Linux-Misc Digest #303, Volume #24 Fri, 28 Apr 00 14:13:03 EDT
Contents:
Re: how to change system time?? (Lew Pitcher)
Re: Problem adding user using userconf in RH 6.1 ([EMAIL PROTECTED])
Re: IPCHAINS Question (John McKown)
Re: Linux Journal Python supplement (Albert Hopkins)
Re: Linux and SCO ("Brian E. Seppanen")
Re: Booting DOS6.1 on a second IDE drive (Vilmos Soti)
LRP distribution problems ("Tim Hicks")
Re: Mail filtering for the Linux console ("Brian E. Seppanen")
Network Problem!! BRANDnewbie (Noah Kuhn)
Disabling the boot logo ([EMAIL PROTECTED])
Netscape/KDE lockups Corrupted files? (Mike Pepera)
Re: Boot Problems (steve)
Re: convert gdbm files to ASCII,... (Jeff Workman)
Re: IPCHAINS Question (Randall Lee Mackie)
Re: Logical Volume Manager for Linux? (Jeff Workman)
Re: Starting xfs (Dave Rolfe)
Re: Script Telnet Sessions (Jeff Workman)
Re: Can't exit gdm in Slack 7 (Jeff Workman)
Re: Floppy format problem. (Harold Bower)
Re: OT -FS:Toshiba SatPro 48megs w/ZipDrive $480 & FDC settlement of $100! (M.
Buchenrieder)
Re: Should I ignore bootps/bootpc packets? (Thomas Boggs)
Re: Linux newbie + Web server (Will)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: how to change system time??
Reply-To: [EMAIL PROTECTED]
Date: Fri, 28 Apr 2000 15:58:04 GMT
On 28 Apr 2000 13:58:40 GMT, <[EMAIL PROTECTED]> wrote:
>when I use date, time, xclock, it always show the wrong time.
>I wonder how I can correct it.
>
>I had used "date" command to change it, it works well. But
>after you restart the system, everything go wrong again.
/sbin/clock or /sbin/hwclock are used to update or read the CMOS
clock. The system clock gets it's original value from the CMOS clock.
Is your CMOS clock off? If not, then perhaps you have a timezone issue
(either your timezone is mis-set, or your CMOS to system clock load
doesn't know about the timezone your CMOS is running in).
Lew Pitcher
System Consultant
Toronto Dominion Financial Group
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Problem adding user using userconf in RH 6.1
Date: Fri, 28 Apr 2000 15:50:01 GMT
In article <[EMAIL PROTECTED]>,
Eric Goforth <[EMAIL PROTECTED]> wrote:
> Hello all,
>
> I'm having problems adding a user to my RH 6.1 system using userconf
> and I can't figure out what I could be doing wrong.
>
> On the Normal tab I pick User Accounts. I then press on the Add
button
> and fill out the Login name and Full name fields. Is there any upper
> limit on the length of these? The login name field that I enter has
> fourteen characters, and the Full name field has nineteen
characters.
> I then click on the Accept button and get a Changing password dialog
> box. For the New UNIX password, I enter a ten digit password and
press
> Accept. I then get a changing password dialog box, and I enter the
same
> ten digit password in the Retype new UNIX password field and press
Accept
> again. I then get a message that "Password was not changed." I've
also
> tried adding this user to the users group, when I attempt to create
his
> account, but get the same message. This user's name is already in my
> dhcpd.conf file and my smbusers files, if that matters. I've
previously
> been able to add users with userconf, any ideas?
>
> Thanks,
> Eric
> --
> Eric Goforth
> Senior Applications Programmer
> SimTek, Inc.
>
Hi,
Loginname not longer than 8 characters will help. If you use a shadow
passwd file, passwords can be very long (don't know the limit), if not,
then also not longer than 8 characters.
So try a shorter loginname. It will probably help a lot.
gr,
Erik.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED] (John McKown)
Subject: Re: IPCHAINS Question
Date: Fri, 28 Apr 2000 06:52:01 -0500
On Mon, 24 Apr 2000 23:30:07 GMT, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
]>I'm a newbie. I am confused and need help on IPCHAINS. Are these the only
]>rules I need to have for my configuration?
]>
]>Your help will be greatly appreciated.
