Linux-Networking Digest #970, Volume #9 Fri, 22 Jan 99 19:13:51 EST
Contents:
Re: Security hole with WU-FTPD (Duncan Simpson)
Re: diald problem (Matt Kressel)
PPP setup/minicom ("News")
Bad Sendmails with otherwise good OSes, Re: Open relays on DNS ("Cameron Spitzer")
Re: httpd (Joseph)
Re: Connect without hub (Criss Hyde)
ipfwadm <--> ipchains rules? (Brian)
Re: kernel: VFS: No free inodes - contact Linus (Matt Corddry)
Yikes!!! Bogons!!! (Tom Reinertson)
Re: DNS problem with IP Masq Gateway (David Moulton)
Re: pppd gurus - help, please! (Clifford Kite)
callin in to my machine... ([EMAIL PROTECTED])
how to setup a driver for sis6326 display card in RH5.2 Linux? ("Alexander")
Re: Bad Sendmails with otherwise good OSes, Re: Open relays on DNS (Ah Clem)
Ethernet address.... ([EMAIL PROTECTED])
No dial tone. PPP. Redhat5.2 (Kaustav Bhattacharya)
Re: Connect without hub (Criss Hyde)
PPP Script for RedHat Linux ("Killer")
Re: Q: Network Linux and OpenDOS ? (Richard Hector)
Re: Compiling ppp-2.3.5 under kernel 2.0.34 (Clifford Kite)
Re: Any program conver text file to g3 format file ? (Dirk Ruediger)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Duncan Simpson)
Crossposted-To:
comp.security,comp.security.unix,redhat.general,redhat.networking.general,aus.computers.linux
Subject: Re: Security hole with WU-FTPD
Date: 22 Jan 1999 20:31:08 GMT
In <[EMAIL PROTECTED]> [EMAIL PROTECTED] (Chris Cover) writes:
[stuff snipped]
>I've seen the same thing in RHL 5.1's mountd. The cracker we saw used it
>to gain root by using the Linux root kit available at rootshell.com. I say
>cracker, but he turned out to be more of a script kiddie. I think the flaw
>is described in CERT advisory CA-98.12.mountd. It's sort of a Pandora's
>box after they are in, because the root kit replaces several binaries in
>order to help avoid detection, such as find, inetd, login, passwd, syslog,
>and many more.
You can detect these with a number of programs. Try tripwire (the free
version or the non-free version) and check-ps (free software with
detials at http://checkps.alcom.co.uk). Combinations are quite
feasable and recommended.
The mountd buffer overrun is a known bug and I think an exploit and
fix information appeared to bugtraq (using the same shell code as all
the other exploits...). RH fixed the bug a few days after the report
on bugtraq.
Duncan (-:
------------------------------
From: Matt Kressel <[EMAIL PROTECTED]>
Subject: Re: diald problem
Date: Fri, 22 Jan 1999 18:25:25 GMT
Emmett Pate wrote:
>
> I've been using diald on RH5.1 for about 5 months now and have it
> basically working pretty well. However, I seem to have one recurring
> problem that I can't quite get solved. There seems to be instances
> that cause the link to stay up that I can't explain.
>
> I tracked down an earlier situation that seemed to be caused by not
> running 'routed'. By running 'routed', the diald standard.filter
> rules were reliably able to ignore 'route' requests. Now it appears
> that if I use a browser (NS4.5) I continue to see 'http' requests from
> a remote address *to* my address even after I exit the browser.
> Incidentally, this doesn't seem to happen if I use a browser on a
> Windows machine that uses the Linux box as a gateway. I haven't been
> able to find a specific site that will cause this. I usually just
> fire up Netscape and hit a couple of Yahoo categories, DejaNews, and
> the Netscape homepage. Then I exit the browser and watch the output
> of 'tcpdump' and /var/log/messages (with diald debugging level 31).
