pppd does not set ppp0 interface

I'm using pppsetup (pppd 2.4.1) on Slackware 9.1. I'm also using a US
Robotics, Courier V.EVERYTHING external modem on /dev/modem
(=/dev/ttyS1).

It works great for me but no ppp0 interface appears (I ran "ppp-go"
without on demand). The modem connects fine (I can see that in
/var/log/messages) but after Connect: ppp0 <--> /dev/modem it just
sits there until a hang up about a minute later (with another ISP it
did not hang up but there was still no ppp0). This is what that part
of the log reports:

Connect: ppp0 <--> /dev/modem
Modem hangup
Connection terminated
Exit

I also find it strange that pppsetup reports (as a warning) that it
can't find ppp as a module or in the kernel. Perhaps the module was
not loaded at the time of the setup. Anyway, since I used a bare
kernel to install (this is an old machine with little ram) I checked
/usr/src/linux/.config and found PPP as a module. And if I do a lsmod
I get:

root@cigar:/usr/src/linux# lsmod
Module Size Used by Not tainted
ppp_async 7552 0 (autoclean)
ppp_generic 15452 0 (autoclean) [ppp_async]
slhc 5040 0 (autoclean) [ppp_generic]
[snip]

It did set up my resolv.conf:

root@cigar:/etc/ppp# cat /etc/resolv.conf
search ig.com.br
nameserver 200.225.159.124

For completeness' sake, this is the pap-secrets file (where I changed
the login info)

root@cigar:/etc/ppp# cat /etc/ppp/pap-secrets
# PAP authentication file: /etc/ppp/pap-secrets
# This file should have a permission of 600.
# ~# chmod 600 /etc/ppp/pap-secrets
# Username Server Password IP addresses
"catbert" * "ktex23#"

Needless to say, this account works from my Windows PC.

This is in my options file:
root@cigar:/etc/ppp# cat options
lock
defaultroute
noipdefault
modem
/dev/modem
57600
crtscts
passive
asyncmap 0

This is in my chat script file:

root@cigar:/etc/ppp# cat pppscript
TIMEOUT 60
ABORT ERROR
ABORT BUSY
ABORT "NO CARRIER"
ABORT "NO DIALTONE"
"" ATZ OK ATX3 OK
"atdt0,33550168"
TIMEOUT 75
CONNECT

If I use minicom to dial, I get to the ppp header:

atz
OK
atx3
OK
atdt0,33655000
CONNECT 31200/ARQ
~ÿ}#.!}!}!} }8}"}&} }*} } }#}$.#}%}&bùól}'}"}(}"å.~ [snip]

Can anyone point me to a solution?

Many thanks.

Miguel

(E-Mail Removed) (miguelk) writes:

]I'm using pppsetup (pppd 2.4.1) on Slackware 9.1. I'm also using a US
]Robotics, Courier V.EVERYTHING external modem on /dev/modem
](=/dev/ttyS1).

OK.

]It works great for me but no ppp0 interface appears (I ran "ppp-go"
]without on demand). The modem connects fine (I can see that in

????? what is this supposed to mean? ppp-go is some script which Slack has
set up. Who knows what is in it.


]/var/log/messages) but after Connect: ppp0 <--> /dev/modem it just

Bad idea. by default there is no debugging set up.
place the word
debug
into /etc/ppp/options.
place the line
daemon.*;local2.* /var/log/ppplog
into /etc/syslog.conf
run the command
killall -1 syslogd

Now run pppd and look in /var/log/ppplog.

You have a ppp0 interface. (Note the message SAID it connected teh ppp0
interface to /dev/modem.)
Does /dev/modem exist and point to the proper serial line?


]sits there until a hang up about a minute later (with another ISP it
]did not hang up but there was still no ppp0). This is what that part
]of the log reports:

]Connect: ppp0 <--> /dev/modem
]Modem hangup
]Connection terminated
]Exit

No debug and wrong file.


]I also find it strange that pppsetup reports (as a warning) that it

Slackware has a habit of purposelessly breakinging things like ppp.