]>
]>
]>
]> 192.168.1.0/24
]> |
]> | ------------
]> | eth0 | |ppp0
]> |<------------>| LINUX |<----------->
]> | 192.168.1.1| PC |10.10.40.202
]> | | | (DHCP)
]> | ------------
]>
]>
]>
]>ipchains -P forward DENY
]>ipchains -P forward -i ppp0 -j MASQ -s 192.168.1.0/24 -d 0.0.0.0/0
]>
]>
I'm not sure. I have something similiar:
ipchains -P forward DENY
ipchains -A forward -s 192.168.1.0/24 -j MASQ
Note that you need the -A on the second ipchains, not a -P (may be a typo).
I don't know if you need the "-i ppp0" and "-d 0.0.0.0/0". I don't have them
and I know that mine works. I've been using it for about 3 months now with
no problems.
Somewhat off-topic, but I will mention that you really need to do some
"firewalling" with ipchain commands. Otherwise, any server on your LINUX
box will be visible on the internet. I just thought that I'd mention it
since you said you are a newbie, and you might have overlooked this.
If so, there are a lot of messages here on the ng about how to setup
a decent firewall on Linux using ipchains.
John
------------------------------
From: [EMAIL PROTECTED] (Albert Hopkins)
Subject: Re: Linux Journal Python supplement
Date: 28 Apr 2000 12:07:56 EDT
Reply-To: [EMAIL PROTECTED]
On 28 Apr 2000 01:56:37 GMT, A.M. Kuchling <[EMAIL PROTECTED]> wrote:
>I've seen no discussion of the Python supplement included with this
>month's Linux journal, either in comp.lang.python or in any of the
>Linux groups. What did people think of it?
Well, I haven't really looked at it yet, but I know a lot of non Linux/Python
people commented about the cover, which I was rather ashamed of.
--
Albert Hopkins
Sr. Systems Specialist
Dynacare Inc.
[EMAIL PROTECTED]
------------------------------
From: "Brian E. Seppanen" <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.sco.misc
Subject: Re: Linux and SCO
Date: Fri, 28 Apr 2000 12:32:57 -0400
On Thu, 27 Apr 2000, Wolf Grossi wrote:
> Try to add the path of libXm.so to LD_LIBRARY_PATH (shared objects library).
> Wolf
>
Would this need to be on the SCO server? On my linux machine libXm.so is
in /usr/X11R6/lib which is in my library search path, so my understanding
is putting it in LD_LIBRARY_PATH is overkill. If it is to be on the SCO
server: Does SCO understand the LD_LIBRARY_PATH environment variable?
/udk/usr/X/lib/WXm/libXm.so
/udk/usr/X/lib/libXm.so
/udk/usr/lib/libXm.so
Whether this is within the library path is up for debate considering I'm
no SCO guru. It doesn't have an /etc/ld.so.conf, which would indicate the
library search path on linux.
I'd appreciate any further assistance.
thanks,
Brian Seppanen
[EMAIL PROTECTED]
------------------------------
Subject: Re: Booting DOS6.1 on a second IDE drive
From: Vilmos Soti <[EMAIL PROTECTED]>
Date: Fri, 28 Apr 2000 16:28:20 GMT
Nick van Stigt <[EMAIL PROTECTED]> writes:
> I have just completed a partition shuffle after buying a new hd. Now
> that the dust has settled, I have 2 DOS6.1 partitions on the second IDE
> drive (hdb1 and hdb2).
IRC DOS has to be on the first drive.
Vilmos
------------------------------
From: "Tim Hicks" <[EMAIL PROTECTED]>
Subject: LRP distribution problems
Date: Fri, 28 Apr 2000 17:36:43 +0100
I am trying to setup my 486 to run linux in order to have NAT/Firewall
service for my lan that is connected to the net by cable modem. The 486
does not have a cd drive, so installing RH6.1 from the cd I have is not an
option. I tried to do an ftp install using the cd and one of my win98se
computers running an ftp server, but (for a reason still to be found!) that
failed...