>
> I'm also using 'named' as a caching nameserver and will occasionally
> see ongoing ".domain" requests even after exiting the browser. They
> don't seem to keep the link up, however, without some other rule match
> (like tcp or www packets).
>
> Any help on interpreting the output of 'tcpdump' and/or the diald
> debugging output would be greatly appreciated.
Firstly, look at the /usr/lib/diald/standard.filter and see what rules
you have and configure them for your settings. Secondly, NS 4.5
occasionally leaves a stale process around after exiting that makes
these requests (this is a bug that I hope gets fixed :) ). If you see
the problem recurr, do a "ps" to see if the stale netscape(s) are
there. Usually they are in parentheses. Just kill them.
-Matt
--
Matthew O. Kressel | INTERNET: [EMAIL PROTECTED]
+--------- Northrop Grumman Corporation, Bethpage, NY ---------+
+--------- TEL: (516) 346-9101 FAX: (516) 346-9740 ------------+
------------------------------
From: "News" <[EMAIL PROTECTED]>
Subject: PPP setup/minicom
Date: Fri, 22 Jan 1999 15:27:33 -0500
Just installed linux 5.0 on dual boot machine. I am trying to get PPP
working using the following references: 10 min guide to ppp,(my pppd and
kernel version match) and Linux ppp help.
Needless to say I can't get it to work. My first problem is in minicom.
when I dial in to my isp(ibm.net) the modem will dial it will connect b/c i
see the welcome message but i get kicked off before reaching the login
prompts. no errors i just here my modem click and the welcome messages
stops. I set my modem timeout to max also did the same with minicom. No
help. I have tried setting up scripts using the referenced material but
they don't work either. so I figure I have to be able to log in from
minicom before I worry about scripts. Any help will be appreciated
------------------------------
From: "Cameron Spitzer" <[EMAIL PROTECTED]>
Crossposted-To: news.admin.net-abuse.email
Subject: Bad Sendmails with otherwise good OSes, Re: Open relays on DNS
Date: 22 Jan 1999 20:35:01 GMT
In article <0h0q2.528$[EMAIL PROTECTED]>,
McWebber <[EMAIL PROTECTED]> wrote:
>What is going on with these morons. It's amazing how much Spam I've gotten
>recently from admins with open relays on what's *supposed* to be their name
>server.
>
>dns1.esplanade3000.net (195.115.0.1)
>Received: from satellite (2Cust6.tnt6.lax3.da.uu.net [153.37.70.6])
> by esplanade3000.net (8.8.5/8.8.5) with SMTP id HAA09091;
> Fri, 22 Jan 1999 07:03:58 +0100 (CET)
Word has gotten out that you can get Linux on CD for US$3 and make a
reasonable DNS out of an old 486 that was good for nothing else,
freeing up more valuable computers for Web or email service.
Unfortunately, a lot of those $3 disks also install Sendmail 8.8
configured as an open relay. Red Hat 5.1 is one such product.
Word hasn't gotten out as widely about this problem.
Older Sun and SGI boxes have the same problem. People are "retiring"
them to DNS and leaving the broken Sendmails running.
Cameron
------------------------------
From: Joseph <[EMAIL PROTECTED]>
Subject: Re: httpd
Date: Fri, 22 Jan 1999 19:40:19 GMT
>From the Apache faq page
66.I'm using RedHat Linux and my .htm files are showing up as HTML
source rather than being formatted!
RedHat messed up and forgot to put a content type for .htm files
into /etc/mime.types. Edit /etc/mime.types, find the line containing
html and add
htm to it. Then restart your httpd server:
kill -HUP `cat /var/run/httpd.pid`
Then clear your browsers' caches. (Many browsers won't re-examine
the content type after they've reloaded a page.)
Unix Adm wrote:
>
> Hi,
>
> We start the Apache httpd daemon and it works fine
> except that it does not accept
> .htm
> (instead of normal .html)
> files which are created by a PC tool.
>
> Is there any configuration or command option
> which can solve this problem?
