Linux-Misc Digest #754, Volume #26 Mon, 8 Jan 01 16:13:02 EST
Contents:
Re: ppp problem with ISP (Jean-David Beyer)
Re: ppp problem with ISP ("Peter T. Breuer")
Re: Crond logging ("Aitch")
Re: $$$ NEW WAY!! TO MAKE FAST CASH $$$ (Jean-David Beyer)
Re: L2-Cache of Pentium2 with Linux (Jean-David Beyer)
Re: konqueror and home key. (Roberto Alsina)
Re: help with compiling kernel 2.4 ("Ninewands")
Re: help with compiling kernel 2.4 ("Ninewands")
Re: Sound editing software (malicorne)
Re: The mailx program (Bob Tennent)
SCSI/IDE/LILO problem (Matthew Fleming)
Re: email software (Harry)
Re: booting Redhat 7.0 from floppy is very slow (Dave Brown)
Re: Kernel 2.4 problem (A Guy Called Tyketto)
Re: ppp problem with ISP (Glitch)
Re: Check for bad pointer ("D. Stimits")
----------------------------------------------------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: ppp problem with ISP
Date: Mon, 08 Jan 2001 15:09:56 -0500
Glitch wrote:
>
> >
> > When I first was setting up Linux on my first machine, I had trouble
> > getting in to my ISP. I had my machine log everything I did into
> > /var/log/messages, and everything seemed fine. I watched the chat script
> > run, and everything was as I expected, except that a little after they
> > told me my IP address and theirs, they hung up on me. Since I have a
> > full service ISP, I called tech support and they snooped the connectioin
> > as I tried to connect. They noticed instantly that I did everything
> > correctly, but did not tell them to go into ppp mode. It took about 15
> > seconds for me to fix that.
> >
>
> I have no way of telling my ISP to go into PPP mode, not anymore.
That cannot be true. Other users can use their service or they would be
out of business already. Windows knows no secrets that you do not, in
principle, know already.
> They
> removed that option from their menu system.
So call them and ask them.
> I'm baffled b/c somehow a
> ppp link (using win98) can still be established betweeen me and my ISP
> w/o me having to worry about whether they have a ppp option anymore. In
> WIn98, it just connects and logs me on. In linux, ppp is not started at
> my ISP, it is on my end but b/c of the first situation pppd on my end
> dies b/c it gets no response from my ISP.
SO CALL THEM AND ASK THEM HOW TO DO IT! If they have a secret way, they
should tell you, or you should get an ISP who will.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 3:05pm up 4 days, 21:44, 4 users, load average: 2.12, 2.09, 2.01
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: ppp problem with ISP
Crossposted-To: comp.os.linux.networking
Date: Mon, 08 Jan 2001 20:15:07 GMT
In comp.os.linux.misc Glitch <[EMAIL PROTECTED]> wrote:
> I know that. The mystery here is its started when i use win98 w/o my
> intervention. Assuming it's started on my end somehow the ISP picks it
> up or it does it on its own and starts communicating to Win98 that it's
> ready for a connection. Or assuming it does it's part first and win98
> picks up a second or 2 later a question arises....how do i my ISP to
> start ppp first if indeed it is the one starting first when i use win98?
Ask them! They changed their protocol. Tell them that you are starting
ppp, and that their end doesn't respond. But you haven't proved that
to me, and so I for one don't believe you. It's easy to prove it to me:
show me the pppd log messages. Nothing more nor less is required. Just
do it instead of rambling on.
>> You can't. You have to do something else, such as starting ppp, or
>> telling the other end to start ppp. There is no magic.
> i have no way to tell my ISP to start ppp. I used to but not anymore.
> That's the problem. In win98 i dont have to tell them but i can still
> get on the Net, not so in Linux. SO what is the difference???
Look and find out.
>> It works fine. Look at the ppp debugging messages if you need more
>> info. There is no need to play at guessing.
> i'll try it as of right now its my only option.
It's your FIRST option. After that you can fall back on random
guesswork, or whatever you are using now.
>> Have a look and find out. There is no need to use your imagination.
>> Just take a look. Possibly you are requiring authentication, or some
>> other minor thing that the other end can't satisfy. Or the other
>> end is demanding a compression mode that you can't satisfy. You have
>> to observe.