]can't find ppp as a module or in the kernel. Perhaps the module was
]not loaded at the time of the setup. Anyway, since I used a bare
]kernel to install (this is an old machine with little ram) I checked
]/usr/src/linux/.config and found PPP as a module. And if I do a lsmod
]I get:

]root@cigar:/usr/src/linux# lsmod
]Module Size Used by Not tainted
]ppp_async 7552 0 (autoclean)
]ppp_generic 15452 0 (autoclean) [ppp_async]
]slhc 5040 0 (autoclean) [ppp_generic]
][snip]

]It did set up my resolv.conf:

]root@cigar:/etc/ppp# cat /etc/resolv.conf
]search ig.com.br
]nameserver 200.225.159.124

]For completeness' sake, this is the pap-secrets file (where I changed
]the login info)

]root@cigar:/etc/ppp# cat /etc/ppp/pap-secrets
]# PAP authentication file: /etc/ppp/pap-secrets
]# This file should have a permission of 600.
]# ~# chmod 600 /etc/ppp/pap-secrets
]# Username Server Password IP addresses
]"catbert" * "ktex23#"

]Needless to say, this account works from my Windows PC.

]This is in my options file:
]root@cigar:/etc/ppp# cat options
]lock
]defaultroute
]noipdefault
]modem
]/dev/modem
]57600
]crtscts
]passive
]asyncmap 0

Put in debug


]This is in my chat script file:

]root@cigar:/etc/ppp# cat pppscript
]TIMEOUT 60
]ABORT ERROR
]ABORT BUSY
]ABORT "NO CARRIER"
]ABORT "NO DIALTONE"
]"" ATZ OK ATX3 OK

Use AT&F (or AT&F1 for a Sportster modem) not ATZ

]"atdt0,33550168"
]TIMEOUT 75
]CONNECT

]If I use minicom to dial, I get to the ppp header:

]atz
]OK
]atx3
]OK
]atdt0,33655000
]CONNECT 31200/ARQ
]~ÿ}#.!}!}!} }8}"}&} }*} } }#}$.#}%}&bùól}'}"}(}"å.~ [snip]

]Can anyone point me to a solution?

No we need more info.

www.theory.physics.ubc.ca/ppp-linux.html


]Many thanks.

]Miguel

miguelk <(E-Mail Removed)> wrote:
> I'm using pppsetup (pppd 2.4.1) on Slackware 9.1. I'm also using a US
> Robotics, Courier V.EVERYTHING external modem on /dev/modem
> (=/dev/ttyS1).

I'm a Slackware fan but that script should be taken out and shot.
....

> This is in my chat script file:

> root@cigar:/etc/ppp# cat pppscript
> TIMEOUT 60
> ABORT ERROR
> ABORT BUSY
> ABORT "NO CARRIER"
> ABORT "NO DIALTONE"
> "" ATZ OK ATX3 OK

Instead of the above line I'd suggest

"" ATZ&F OK

unless there is a good reason for ignoring the dial-tone (ATX3).

> "atdt0,33550168"
> TIMEOUT 75
> CONNECT

This is very likely the problem. CONNECT followed by only a newline
will cause chat to send a carriage-return and often, but not always,
this is a signal to the ISP that a login is desired.

Use

CONNECT \d\c

instead. That causes chat to delay by one second and then send nothing.
Pppd will start PPP negotiations, the ISP should also start them, and the
PPP link established - in the absence of any other problem.

> If I use minicom to dial, I get to the ppp header:

> atz
> OK
> atx3
> OK
> atdt0,33655000
> CONNECT 31200/ARQ
> ~ÿ}#.!}!}!} }8}"}&} }*} } }#}$.#}%}&bùól}'}"}(}"å.~ [snip]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^
Here the ISP starts pppd and expects it's client to do the same.
If I'm right then the carriage-return sent by pppd causes the ISP to
present a prompt and it does not respond to any PPP link negotiations
it receives from pppd or send any itself. Stalemate, and timeout on
one side or the other.

--
Clifford Kite Email: "echo xvgr_yvahk-(E-Mail Removed)|rot13"
PPP-Q&A links, downloads: http://ckite.no-ip.net/
/* My confidence in this answer (X), on a scale of 0 to 10:
|----|----|----|----|----|----|----|----|----|----|
0----1----2----3----4----5----6----7----8----9-X--10 */

Bill Unruh <(E-Mail Removed)> wrote:

> Slackware has a habit of purposelessly breakinging things like ppp.