>
> Thanks,
>
> Arthur
> ======
------------------------------
From: Criss Hyde <[EMAIL PROTECTED]>
Crossposted-To:
comp.dcom.lans.ethernet,comp.sys.sun.admin,comp.os.ms-windows.networking.win95
Subject: Re: Connect without hub
Date: Thu, 21 Jan 1999 10:34:34 -0500
Nico Kadel-Garcia wrote:
> On 19 Jan 1999 23:40:39 GMT, Chris Cappuccio <[EMAIL PROTECTED]> wrote:
> >What about the situation where you are connecting together two machines
> >via RJ45 ? Don't ethernet cards with 10bT interfaces rely on the hub
> >for collision detection? If a hub detects a collision, it sends out a signal
> >which causes the cards to retransmit...Otherwise packets are lost?
>
> No. You cannot "collide" packets on a single segment of 10BaseT: one wire
> carries the signal in one direction, the other wire carries signals
> in the *other* direction. It is when you have to mix the signals together
> to multiple receivers that the hub or repeater if you're cheap has to
> say "the receivers are busy getting the previous packet, please wait".
Folks:
I have a stranger view of Ethernet than many but the following has
helped me solve some problems.
Ethernet collisions happen at half-duplex transceivers when a receiving
event overlaps a transmitting event. This is why collisions happen when
using non-coaxial Ethernet media, i.e. media with physical paths devoted
to transmission and reception. Signal "collisions" on a properly designed
cable plant except at the transceivers are undefined regardless of the
medium.
Thanks. /criss
------------------------------
From: Brian <[EMAIL PROTECTED]>
Subject: ipfwadm <--> ipchains rules?
Date: Thu, 21 Jan 1999 10:50:51 -0500
Reply-To: [EMAIL PROTECTED]
Hello!
I don't have the slightest clue how to get these ipfwadm rules to
ipchains....I've read the documentation but, this is way outside my
expertise. Can some one help? Thanks in advance! The ipfwadm commands
are listed below.....
/sbin/ipfwadm -F -p deny
/sbin/ipfwadm -F -a m -S 192.168.1.0/24 -D 0.0.0.0/0
/sbin/ipfwadm -M -s 7200 120 900
/sbin/ipfwadm -I -a accept -P tcp -D 0.0.0.0/0 80 -r 987
/sbin/ipfwadm -I -a accept -P tcp -D 0.0.0.0/0 8000 -r 987
/sbin/ipfwadm -I -a accept -P tcp -D 0.0.0.0/0 8080 -r 987
#/sbin/ipfwadm -I -a accept -P udp -D 0.0.0.0/0 53 -r 986
#/sbin/ipfwadm -I -a accept -P tcp -D 0.0.0.0/0 53 -r 986
/sbin/modprobe ip_masq_ftp
/sbin/modprobe ip_masq_raudio
/sbin/modprobe ip_masq_vdolive
/usr/sbin/redir --syslog 192.168.1.53 25 25 &
#/usr/sbin/diald
/usr/sbin/in.tproxyd -s 987 -r nobody 127.0.0.1 3128
------------------------------
Date: Sat, 23 Jan 1999 00:28:24 -0800
From: Matt Corddry <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.system,comp.os.linux.development.apps,comp.os.linux
Subject: Re: kernel: VFS: No free inodes - contact Linus
Hmmm... Well, I crashed it again (easy to do -- I'm simulating clients).
Here's the output from 'df -i' when it crashed:
Filesystem Inodes IUsed IFree %IUsed Mounted on
/dev/sda1 1081344 30952 1050392 3% /
So I'm not running out of inodes. However, the running out of memory
makes _more_ sense, although according to 'top', I still have some
buffers left, and it doesn't appear to be hitting swap too much.
I'll try the 2.2.0-preX kernel and see if that makes a difference.
Thanks to all who have helped with this!
matt corddry
Ingo Saitz wrote:
>
> MoiN
>
> On Thu, 21 Jan 1999, Matt Corddry wrote:
>
> > Well... the subject really explains my problem. I get hundreds of
> > thousands of these at console and in /var/log/messages.