So I found myself with the LRP distribution. I have been able to create an
LRP disk and boot to linux just fine. However, I am having real trouble
getting my 3com 3c509B TPO card detected. I have used the 3com dos utility
to disable the PnP feature on the card, and I made a note of the io (240)
and the irq (10) of the card. I created my own module.lrp file using the
utility on www.linuxrouter.org and copied this to disk. Having read
numerous faq's, howto's etc, I think that I need to tell linux what the io
and irq is for the card, but I don't know which file to put this into or
*what format to type it in*. The howto implies that I should edit
etc/conf.modules, only lrp does not have this file as far as I can tell. I
tried editing etc/modules but I don't really know actually what form this
should take.
If anyone can provide any pointers, I'd really appreciate it.
tim
p.s. This is what my etc/modules file looks like
# Generated by modules.lrp creator on www.linuxrouter.org
ext2
vfat
smbfs
ipip
ip_masq_ftp
ip_masq_irc
ip_masq_raudio
ip_masq_cuseeme
ip_masq_vdolive
ip_masq_quake
ip_alias
router
serial
plip
3c509
slhc
ppp
slip
------------------------------
From: "Brian E. Seppanen" <[EMAIL PROTECTED]>
Subject: Re: Mail filtering for the Linux console
Date: Fri, 28 Apr 2000 12:46:13 -0400
On 23 Apr 2000, Bryan Hoyt wrote:
> I want to be able to filter mail into separate folders, like you can in
> Netscape Messenger, but with a command from the Linux console. Is there
> a way to do this using common programs like fetchmail, mail, pine, and
> others with bash? Any help would be very nice indeed.
> Thanks.
I do that myself with procmail, and I also use junkbuster to catch
spam (junkbuster is nothing but preformulated procmail recipes that
effectively filter spam). The procmail man page is actually helpful
compared to some of the other man pages I've seen.
HTH
Brian Seppanen
[EMAIL PROTECTED]
------------------------------
From: Noah Kuhn <[EMAIL PROTECTED]>
Subject: Network Problem!! BRANDnewbie
Date: Fri, 28 Apr 2000 12:38:39 -0400
Hello!!
I'm a Linux newbie running Mandrake 7.0 and am trying to set up my
internet connection through my school's network. I have setup the
settings such that when I open Netscape and go to my school's website,
everything works fine, pages load like twice to 3 times faster than in
windows so i'm very excited, the problem is that I cannot access pages
beyone my internal www. I get an error that the Server is Unreachable
and there is a TCP error. I was told there could be many reasons for
this (firewall, invisible proxy etc.) However, no one I spoke to in irc
could figure out my problem. If anyone has ever seen this problem
before PLEASE help. I may even be putting the wrong values in the DNS
setup in Mandrake. I took the values from my winipcfg in win98 and put
them in where I thought they went. Anyway, any response would be very
appreciated, i don't want to have to give up on Linux just because of
this. Thank You!!
Noah
------------------------------
From: [EMAIL PROTECTED]
Subject: Disabling the boot logo
Date: Fri, 28 Apr 2000 16:47:31 GMT
Hi everyone,
I run the VESA framebuffer in my laptop but unfortunately the boot logo
causes my screen to resize; I grep'ed the source tree and looked at some
files but I am just a hobbyist programmer ... :-) Can anyone give me any
tips on how to disable the boot logo ? Is there any #define switch I can
use?
Thank you for any tips!
bye,
-joane
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Mike Pepera <[EMAIL PROTECTED]>
Subject: Netscape/KDE lockups Corrupted files?
Date: Fri, 28 Apr 2000 16:54:51 GMT
Hello all,
I am a relatively new Linux user. I have been using the Red Hat 6.2
distribution with the K desktop Environment.
I have been having problems with Netscape locking up, and I am unable
to use the kill command to stop the process. The rest of my
applications, including X-Windows seem to work fine. However, after
waiting 10-15 minutes, and Netscape is still sitting open, but not
functioning, I try to logout of X to go back to the command line, at
this point the whole system seems to hang. After waiting another ten to
fifteen minutes, I hit the power switch and do a forced reboot. I know
that this is not a Kosher thing to do, but I saw no other alternative.
Most of my problems seem to have started when my cousin, a Windoze
user, was using my machine and he had initialized over forty sessions
of netscape which quite effectively stopped my machine. The only way I
could do anything, was to reboot the machine.