And no other distribution does?

--
Clifford Kite Email: "echo xvgr_yvahk-(E-Mail Removed)|rot13"
PPP-Q&A links, downloads: http://ckite.no-ip.net/
/* For every credibility gap, there is a gullibility fill.
-- R. Clopton */

Clifford Kite <(E-Mail Removed)> writes:

]Bill Unruh <(E-Mail Removed)> wrote:

]> Slackware has a habit of purposelessly breakinging things like ppp.

]And no other distribution does?

Redhat also seems to have that habit. mandrake seemingly does not.

Clifford;

Thanks for your suggestions. Here are the results:

>
> I'm a Slackware fan but that script should be taken out and shot.
> ...

I used it for the initial setup of the scripts but it seems to do the
job. :-) I later abandoned it.

> Instead of the above line I'd suggest
>
> "" ATZ&F OK
>
> unless there is a good reason for ignoring the dial-tone (ATX3).

I found it helps sometimes, so it's really out of habit.

> > "atdt0,33550168"
> > TIMEOUT 75
> > CONNECT
>
> This is very likely the problem. CONNECT followed by only a newline
> will cause chat to send a carriage-return and often, but not always,
> this is a signal to the ISP that a login is desired.
>
> Use
>
> CONNECT \d\c

I now have this chat script:

root@cigar:/etc/ppp# cat pppscript
TIMEOUT 30
ABORT ERROR
ABORT BUSY
ABORT "NO CARRIER"
ABORT "NO DIALTONE"
"" ATZ&F OK
"atdt0,33550168"
CONNECT \d\c

and this options file:

root@cigar:/etc/ppp# cat options
lock
defaultroute
noipdefault
modem
/dev/ttyS1
57600
crtscts
debug
noauth
passive
asyncmap 0
name "catbert_ktex"


> Here the ISP starts pppd and expects it's client to do the same.
> If I'm right then the carriage-return sent by pppd causes the ISP to
> present a prompt and it does not respond to any PPP link negotiations
> it receives from pppd or send any itself. Stalemate, and timeout on
> one side or the other.

I now stopped using "ppp-go" and start the pppd like this:

/usr/sbin/pppd -detach file /etc/ppp/options connect "/usr/sbin/chat
-v -f /etc/ppp/pppscript"

The output I get in the log file (/var/log/ppp) is:

Sep 20 16:12:25 cigar pppd[1008]: pppd 2.4.1 started by root, uid 0
Sep 20 16:12:26 cigar chat[1010]: timeout set to 30 seconds
Sep 20 16:12:26 cigar chat[1010]: abort on (ERROR)
Sep 20 16:12:26 cigar chat[1010]: abort on (BUSY)
Sep 20 16:12:26 cigar chat[1010]: abort on (NO CARRIER)
Sep 20 16:12:26 cigar chat[1010]: abort on (NO DIALTONE)
Sep 20 16:12:26 cigar chat[1010]: send (ATZ&F^M)
Sep 20 16:12:26 cigar chat[1010]: expect (OK)
Sep 20 16:12:26 cigar chat[1010]: ATZ&F^M^M
Sep 20 16:12:26 cigar chat[1010]: OK
Sep 20 16:12:26 cigar chat[1010]: -- got it
Sep 20 16:12:26 cigar chat[1010]: send (atdt0,33550168^M)
Sep 20 16:12:27 cigar chat[1010]: expect (CONNECT)
Sep 20 16:12:27 cigar chat[1010]: ^M
Sep 20 16:12:49 cigar chat[1010]: atdt0,33550168^M^M
Sep 20 16:12:49 cigar chat[1010]: CONNECT
Sep 20 16:12:49 cigar chat[1010]: -- got it
Sep 20 16:12:49 cigar chat[1010]: send (\d)
Sep 20 16:12:50 cigar pppd[1008]: Serial connection established.
Sep 20 16:12:50 cigar pppd[1008]: using channel 3
Sep 20 16:12:50 cigar pppd[1008]: Using interface ppp0
Sep 20 16:12:50 cigar pppd[1008]: Connect: ppp0 <--> /dev/ttyS1
Sep 20 16:12:51 cigar pppd[1008]: sent [LCP ConfReq id=0x1 <asyncmap
0x0> <magic 0xb75677cd> <pcomp> <accomp>]
Sep 20 16:13:19 cigar last message repeated 9 times
Sep 20 16:13:22 cigar pppd[1008]: LCP: timeout sending Config-Requests
Sep 20 16:13:26 cigar pppd[1008]: Terminating on signal 2.
Sep 20 16:13:26 cigar pppd[1008]: Connection terminated.
Sep 20 16:13:27 cigar pppd[1008]: Exit.