>
> It seems your filesystem has gone out of inodes. This can happen if you
> use to have a lot of small files in one partition. The solution is to find
> the dirs which contain many small files (usually /tmp or under /var) and
> increase the inodes in that partition. ext2fs defaults to one inode per
> 4096 bytes of drive space. So you sould use "mkfs.ext2 -i 2048" or even
> "mkfs.ext2 -i 1024" on that partition *back up your data first!*
>
> Ingo
> --
> Starting Java...
------------------------------
Date: Thu, 21 Jan 1999 03:17:30 +0000
From: Tom Reinertson <[EMAIL PROTECTED]>
Subject: Yikes!!! Bogons!!!
Okay, I turned on arpwatch just to see what it would give me, and now
/var/log/messages is being flooded with the message:
... arpwatch: bogon: {ip address} {ethernet address}
I cannot find any documentation which describes this message. Can
anyone point me in the right direction?
Thanks.
Tom
------------------------------
Date: Fri, 22 Jan 1999 15:44:22 -0700
From: David Moulton <[EMAIL PROTECTED]>
Subject: Re: DNS problem with IP Masq Gateway
Matthew Ho wrote:
...
> applications like Web and ICQ. However, my linux gateway cannot resolve
> internet domain name. I can resolve names within the LAN thru. the
> hosts file. Everytime i try "ping www.yahoo.com" there is no reply.
> when i try "nslookup", i got the following
>
> looks like resolver cannot talk to the dns i specified.
Can you ping outside machines using the IP address and not the domain name?
It might be a routing issue.
Also, just a suggestion, I set up my gateway to be a caching DNS as well.
This seems to speed up the lookups at least a little.
dave
------------------------------
From: [EMAIL PROTECTED] (Clifford Kite)
Crossposted-To: comp.os.linux.setup
Subject: Re: pppd gurus - help, please!
Date: 21 Jan 1999 10:41:29 -0600
Richard R Urena ([EMAIL PROTECTED]) wrote:
: My problem: after the modem connects and pppd starts up,
: it dies after 10 seconds, without having established
: a connection.
: Here's the very short output on the system log:
: [ chat script to dial, etc. ]
: Jan 20 18:45:48 sebastian chat[1221]: CONNECT -- got it
: Jan 20 18:45:48 sebastian chat[1221]: send (^M)
: Jan 20 18:45:48 sebastian pppd[1217]: Serial connection established.
: Jan 20 18:45:49 sebastian pppd[1217]: Using interface ppp0
: Jan 20 18:45:49 sebastian pppd[1217]: Connect: ppp0 <--> /dev/modem
: Jan 20 18:45:58 sebastian pppd[1217]: Hangup (SIGHUP)
: Jan 20 18:45:58 sebastian pppd[1217]: Modem hangup
: Jan 20 18:45:58 sebastian pppd[1217]: Connection terminated.
: Jan 20 18:45:58 sebastian pppd[1217]: Exit.
Best guess at this point would be to replace the chat expect/send
<CONNECT ''> (or <CONNECT ""> with <CONNECT \\c>. This might have
to be <CONNECT "\\c"> depending on how the script is implemented and
read. Instead of the <send (^M) above you should get <send ()> since
the carriage return is suppressed.
If this doen't do it, then change teh pppd option "kdebug 1" to "debug"
and post the appropriate log messages if need be.
: Distribution: Slackware 3.5.0
: Kernel version: 2.0.34
: pppd version: 2.2.0
: Modem: Cardinal MVPV34ILC
: About the modem: I don't know if this is a so-called
: "Winmodem". Calls and email to the company tech
: support have failed (no answer on either). Although the
: front of the box says "Works with Windows 95", it also
: says it works with "Windos 3.1 or higher / DOS 5.0 or higher".
: And as I say, minicom works fine.
If minicom works, then it's not of family winmodem.