> answer this: why is it as easy as supplying a phone number, username,
> and password with Win98 but it isnt like that in Linux? It seems i have
Eh?
> to look at debug messages and fool around with who knows what else for
> the same thing to happen.
Of course (why do you view finding out what is going on as some sort of
penance?) Try it in win95 or win2000 or NT and let us know how you
get on! If your ISP changes things so that you can't start PPP their
end and starting PPP your end doesn't elicit a response, then you
and they have a problem to resolve! Get to it.
Peter
------------------------------
From: "Aitch" <[EMAIL PROTECTED]>
Subject: Re: Crond logging
Date: Mon, 08 Jan 2001 20:13:39 GMT
"Jeff Lacy" <[EMAIL PROTECTED]> wrote in message
news:ehe66.20024$[EMAIL PROTECTED]...
| Hello everyone. I'm kind of new to the list, so please bear with me if I
| sound stupid or something. Thanks.
|
| I am running RH7.0, and it seems like most of /var/log/messages is
consumed
| with 'dumb' logs from crond. I was wondering if I could maybe tell crond
to
| log it's output to /var/log/cron or something, and if it can be done, how
do
| I do it? I would really appreciate any advice/help someone could give me.
| Thanks a bunch in advance :-D
In your /etc/syslog.conf look for a line:
*.=info;*.=notice /var/log/messages
change it to:
*.=info;*.=notice;cron.none /var/log/messages
let syslogd re-read it's configuration by doing a killall -HUP syslogd
--
2:06pm up 25 min, 1 user, load average: 0.99, 0.69, 0.31
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: $$$ NEW WAY!! TO MAKE FAST CASH $$$
Date: Mon, 08 Jan 2001 15:16:51 -0500
> [EMAIL PROTECTED] wrote:
>
> c
>
> NEED CASH, FAST MONEY IT REALLY WORKS!!!!
>
This turkey spammed comp.mail.sendmail, too. I already have notified his
ISP and various other anti-spamming sites.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 3:15pm up 4 days, 21:54, 2 users, load average: 2.31, 2.25, 2.12
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: L2-Cache of Pentium2 with Linux
Date: Mon, 08 Jan 2001 15:17:50 -0500
Ralf Render wrote:
>
> It seems, that Linux does not use the L2-cache of my Pentium2.
> Do I need a special kernel?
>
> Thanks.
I doubt it. How did you determine this? Is your BIOS set up so the L2
cache is used?
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 3:15pm up 4 days, 21:54, 2 users, load average: 2.31, 2.25, 2.12
------------------------------
From: Roberto Alsina <[EMAIL PROTECTED]>
Subject: Re: konqueror and home key.
Date: Mon, 08 Jan 2001 20:14:29 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> Sometimes when I am searching a document, I want to go to the top a
> the document. When i do this I press the home key.
> Unfortunately konqueror takes me to the top of the document.
> Is there any way to change this behaviour?
Excuse me, but isn't Konqueror doing exactly what you want? ;-)
--
Roberto Alsina
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Ninewands" <[EMAIL PROTECTED]>
Subject: Re: help with compiling kernel 2.4
Date: Mon, 8 Jan 2001 14:07:59 -0600
Reply-To: "Ninewands" <newsgroup>
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Upgrade modutils to the version specified in
/usr/src/linux/Documentation/Changes. I believe it has to be 2.4.0
or later (I'm using 2.4.1). That should make the problem go away.
What's happening is that your depmod can't read the dependencies for
the 2.4.0 modules, and therefore, make can't produce the
/lib/modules/2.4.0/modules.dep file.
Regards,
ninewands
"Hung Ngoc Lai" <[EMAIL PROTECTED]> wrote in message
news:93d240$[EMAIL PROTECTED]...