My computer is on a 100Mb LAN with my mother's Windows machine serving
as an internet gateway using WinRoute, a DHCP/DNS server. My sister and
her friend often use the Windows machine to access the internet at the
same time as I am accessing it from my machine. Do to their complete
lack of computing knowledge, and despite my repeated pleas, they keep
shutting off the server software, killing my internet connection, and
other activities which lock up the Windows machine( This machine only
locks up for them, no one else in the family!) Could the repeated loss
of the server on the network cause Netscape to lock up?
In addition to these problems, I have had problems unmounting the file
systems of my computer when rebooting the machine.
Could my repeated destructive reboots of the computer have caused
corruption of various files?
Thanks,
Mike Pepera
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: steve <[EMAIL PROTECTED]>
Subject: Re: Boot Problems
Date: Fri, 28 Apr 2000 11:04:03 -0600
John Battle wrote:
> I vrecently installed a new IDE drive on my Dell WS-400 Workstation
> shich was set up to dual boot Linus/NT on its original 4-Gig SCSI
> drive. According to Dell, there is no way to make the BIOS boot from
> the SCSI if the IDE drive is turned on. I prefer not to re-install
> Windows NT or Linux, since everything is working good now on both. I
> tried copying the entire NT partition (using Partition Magic) to the IDE
> drive, but when I try to boot from it now I get the Blus Screen of Death
> if I choose NT and the good old LI... when I try to boot Linux. Any
> ideas (besides reinstalling) Help!!!!!!
I assume you used LILO with the scsi drive, correct? Did you have a
boot-disk
for your scsi system? If so, you need to install LILO on the MBR of the IDE
drive.
Something like:
boot=/dev/hda
root=/dev/sda1
There is no reason that you cannot boot off the IDE drive and mount the
SCSI drive as
root, so you don't need to move any partitions around. You don't need to
mount the IDE
drive at all, if you don't want to. I've had systems with just this
configuration for years.
------------------------------
Subject: Re: convert gdbm files to ASCII,...
From: Jeff Workman <[EMAIL PROTECTED]>
Date: 28 Apr 2000 12:16:47 -0400
"Hanno Schulz" <[EMAIL PROTECTED]> writes:
> Hi!
>
> Can someone tell me the location of an pogramm that convert gdbm files to
> ascii or an other usable database format.
> Thank you.
>
> Hanno
>
#!/usr/bin/perl
use NDBM_File;
use POSIX;
tie(%DBM, 'NDBM_File','/path/to/dbm/file', O_RDONLY, 0644) or die "Cannot
open DBM file $!\n";
while(($key,$value) = each %DBM) {
print "$key\t$value\n";
}
untie(%DBM);
--
Jeff Workman | [End of diatribe. We now return you to your
UNIX System Administrator | regularly scheduled programming...]
Gibralter Publishing |
(910) 455-6446 ext. 3034 | -- Larry Wall, in "Configure" from the
http://www.gibralter.com | perl distribution.
------------------------------
From: Randall Lee Mackie <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: IPCHAINS Question
Date: Fri, 28 Apr 2000 10:22:43 -0700
The easiest thing to do is to use one of the firewalling
utilities that asks you questions about your set up
and then writes the rules. I have use pmfirewall:
http://www.pointman.org/pmfirewall/
This writes rules based on answers to questions it
asks you. You can then either start pmfirewall manually
or have it started automatically, depending on your setup.
If you are using a static IP connection, it can be
started at boot-time. If you have a ppp connection,
you can have it started each time the ppp connection
is brought up by adding the following line to the
/etc/ppp/ip-up.local file (making sure it has executable
permissions):
#!/bin/sh
#
# start pmfirewall
#
sh /usr/local/pmfirewall/pmfirewall start
Using pmfirewall, I've successfully set up the firewall
and IP masquerading for a Windows 98 PC I have connected
to my system. It works very well.
Randy
------------------------------
Crossposted-To:
alt.os.linux,comp.os.linux.development.apps,linux.redhat.install,linux.redhat.misc
Subject: Re: Logical Volume Manager for Linux?
From: Jeff Workman <[EMAIL PROTECTED]>
Date: 28 Apr 2000 12:24:08 -0400
[EMAIL PROTECTED] (M Sweger) writes:
> Any free Veritas equivalent? Need to have choices! :)
Of course it won't be free, but I've heard rumors that Veritas is going to be
porting their products to Linux soon.