--
Clifford Kite <[EMAIL PROTECTED]> Not a guru. (tm)
/* Better is the enemy of good enough. */
------------------------------
From: [EMAIL PROTECTED]
Subject: callin in to my machine...
Date: Fri, 22 Jan 1999 21:53:36 GMT
i have installed redaht 5 on my machien at my place..
now i need to acces my mahicne form my office how do i dial in...
my office machien has 98...
Please redirect ne help to [EMAIL PROTECTED]
n thanks
have a beer on me...
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "Alexander" <[EMAIL PROTECTED]>
Crossposted-To:
apana.lists.os.linux.redhat,hk.comp.os.linux,linux.redhat.applixware,linux.redhat.axp,linux.redhat.install,linux.redhat.list,linux.redhat.misc,linux.redhat.ppp,linux.redhat.rpm,tw.bbs.comp.linux
Subject: how to setup a driver for sis6326 display card in RH5.2 Linux?
Date: Sat, 23 Jan 1999 01:04:40 +0800
i was already setup a Redhat 5.2 linux but it can't recognize my display
card
SIS6326 AGP 8mb RAM.
When i start "startx", the screen is very large, all icon is enlarge.
So, how can i update my display card driver.
Pls give me some advice, thx
I'm so poor, pls tell me step by step, thx
------------------------------
From: Ah Clem <[EMAIL PROTECTED]>
Crossposted-To: news.admin.net-abuse.email
Subject: Re: Bad Sendmails with otherwise good OSes, Re: Open relays on DNS
Date: Fri, 22 Jan 1999 21:12:06 GMT
On 22 Jan 1999 20:35:01 GMT, "Cameron Spitzer" <[EMAIL PROTECTED]>
wrote:
>Unfortunately, a lot of those $3 disks also install Sendmail 8.8
>configured as an open relay. Red Hat 5.1 is one such product.
>
My copy of RH 5.1 came with sendmail 8.8.7 and installed with relaying
denied as the default.
------------------------------
From: [EMAIL PROTECTED]
Subject: Ethernet address....
Date: Fri, 22 Jan 1999 01:38:29 -0800
> I am having 2 related queries to which you gurus may answer quickly.
1. How do I know the ethernet
> address of the adapters in my own machine from
> within a program( say from a C/C++ program - there must be a system
call for
> this purpose.) Also pls. let me know what is the handy command I can
give from
shell prompt to know the same. 2. Given a ethernet address(which as per
my knowledge is unique
use ifconfig
>over the world)
Not necessarily. But it is *very* unlikely to have the same MAC-adresses in
the same subnet
> how do I know whether the interface( whatever machine it is attached
to) is up or
> not. Is their any system call for doing that? It wil be very helpful
for me if you also
> let me know whether there is any command line utility to do that. At
least I should
use ifconfig
Theres is a line stating: UP BROADCAST RUNNING
> be able to do that in my subnet.
> Thank you in advance -
You're welcome
> Sudip
> PS. Your response at my email: [EMAIL PROTECTED] will highly be
appreciated.
*** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ***
------------------------------
From: Kaustav Bhattacharya <[EMAIL PROTECTED]>
Subject: No dial tone. PPP. Redhat5.2
Date: Fri, 22 Jan 1999 11:35:02 +0000
Reply-To: k, dot, bhattacharya, at, bbc, dot, co, dot, uk
I just installed RH5.2 on my Intel box, been using the graphical tool
which comes with it in Xwindows to configure my PPP settings. I
followed the instructions in the manual rather carefully and used Win98
to identify which com port my modem was dialing out through. It was Com
port 3. So I set up all the relevant info in the Xwindow PPP config
util and hit the connect button. No dialtone, no connect, just
nothing. Clicked disconnect and tried all other available ports....
nothing :-( What might be going wrong? The modem connects fine when I
get back to Win98 and start moaning at people in #linux on IRC *lol* :-)
Regards,
Kozzey
------------------------------
From: Criss Hyde <[EMAIL PROTECTED]>
Crossposted-To:
comp.dcom.lans.ethernet,comp.sys.sun.admin,comp.os.ms-windows.networking.win95
Subject: Re: Connect without hub
Date: Thu, 21 Jan 1999 11:52:21 -0500
Mike Humski wrote:
> I am planning to connect the following two machines with ethernet
> without using a hub.