> Hi everyone,
>
> I am running RedHat Linux version 6.1 with kernel 2.2.18. I would
> like to upgrade the kernel to version 2.4. I download
> linux-2.4.0.tar.gz to /usr/src directory the following steps:
> 1) gunzip linux-2.4.0.tar.gz
> 2) tar -xvpf linux-2.4.0.tar
> 3) cd /usr/src/linux
> 4) make mrproper
> 5) make menuconfig (in this step, I uncheck the Cardbus support
> (pcmcia) because
> I am running linux on a destop [dell 300MHz])
> 6) make dep
> 7) make clean
> 8) make bzImage
> 9) make modules
> 10)make modules_install
>
> in step 10, I get this error:
>
> make[1]: Entering directory `/usr/src/linux-2.4.0/arch/i386/lib'
> make[1]: Nothing to be done for `modules_install'.
> make[1]: Leaving directory `/usr/src/linux-2.4.0/arch/i386/lib'
> cd /lib/modules/2.4.0; \
> mkdir -p pcmcia; \
> find kernel -path '*/pcmcia/*' -name '*.o' | xargs -i -r ln -sf
> ../{} pcmcia if [ -r System.map ]; then /sbin/depmod -ae -F
> System.map 2.4.0; fi /sbin/depmod: invalid option -- F
> Usage: depmod [-e -s -v ] -a [FORCED_KERNEL_VER]
> depmod [-e -s -v ] MODULE_1.o MODULE_2.o ...
> Create module-dependency information for modprobe.
>
> -a, --all visit all modules
> -d, --debug run in debug mode
> -e output unresolved symbols
> -i ignore symbol versions
> -m, --system-map <file> use the symbols in <file>
> -s, --system-log use the system log for error reporting
> --help display this help and exit
> -v, --verbose run in verbose mode
> -V, --version output version information and exit
> make: *** [_modinst_post] Error 1
>
> *******
>
> At this point, I copy the bzImage and System.map to /boot, modify
> the /etc/lilo.conf, and run the /sbin/lilo command and reboot the
> box. When the system is restarted, it hangs. I could not get
> passed the line "decompressing the linux kernel".
>
> Going back to the error message, it has to do with the pcmcia
> modules which I don't need. How do I get around this problem?
> When I recompile the box with kernel 2.2.18, I didn't see the
> pcmcia directory under
> /lib/modules/2.2.18; however, I do see the pcmcia directory
> /lib/modules/2.4.0/pcmcia under kernel 2.4. Can anyone help me
> with this problem? Right now, I am running kernel 2.2.18 but I
> really would like to use kernel 2.4. I am a newbie here so please
> be gentle with me.
>
> Many Thanks....
> David
=====BEGIN PGP SIGNATURE=====
Version: PGPfreeware 6.5.8 for non-commercial use <http://www.pgp.com>
iQA/AwUBOloeHqyFJnQfZdqfEQLBrgCg45r3TS+ynj9+1We5S1IwZzWzKkQAnjY3
P32UTkDIAfMg+ufyaZxljatx
=PeEi
=====END PGP SIGNATURE=====
------------------------------
From: "Ninewands" <[EMAIL PROTECTED]>
Subject: Re: help with compiling kernel 2.4
Date: Mon, 8 Jan 2001 14:11:41 -0600
Reply-To: "Ninewands" <newsgroup>
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Upgrade modutils to the version specified in
/usr/src/linux/Documentation/Changes. I believe it has to be 2.4.0
or later (I'm using 2.4.1). That should make the problem go away.
What's happening is that your depmod can't read the dependencies for
the 2.4.0 modules, and therefore, make can't produce the
/lib/modules/2.4.0/modules.dep file.
Regards,
ninewands
"Hung Ngoc Lai" <[EMAIL PROTECTED]> wrote in message
news:93d240$[EMAIL PROTECTED]...
> Hi everyone,
>
> I am running RedHat Linux version 6.1 with kernel 2.2.18. I would
> like to upgrade the kernel to version 2.4. I download
> linux-2.4.0.tar.gz to /usr/src directory the following steps:
> 1) gunzip linux-2.4.0.tar.gz
> 2) tar -xvpf linux-2.4.0.tar
> 3) cd /usr/src/linux
> 4) make mrproper
> 5) make menuconfig (in this step, I uncheck the Cardbus support
> (pcmcia) because
> I am running linux on a destop [dell 300MHz])
> 6) make dep
> 7) make clean
> 8) make bzImage
> 9) make modules
> 10)make modules_install
>
> in step 10, I get this error:
>
> make[1]: Entering directory `/usr/src/linux-2.4.0/arch/i386/lib'
> make[1]: Nothing to be done for `modules_install'.
