memory ( or certain types of processors) you wouldnt be needing the
hugemem kernel.

Partitioning SATA disk

Posted: 26 Dec 2007 04:01 AM PST

Dave Uhring wrote: 

Well a program kike

fred(int a, int b)
{
return a*b;
}


That will compile OK.

Won't run mind you.

But not all compiled programs are there to run on any given set of
hardware and OS.


 

Actually for practical application purposes they are.

with the sole exception of when you are bootstrapping a kernel, when you
are expecting to compile any libraries, and they are associated with the
source files of such.

Headers go with source files or with precompiled source files that are
not yet linked = libraries.

The compiler, the kernel development source and ITS headers and the OS
libraries and THEIR headers are not a one package thing: I can think of
many examples where you might want the compiler alone. Or the compiler
and the libraries-headers,but not the kernel source, or any combination.

 

Well I think compilers without header files are perfectly FINE.

*Libraries* without them are effing useless though.

Booting from PCMCIA CompactFlash

Posted: 25 Dec 2007 10:45 AM PST

On Thu, 27 Dec 2007 13:37:27 -0500, Chick Tower wrote:
 

Well, you could just try the install and see what happens. Worse that
can happen is it's not usable. Just don't pick KDE, GNOME or XFCE for
the GUI environment. Use just a window manager like Fluxbox or IceWM to
keep the GUI overhead low.

FWIW, I ran Mandrake 7.0, circa 2001, with KDE on a 166MHz machine with
just 64MB of RAM, 256MB swap, and it worked fine, quite usable. However,
it was a lean, no frills system: Nothing running that didn't absolutely
have to be. However, I know running a contemporary distro on that same
system today would probably not work or, if it did, would be too slow due
to swapping to be usable.

Stef