>
> (1) PC (Pentium II 266 Mhz; Intel AL440LX mother board) with
> multiple-boot set up for Win95/Linux/FreeBSD
> Ethernet card: 10/100 Mb (will be Intel EtherExpress Pro or 3Com)
>
> (2) Sun Sparc Classic with Solaris 2.5.1
> Ethernet adapter: 10 Mb (preinstalled in the system)
>
> These machines are only two feet apart.
> I heard it is possible to directly connect two machines with
> ethernet cards if you use a crossover 10 baseT cable.
>
> But I would like to make sure it is really possible in my case,
> since I have no experience in setting up an ethernet connection.
>
> There is no possibility to add another platform in the future,
> hence no need to use a hub.
>
> However, I would like to know the following points:
> (1) Will there be any degradation in performance or stability if
> I do not use a hub?
> (2) Will there be any danger in connecting 10 Mb only (Sparc) directly
> to 10/100 Mb (PC) with a crossover cable?
> (3) Will there be any other problems if I use a crossover cable
> without a hub?
>
> I would appreciate any help, comment, or pointer related to this subject.
> Thanks in advance.
>
> ---------
> Mike
Mike:
I responded to one of your responders before reading your initial post.
Sorry.
1) Given two Ethernet stations zero repeaters between them will give a tiny
performance advantage over more than zero repeaters.
2) Your 10/100baseT station may need to be statically set to 10baseT
half-duplex to match your Sparc if auto-negotiation fails.
3) In my shop crossover cables tend to be hand made and lower quality than
straight through and for this reason alone we sometimes use two standard
cables and a repeater instead.
The need for repeaters to participate in collision detection by emitting jam
signals disappears when repeaters are not used.
Does any of this help?
Thanks. /criss
--
///// Criss Hyde Desk: 703/560-5000 x2348
///// Raytheon Systems Company Fax: 703/208-1234
///// 7700 Arlington Boulevard
///// Falls Church VA 22042
------------------------------
From: "Killer" <[EMAIL PROTECTED]>
Subject: PPP Script for RedHat Linux
Date: Mon, 18 Jan 1999 21:39:05 +0100
Heya.. I Need a PPP Script witch makes me capeable of getting on
to the net, with a Eicon Diva 2 card.. Can Someone help me ?
------------------------------
Date: Sat, 23 Jan 1999 11:41:14 +1300
From: Richard Hector <[EMAIL PROTECTED]>
Subject: Re: Q: Network Linux and OpenDOS ?
Pat Thoyts wrote:
>
> Raymond Doetjes <[EMAIL PROTECTED]> writes:
>
> > > I've got two systems, they both run Linux, but sometimes I run
> > > OpenDOS on the smaller one. Can I use Samba to let the OpenDOS
> > > system talk with the other Linux system? (to avoid a floppy-net)
> > >
> > > Any and all pointers will be greatly appreciated!
> > > Thanks!
> > > --
> > > Mike Hall
> > > [EMAIL PROTECTED]
>
> OpenDOS has Personal Netware included. This isn't Micro$oft networking
> and uses IPX and NCP not SMB, hence you can't make it work with samba
> unless you are using Windows for Workgroups on top of OpenDOS.
>
> What you need to do is run a netware server on your linux box. Then
> your DOS PC can connect to that. A good one is mars_nwe available at
> all good linux software sites!
Alternatively you could probably use a DOS NFS client, such as XFS. I
think it's shareware, and uses Packet Drivers, but I'm not sure where to
get it. There are no doubt others as well.