Jeff
--
Jeff Workman | [End of diatribe. We now return you to your
UNIX System Administrator | regularly scheduled programming...]
Gibralter Publishing |
(910) 455-6446 ext. 3034 | -- Larry Wall, in "Configure" from the
http://www.gibralter.com | perl distribution.
------------------------------
From: Dave Rolfe <[EMAIL PROTECTED]>
Subject: Re: Starting xfs
Date: Fri, 28 Apr 2000 13:20:59 -0400
Patrik Vahtras wrote:
> On Tue, 25 Apr 2000, David Rolfe wrote:
>
> >
> > Hal,
> > It is working. I know about the font paths in XF86Config. And I know about X
>looking
> > for xfs. And it is a little strange that the ONLY invocation in XF86Config that
> > allows X to find xfs is unix/:-1. Any reference to port 7100 just fails. I don't
> > know why. Maybe it is a bug in X. But my stuff is definitely working now. I have
>the
> > kde font manager up right now and am looking at the ms verdana font. So why does it
> > work? Who knows?
> >
> > Dave
> >
> It's a RedHat thing
> http://www.redhat.com/knowledgebase/newfontsystem/
As a final parting word ... I got tired of fooling around with RedHat 6.0 and installed
Mandrake 7.0 ... Very Nice. Very Very Nice. No more fooling around and my sound card...
works and my graphics adapter works without any un-natural acts. AND it come with LyX
(among other things).
Dave
------------------------------
Subject: Re: Script Telnet Sessions
From: Jeff Workman <[EMAIL PROTECTED]>
Date: 28 Apr 2000 12:30:33 -0400
[EMAIL PROTECTED] (Peter Alliett) writes:
> I am wondering if there is a way to script the following or if someone
> has a better idea for automating this process.
>
> Essentially I telnet into 5 LINUX boxes, RH 6.1 to check the disk
> space on the drives.
>
> I am wondering how I can automate the process into 1 script file to
> telnet to the server specifing a username and password then issue the
> command df -k and redirect it to a file then close the connection and
> continue on to the next server and do the same thing and append the
> results of df -k.
Sounds like a job for expect (http://expect.nist.gov) or, if you do perl,
the Net::Telnet module (ftp://ftp.cpan.org).
Jeff
--
Jeff Workman | [End of diatribe. We now return you to your
UNIX System Administrator | regularly scheduled programming...]
Gibralter Publishing |
(910) 455-6446 ext. 3034 | -- Larry Wall, in "Configure" from the
http://www.gibralter.com | perl distribution.
------------------------------
Crossposted-To: comp.os.linux.setup
Subject: Re: Can't exit gdm in Slack 7
From: Jeff Workman <[EMAIL PROTECTED]>
Date: 28 Apr 2000 12:38:25 -0400
[EMAIL PROTECTED] (Ron Gibson) writes:
> I'd like the ability to use GDM to try out or chose one X window manager
> or another, as I have Gnome, Enlightenment, KDE, the whole ball of wax
> on Slack 7 installed, for various reasons.
>
> GDM has the ability to chose the manager you want but I can't figure a
> way to exit the login profile other than to open another terminal do
>
I'm not sure if you are talking about exiting your X session or exiting X
altogether. If you just want to exit your X session and be presented
with another GDM login, try pressing CTL-ALT-BACKSPACE. If you want to
get completely out of X and back to the console. Type /sbin/init 3 as
root.
Jeff
--
Jeff Workman | [End of diatribe. We now return you to your
UNIX System Administrator | regularly scheduled programming...]
Gibralter Publishing |
(910) 455-6446 ext. 3034 | -- Larry Wall, in "Configure" from the
http://www.gibralter.com | perl distribution.
------------------------------
From: Harold Bower <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: Floppy format problem.
Date: Fri, 28 Apr 2000 17:38:34 GMT
Charles Sullivan wrote:
>
> Harold Bower <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
[snip]
> > I believe that you may be confused by thinking that the 360K diskettes
> > are writted in a 1.2M drive by writing two adjacent tracks with the same
> > data...they are NOT. In a 1.2M drive, the drive steps twice for each
> > track increment, writing every other track. If it were not so, your
> > times for Linux would be more than twice those for DOS. If you sketch a
> > couple of tracks on a piece of paper, label one on an end as track 0,
> > and increment numbers for the adjacent 'tracks'. Mark a large swath
> > around track 0, 2, 4, etc. That would correspond to tracks 0, 1 and 2
> > on 360K. The narrower tracks will be reliable, but writing narrow and
> > reading wide (with a 'real' 360k drive) might not if the disk were
> > previously written with a 'real' 360k.