> make[1]: Leaving directory `/usr/src/linux-2.4.0/arch/i386/lib'
> cd /lib/modules/2.4.0; \
> mkdir -p pcmcia; \
> find kernel -path '*/pcmcia/*' -name '*.o' | xargs -i -r ln -sf
> ../{} pcmcia if [ -r System.map ]; then /sbin/depmod -ae -F
> System.map 2.4.0; fi /sbin/depmod: invalid option -- F
> Usage: depmod [-e -s -v ] -a [FORCED_KERNEL_VER]
> depmod [-e -s -v ] MODULE_1.o MODULE_2.o ...
> Create module-dependency information for modprobe.
>
> -a, --all visit all modules
> -d, --debug run in debug mode
> -e output unresolved symbols
> -i ignore symbol versions
> -m, --system-map <file> use the symbols in <file>
> -s, --system-log use the system log for error reporting
> --help display this help and exit
> -v, --verbose run in verbose mode
> -V, --version output version information and exit
> make: *** [_modinst_post] Error 1
>
> *******
>
> At this point, I copy the bzImage and System.map to /boot, modify
> the /etc/lilo.conf, and run the /sbin/lilo command and reboot the
> box. When the system is restarted, it hangs. I could not get
> passed the line "decompressing the linux kernel".
>
> Going back to the error message, it has to do with the pcmcia
> modules which I don't need. How do I get around this problem?
> When I recompile the box with kernel 2.2.18, I didn't see the
> pcmcia directory under
> /lib/modules/2.2.18; however, I do see the pcmcia directory
> /lib/modules/2.4.0/pcmcia under kernel 2.4. Can anyone help me
> with this problem? Right now, I am running kernel 2.2.18 but I
> really would like to use kernel 2.4. I am a newbie here so please
> be gentle with me.
>
> Many Thanks....
> David
=====BEGIN PGP SIGNATURE=====
Version: PGPfreeware 6.5.8 for non-commercial use <http://www.pgp.com>
iQA/AwUBOloe+6yFJnQfZdqfEQLeHACgoCnncq1RlFQsTp+yONzkN8dvjYsAoJwQ
hlIefOrEJV08eGK+gclxOT1L
=Wzhu
=====END PGP SIGNATURE=====
------------------------------
From: malicorne <[EMAIL PROTECTED]>
Subject: Re: Sound editing software
Date: Mon, 8 Jan 2001 21:24:13 +0100
Reply-To: [EMAIL PROTECTED]
Steve Connet wrote:
> [EMAIL PROTECTED] (Bill Unruh) writes:
>
> > I would like to get some sound editing software--ie I can read in an
> > audio track say, and cut out a piece of it to then reburn to a CD.
> >
> > This is for teaching so I can take samples of the sound to illustrate
> > various musical things.
>
> Did you ever find anything? I am looking for a program that is
> comparable to N-TRACK studio in Windows. It's a multi-track recording
> program. Allows you to record from the mic, then play it back with
> recording another track. Then again and again. And finally you can mix
> down all tracks to a single audio file that can be burned to CD.
> Anything like that exist for linux do you know?
I think ecasound does this. The guy who made it has recorded some tunes
available at MP3.com and it sure sounds like a multi-track setup. Do a
search on ecasound and you'll find it.
Alain
------------------------------
From: [EMAIL PROTECTED] (Bob Tennent)
Subject: Re: The mailx program
Date: 8 Jan 2001 20:26:50 GMT
Reply-To: [EMAIL PROTECTED]
On 08 Jan 2001 15:44:44 -0400, * Tong * wrote:
>There are two version of command line email programs: mail and
>mailx. At least it is true for Solaris. I'm wondering where I can
>find the mailx for Linux.
>
>The reason I'm asking here is that the mail tool come with Redhat is
>called mailx-8.1.1-10.i386.rpm, but it is actually mail.
Mail works much like mailx on Solaris. If you want attachments and other
goodies, look at nail:
http://omnibus.ruf.uni-freiburg.de/~gritter
Bob T.