Richard Hector
------------------------------
From: [EMAIL PROTECTED] (Clifford Kite)
Subject: Re: Compiling ppp-2.3.5 under kernel 2.0.34
Date: 21 Jan 1999 09:46:10 -0600
Simon Annetts ([EMAIL PROTECTED]) wrote:
: Has anyone tried ppp-2.3.5 under slackware with kernel 2.0.34.
: I can compile ppp fine but when I try to recompile the kernel with the new
: ppp module
: I get the following somewhere during kernel make:
: In file included from /usr/src/linux-2.0.34/include/linux/string.h:39,
: from /usr/src/linux-2.0.34/include/asm/termios.h:58,
: from /usr/src/linux-2.0.34/include/linux/termios.h:5,
: from /usr/src/linux-2.0.34/include/linux/tty.h:20,
: from /usr/src/linux-2.0.34/include/linux/sched.h:26,
: from ppp.c:65:
: /usr/src/linux-2.0.34/include/asm/string.h:443: warning: control reaches end
: of non-void function
: /usr/src/linux-2.0.34/include/asm/string.h: In function
: `__constant_c_and_count_memset':
: /usr/src/linux-2.0.34/include/asm/string.h:594: warning: control reaches end
: of non-void function
: ppp.c: In function `ppp_init_ctrl_blk':
: ppp.c:465: structure has no member named `ddinfo'
: ppp.c:466: structure has no member named `ddinfo'
The ppp-2.3.5 package has a couple of problems with kernels 2.0.34 to
2.0.36 . One is that the ppp-2.3.5/linux/ppp.c isn't copied to the kernel
source tree during the "make kernel" phase of ppp-2.3.5 installation.
The other is a bug in a section of ppp.c that detects the kernel version
and selects kernel-dependent code.
A solution is to fix the versioning code in ppp-2.3.5/linux/ppp.c and
then copy it to /usr/src/linux/drivers/net, overwriting the older ppp.c
there.
Look for this code beginning at line 3079 in ppp-2.3.5/linux/ppp.c and
change the < to >= . Copy this ppp.c to /usr/src/linux/drivers/net and
recompile the kernel.
#if LINUX_VERSION_CODE < VERSION(2,1,86)
#define FREE_SKB(skb) dev_kfree_skb(skb)
#else
#define FREE_SKB(skb) dev_kfree_skb(skb, FREE_WRITE)
#endif
NOTE: This problem goes away with 2.1.131+ kernels.
--
Clifford Kite <[EMAIL PROTECTED]> Not a guru. (tm)
/* I gave up on politics when no matter who I voted for, I regretted it.
* -- Pepper...and Salt, WSJ */
------------------------------
From: [EMAIL PROTECTED] (Dirk Ruediger)
Crossposted-To: comp.os.linux.misc,de.alt.comm.mgetty
Subject: Re: Any program conver text file to g3 format file ?
Date: 22 Jan 1999 18:43:49 GMT
Reply-To: [EMAIL PROTECTED]
On 22 Jan 1999 13:39:30 GMT, Wilson Lam <[EMAIL PROTECTED]> wrote:
>Hi,
>I have tried pbmtext comes with mgetty. It seems it not working
>correctly. Is there any other program can do the job correctly ?
>Any help is appreciated !
>
>Wilson
>
Hi,
I assume, you want to convert the text into faxg3 format. So convert it to
postscript and then w/ ghostscript to output format "faxg3".
Alternatively, you can convert the file into dvi dvi-format (output of latex)
and let dvi2fax (included in tetex-pkg) the conversion to faxg3.
Good luck!
Ciao for now, Dirk
--
Dirk Ruediger, Magdeburg, Germany
"I used to be interested in Windows NT, but the more I see of it the more it
looks like traditional Windows with a stabler kernel. I don't find anything
technically interesting there. In my opinion, MS is a lot better at making
money than it is at making good operating systems." -- Linus Torvalds
------------------------------
** 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.networking) 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-Networking Digest
******************************