> >
> > If you can find some old 720K drives by TEAC, you will find a jumper
> > that actually forces a double-step the drive. It appears that they used
> > a common stepper and electronics between 40 and 80 track drives (at
> > least the ones I had in the mid 80s), and the technology was carried
> > forward. You can find assembly code for Z80/8080 systems which do the
> > double-stepping in CP/M archives as part of BIOS code for CP/M.
>
> You're right, the discussion here did lead me to believe that two adjacent
> tracks were written with the same data. But with the measured format times
> that seems not to be the case. However my experience and that of David C.
> was that 360K diskettes written on 1.2M drives on clone PCs available a few
> years after the introduction of the IBM-AT _were_ readable on MFM drives.
>
> I was told years ago that the later 1.2M drives had some sort of dual-head
> arrangement for write-compatibility with MFM drives. And to the best of my
> recollection I had no problem reading 360K diskettes on an older MFM-only
> system which were written on my circa-1990 clone PC (which I still have)
> and which was/is equipped with a Teac FD55GFR-142-U of the same vintage.
>
> My confusion now is that the data sheet for that old Teac FD55GFR-142-U
> says the same thing as the data sheet for a new Teac FD55GFR-7193-U,
> i.e., that 360K diskettes are read-only. So how did it write MFM
> compatible diskettes? The story about the dual heads certainly doesn't
> seem to apply to either of these drives.
The data sheet may only say that to protect their interests. For
practical purposes, I would read that as "..will *reliably* read 360K
diskettes.." I don't believe that they want to get into the position of
saying that you can write (narrow-track write), then have the disk
placed in a normal "wide-track" 360k (more appropriately a 40-track,
double-sided, double-dendity, MFM) drive, then write again with the
narrow track (on the 1.2M drive) and try to read yet again on a
"wide-track" drive. This is where the garbage residue on both sides of
the second "narrow" write did not obliterate the peripheral remnants of
the "wide" write, and the drive may issue errors.
In the CP/M 8-bit community, many of us have been reliably transferring
data between drives in just this fashion. I normally place a BIG label
on diskettes with narrow (80-track, but double-stepped to appear as
40-track) tracks, and insure that I never write (usually by adding a
write-protect tab) before placing the diskette in a "wide" 40-track
drive.
If you want to try this, I would advise you to first bulk-erase a
diskette, then format it in a 1.2M drive as a 360K diskette. The bulk
erase should remove existing format info and data down to a level where
it won't cause problems. I believe that you will find out that the
drive *will* allow writes, and be relatively reliable as long as you
understand what you are doing and exercise caution. Also remember that
most floppy drives use true stepper motors and rely on physical
positioning for track alignment. If you are shooting for reliable data
exchange between machines, you might want to check the head alignment (a
rather laborious procedure without special tools) if you experience
random errors, particularly on inner tracks.
Good Luck.
Hal
> Regards,
> Charles Sullivan
------------------------------
From: [EMAIL PROTECTED] (M. Buchenrieder)
Subject: Re: OT -FS:Toshiba SatPro 48megs w/ZipDrive $480 & FDC settlement of $100!
Date: Fri, 28 Apr 2000 14:27:03 GMT
[Note FollowUp-To: header]
[Posted & mailed]
Chris T <[EMAIL PROTECTED]> writes:
>I have my Toshiba Satellite Pro 430CDT with an active matrix LCD for
[drivel snipped]
Please stop that. This is not a for-sale newsgroup, and if you continue
posting off-topic ads, then you'll possibly lose your university account.
Michael
--
Michael Buchenrieder * [EMAIL PROTECTED] * http://www.muc.de/~mibu
Lumber Cartel Unit #456 (TINLC) & Official Netscum
Note: If you want me to send you email, don't munge your address.
------------------------------
From: Thomas Boggs <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.security
Subject: Re: Should I ignore bootps/bootpc packets?