------------------------------
From: [EMAIL PROTECTED] (Matthew Fleming)
Subject: SCSI/IDE/LILO problem
Date: 8 Jan 2001 20:27:38 GMT
Netmeisters,
I would appreciate your help with the following problem: I have a
system with a SCSI disk, which holds the root partition, and a 120 MB
IDE superdisk, which is connected to the first IDE channel (so
recognized as hda). I have just connected a 20 GB IDE disk to the
second IDE channel (so hdc). The problem is that the system will now
boot only when I have told the BIOS that the IDE HD is not installed.
This is despite rerunning LILO with the following lilo.conf:
append="hdc=16383,16,63"
# Start LILO global Section
boot=/dev/hda
disk = /dev/sda
bios=0x80
disk = /dev/hdc
bios=0x81
#compact # faster, but won't work on all systems.
read-only
prompt
timeout=10
vga = normal # force sane state
# End LILO global section
# Linux bootable partition config begins
image = /boot/vmlinuz
root = /dev/sda2
label = Linux
# Linux bootable partition config ends
#
If it matters, fstab looks like this:
/dev/sda2 / ext2 defaults
1 1
/dev/sda4 /home ext2 defaults
1 2
/dev/sda3 swap swap defaults
0 0
/dev/hdc3 /home2 ext2 defaults
1 2
/dev/hdc2 /fromboth msdos defaults
1 2
/dev/hda1 /mnt/floppy ext2 noauto
0 0
/dev/cdrom /mnt/cdrom iso9660 noauto,ro
0 0
none /proc proc defaults
0 0
Oddly, with the BIOS set to know about the new disk, I can't even boot
the system with a boot disk; when I try to do so, it starts to load
the kernel but then reboots itself, and so on ad infinitum.
The system also has Win NT on it, which recognizes the new IDE disk
without any trouble.
Your assistance much appreciated.
Matthew Fleming
==============================================================================
Matthew G. Fleming, MD phone : 414.456.4072
Associate Professor fax : 414.456.6518
Department of Dermatology s-mail: Dept. of Dermatology
Medical College of Wisconsin Medical College of Wisconsin
MFRC Room 4061
e-mail: [EMAIL PROTECTED] Milwaukee, WI 53226-4810
==============================================================================
------------------------------
From: Harry <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux
Subject: Re: email software
Date: Mon, 08 Jan 2001 20:37:02 +0000
On 8 Jan 2001 04:23:24 GMT, [EMAIL PROTECTED] (Dances With Crows)
wrote:
>On Mon, 08 Jan 2001 01:55:50 +0000, Harry staggered into the Black Sun
>and said:
>>I'm certain this is going to sound terribly naive, but.....
>>what's the bare minimum required for sending and receiving email over
>>IP? The sofware is so wide and varied it's bewildering. Sendmail,
>>procmail, mailx, pine, elm, mutt, metamail, imapd, smail etc, plus
>>umpteen X-based efforts. Jesus, what's it all about?
>
>Choice. Freedom is just chaos with better lighting, after all :-]
>
>>From what I've read so far sendmail seems to be the bee's knees, but
>>isn't it rather a complex server system for ISPs and the like? All I
>>want to do is send ascii to my ISP's smtp box and pull whatever is
>>waiting from their pop3 box.
>
>What you want is a Mail User Agent (MUA). There are (predictably) tons
>of those for Linux, among them pine, mutt, Kmail, Netscape Messenger,
>and the prototype Evolution. So pick one of them and go to it.
>Netscape might be the best answer if you're coming from a Windoze
>worldview, since the configuration/interface is almost identical to how
>it is on 'Doze. I like Pine, but then I'm partially insane.
>
>sendmail, exim, qma!l, postfix, and the like are Mail Transfer Agents,
>generally not run by Joe User. They are powerful and nifty, but not
>really required if you have a POP/IMAP + SMTP box you can connect to.
>If you have a permanent Net drop and multiple users who all want mail,
>you may want to read up on MTAs and use one of them. Otherwise don't
>bother.
>
>>The sweetest answer would be for somebody to point me to a unix port
>>of Forte's Agent ;o)
>
>Isn't that a newsreader, not a MUA? Um, try PAN, the Pimp-Ass
>Newsreader, at http://superpimp.org/ . HTH,
Thanks to all of you lads for the advice.... (I confess to still being
a bit in the dark though).
To compare the situation with a windows environment, what equals what?
Or to put it another way, it seems to me that the unix "MTA" component
is missing from the windows scene, and so what would I lose by not
using one of these on my linux box? I've been reading the two email
HOWTOs, and so my head is spinning somewhat. Mutt for the MUA appeals
to me with being so configurable and not forcing an editor upon you.
Also one of the authors highly recommends qmail (which I'd never heard
of) as an alternative to sendmail for security, simplicity and speed.
Does this all seem over the top for one guy on one small home-based
network?
I also read in those HOWTOs not to post mail related questions to the
linux newsgroups <cringe> and so I thank you also for not burning my
arse. I'll bow out after this and go lurk in comp.mail.?.?
Free Agent is a news reader by the way - the paid for version is a
news reader combined with the best POP3 client I've ever seen.
Cheers,
Harry
------------------------------
From: [EMAIL PROTECTED] (Dave Brown)
Subject: Re: booting Redhat 7.0 from floppy is very slow
Reply-To: [EMAIL PROTECTED]
Date: 8 Jan 2001 15:52:50 -0600
In article <[EMAIL PROTECTED]>, Leonard Evens wrote:
>Frank wrote:
>>
>> I have installed Redhat 7.0 on two machines, IBM 300GL with PII350 and
>> 128MB, Dell XPS T450 with 196MB. They have the same problem. It is very
>> slow to boot from floppy. It takes more than 5 minutes to read some
>> information in. As soon as finishing the reading part, Redhat is working
>> fine.
>>
>> Did anybody experience same kind of problem before? What kind of problem
>> it could be? I have another machine installed RH 7.0. It boots normally
>> from floppy.
>>
>> --
>> Posted via CNET Help.com
>> http://www.help.com/
>
>The floppy made with mkbootdisk boots very slowly on many PCs.
>Surprisingly, this is less of a problem with really old PCs. It
>may have something to do with the BIOS routines for reading the
>floppy. It can take ten minutes or longer to start booting. It
>takes forever to get the first dot and then they come painfully
>slowly.
>
>But there are a couple of ways around this.
>
>First, you can put the lilo boot loader on a floppy by creating
>a lilo.conf file with first line
>boot=/dev/fd0
>and everything else the same. It is best to give this a separate
>name and run /sbin/lilo with the -C option to specify that name.
>Then the initial lilo boot sequence will load quickly from the
>floppy and the rest will come from the hard drive.
>
>Second, you can use
>dd if=your_kernel of=/dev/fd0 bs=1440k
>to write your kernel directly to a floppy and
>rdev /dev/fd0 your_boot_partition
>to tell the kernel on your floppy where to find the boot partition.
>This floppy will load relatively quickly. It may also be used
>if there is some problem with the kernel on the hard disk.
Actually, it's still painfully slow with nothing but the kernel
image on it (your "second case"). When I teach SA classes, I
usually tell them about "trying a kernel first" after compilation
by dd'ing it to a floppy. But on some machines (particularly IBM's)
this is excruciatingly slow. There's something about the bios that's
different. As I recall, Dell's (tower versions) are also quite slow.
(Maybe an unrelated observation, but I've noticed on some Windows
machines that diskette operations have gotten awfully slow as well.)
I have MBs made by ASUS, Gigabyte, and ABIT, (all Socket 7), and none
show these characteristics with the diskette drive reading a stream,
making me think it's BIOS-related.
--
Dave Brown Austin, TX
------------------------------
From: A Guy Called Tyketto <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4 problem
Date: Mon, 08 Jan 2001 19:16:25 GMT
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Lack Mr G M <[EMAIL PROTECTED]> wrote:
> There are several ppp* modules (pp_async, pp_deflate, and about 3
> others). If you modprobe these ppp will work (look in the drivers/net
> modules dir for ppp* files to get the list).
Even more than that, did you take a look at
linux/Documentation/Changes? there's an important blurb about using ppp
as a module, and making changes to /etc/modules.conf. If that's not
done, and upgrading to at least modutils 2.4.0 (2.4.1 is the latest),
ppp won't go.