Date: Fri, 28 Apr 2000 13:51:13 -0400
Manfred Bartz wrote:
>
> Thomas Boggs <[EMAIL PROTECTED]> writes:
>
> > I use dhclient to maintain an internet connection to my local network
> > via a cablemodem. I am using ipchains on my firewall machine. Every so
> > often, my firewall log shows blocked packets using the bootps and bootpc
> > protocols. So far I haven't experienced any problems due to blocking
> > these packets. Is there a reason why I should allow these packets from
> > the dhcp server to enter?
>
> With those ports blocked, your IP-address lease cannot be renewed and
> will eventually expire.
> You only need to have the ports open for UDP, e.g.:
>
> ipchains -I input -i $CABLE -p udp -s $DHCPHOST 67 --dport 68 -j ACCEPT -l
>
> the output chain is always open I assume.
>
> --
> Manfred
That makes sense to me but the packet source addresses aren't from my
dhcp server. Here are a couple of the log entries (which I should have
included in the original post):
Apr 26 21:56:19 gate kernel: Packet log: input DENY eth0 PROTO=17
10.28.214.1:67
255.255.255.255:68 L=328 S=0x00 I=53372 F=0x0000 T=255 (#4)
Apr 26 22:10:26 gate kernel: Packet log: input DENY eth0 PROTO=17
10.28.214.1:67
255.255.255.255:68 L=335 S=0x00 I=54366 F=0x0000 T=255 (#4)
10.28.214.1 is _not_ my dhcp server. Here are the ipchains rules I
currently have regarding dhcp:
BROADCAST_SRC="0.0.0.0" # broadcast source address
BROADCAST_DEST="255.255.255.255" # broadcast destination
address
<skipping....>
# DHCP client (67, 68)
# --------------------
# allow dhcp server (67) to connect to dhcp client (68)
# Note: the DHCP server is the only externel source of broadcast
# messages we should see, ever.
ipchains -A input -i $EXTERNAL_INTERFACE -p udp \
-s $DHCP_SERVER 67 \
-d $IPADDR 68 -j ACCEPT
ipchains -A output -i $EXTERNAL_INTERFACE -p udp \
-s $IPADDR 68 \
-d $DHCP_SERVER 67 -j ACCEPT
ipchains -A input -i $EXTERNAL_INTERFACE -p udp \
-s $DHCP_SERVER 67 \
-d $BROADCAST_DEST 68 -j ACCEPT
ipchains -A output -i $EXTERNAL_INTERFACE -p udp \
-s $BROADCAST_SRC 68 \
-d $DHCP_SERVER 67 -j ACCEPT
# Getting renumbered
ipchains -A input -i $EXTERNAL_INTERFACE -p udp \
-s $BROADCAST_SRC 67 \
-d $BROADCAST_DEST 68 -j ACCEPT
ipchains -A output -i $EXTERNAL_INTERFACE -p udp \
-s $BROADCAST_SRC 68 \
-d $BROADCAST_DEST 67 -j ACCEPT
# As a result of the above, we're supposed to change our IP address
with
# this message, which is addressed to our new address before the
dhcp
# client has received the update.
ipchains -A input -i $EXTERNAL_INTERFACE -p udp \
-s $DHCP_SERVER 67 \
--destination-port 68 -j ACCEPT
ipchains -A input -i $EXTERNAL_INTERFACE -p udp \
--source-port 67 \
-d $IPADDR 68 -j DENY -l
The only thing I can guess is that 10.28.214.1 is supposed to be the
value of BROADCAST_SRC. But I don't want to assume that if it might not
be true.
Thanks,
Thomas
------------------------------
From: Will <[EMAIL PROTECTED]>
Subject: Re: Linux newbie + Web server
Date: Fri, 28 Apr 2000 13:02:02 -0500
> Apache is THE Web server...
> www.apache.org
> --
> Alexis Bilodeau
> eMagiK Technologies
> 819.371.9273
> [EMAIL PROTECTED]
I've found Apache to be, well besides one of the better, very memory
intensive.. is there any way to reduce the amount of memory it requires??
I know if I recompile it, without all the extra modules, that will help, but
what else can I do?
Thanks
Will
------------------------------
** 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
******************************