BL.
- --
Brad Littlejohn | Email: [EMAIL PROTECTED]
Unix Systems Administrator, | [EMAIL PROTECTED]
Web + NewsMaster, BOFH.. Smeghead! :) | http://www.wizard.com/~tyketto
PGP: 1024D/E319F0BF 6980 AAD6 7329 E9E6 D569 F620 C819 199A E319 F0BF
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.4 (FreeBSD)
Comment: For info see http://www.gnupg.org
iD8DBQE6WhLiyBkZmuMZ8L8RAsqHAJ0UcBX87bVAEwdkvaLCKXwcGDpncQCeNjCD
WkLZKRan0PQHZqFoo5LDgLc=
=97In
=====END PGP SIGNATURE=====
------------------------------
Date: Mon, 08 Jan 2001 16:07:24 -0500
From: Glitch <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: ppp problem with ISP
"Peter T. Breuer" wrote:
>
> In comp.os.linux.misc Glitch <[EMAIL PROTECTED]> wrote:
>
> >> Tell your isp that you start ppp and they don't, if that is what
> >> happens. But I don't believe it.
>
> > That's what KPPP tells me. If I take out my chat script, and use PAP,
>
> kppp is NOT a debugger. Look at the pppd messages only. They must be
> directed to one of your log files via syslogd. You must add debug as a
> pppd option.
>
> Peter
I didn't say it was. I've already said what's in /var/log/messages
I made some headway now. With no help from my ISP i fooled around and
got a prompt where i can choose what connection I want to start. If i
type in a "0" at their menu prompt I get shell like prompt. If i type in
'ppp' i get a remote IP and it tells me what the current MTU is set at.
However I don't get the weird ascii characters , which mean the server
is ready to make a ppp connection, until i hit Enter after I type 'ppp'
and also after hitting another key, no matter what key it is. THe
problme is i can't remember how to send carriage returns in the Send
script (which is what i'm having to use in order to get to the
shell-like prompt).
I thought /r or \r was a carriage return but it didnt seem to help
I did this:
<above this are lines that already work>
expect %
send ppp/r and I also tried ppp\r
send /r and also \r
should something similar to the above work in order to get the server to
start sending those weird ascii characters?
I can get that far using minicom but actually doing it in kppp is not as
easy although im almost there.
------------------------------
Date: Mon, 08 Jan 2001 14:03:13 -0700
From: "D. Stimits" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Check for bad pointer
Igor Lavrinenko wrote:
>
> "D. Stimits" wrote:
>
> > Igor Lavrinenko wrote:
> > >
> > > Let's say I have a pointer of type char*. Is there any way to tell if I
> > > am going to get SIGSEGV or SIGBUS if I try to dereference this
> > > pointer before I actually do (dereference) it. Of course, I can catch
> > > these signals. What if I can't simply exit() from the program but
> > > would rather return some error from the routine that caused
> > > (or would cause if I did not prevent it) SIGSEGV/SIGBUS.
> >
> > A good thing to do is to always assign new pointers to NULL, until they
> > point at something. Once they are freed, they should again be assigned
> > to NULL. You can then trust a check against mypointer == '\0' to tell
> > you if it is valid or not. It is unfortunate if you have to deal with
> > someone else's code and they don't follow that practice.
>
> This is a great idea except it does not work in cases when you receive
> a pointer as an argument in the function call (say a pointer to an array
> of characters) and user of the library is responsible for allocating of the
> memory. The only thing one can do in this situation is to check if p != 0.
> This is certainly not enough. The problem stays.
If an array is passed by reference or pointer, a check for NULL can
still be used. If it was allocated in stack space, it is already
allocated, and will not be NULL. But still whoever passes it must agree
that they will not create any pointers without setting them to NULL
until allocated, or else the whole thing fails. There isn't a whole lot
you can do about this problem without cooperation of various coders. If
you are using C++, you can try to use smart pointers (the description is
available in books on design patterns). If you were to use a normal
pointer, and it had some address that is random and it just happens to
be within your program's data space, you'd get bizarre behavior by using
it, rather than instant program death...or a signal that could be
caught.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