Where it says "Terminating..." is where I gave the console a
[crtl][c].
]

So it looks like the remote host is not recognising or refusing to
send information.

I also did another test like this:

/usr/sbin/pppd /dev/ttyS1 57600 debug connect "/usr/sbin/chat -v ''
ATDT0,33655000 CONNECT '\d\c'"

and this is the log:

Sep 20 16:35:36 cigar pppd[1054]: pppd 2.4.1 started by root, uid 0
Sep 20 16:35:37 cigar chat[1055]: send (ATDT0,33655000^M)
Sep 20 16:35:37 cigar chat[1055]: expect (CONNECT)
Sep 20 16:36:00 cigar chat[1055]: ATDT0,33655000^M^M
Sep 20 16:36:00 cigar chat[1055]: CONNECT
Sep 20 16:36:00 cigar chat[1055]: -- got it
Sep 20 16:36:00 cigar chat[1055]: send (\d)
Sep 20 16:36:01 cigar pppd[1054]: Serial connection established.
Sep 20 16:36:01 cigar pppd[1054]: using channel 4
Sep 20 16:36:01 cigar pppd[1054]: Using interface ppp0
Sep 20 16:36:02 cigar pppd[1054]: Connect: ppp0 <--> /dev/ttyS1
Sep 20 16:36:03 cigar pppd[1054]: sent [LCP ConfReq id=0x1 <asyncmap
0x0> <magic 0x244ee6f0> <pcomp> <accomp>]
Sep 20 16:36:30 cigar last message repeated 9 times
Sep 20 16:36:33 cigar pppd[1054]: LCP: timeout sending Config-Requests
Sep 20 16:36:49 cigar pppd[1054]: Terminating on signal 15.
Sep 20 16:36:49 cigar pppd[1054]: Connection terminated.
Sep 20 16:36:49 cigar pppd[1054]: Hangup (SIGHUP)
Sep 20 16:36:49 cigar pppd[1054]: Exit.

Terribly frustrating. 4 years ago I set this up with diald on Suse 5.3
and Red Hat 6.0 without any problems at all (except figuring out how
it works the first time). However then I was also controlling the
other end (Win NT RAS).

Do you have any comments or clues?

Miguel

Bill;

Thanks for the insight. Here are some comments. I had also sent you a
direct email last night so I'll also answer that here so it's of use
to others in the future.

> ]It works great for me but no ppp0 interface appears (I ran "ppp-go"
> ]without on demand). The modem connects fine (I can see that in
>
> ????? what is this supposed to mean? ppp-go is some script which Slack has
> set up. Who knows what is in it.

I stopped using ppp-go and start pppd like this:

/usr/sbin/pppd -detach file /etc/ppp/options connect "/usr/sbin/chat
-v -f /etc/ppp/pppscript"

with this /etc/ppp/options file:

root@cigar:/etc/ppp# cat options
lock
defaultroute
noipdefault
modem
/dev/ttyS1
57600
crtscts
debug
noauth
passive
asyncmap 0
name "catbert_ktex"

and this chat script file /etc/ppp/pppscripts:

root@cigar:/etc/ppp# cat pppscript
TIMEOUT 30
ABORT ERROR
ABORT BUSY
ABORT "NO CARRIER"
ABORT "NO DIALTONE"
"" ATZ&F OK
"atdt0,3365000"
CONNECT \d\c

This is the log output:

Sep 20 16:48:43 cigar pppd[1092]: pppd 2.4.1 started by root, uid 0
Sep 20 16:48:44 cigar chat[1094]: timeout set to 30 seconds
Sep 20 16:48:44 cigar chat[1094]: abort on (ERROR)
Sep 20 16:48:44 cigar chat[1094]: abort on (BUSY)
Sep 20 16:48:44 cigar chat[1094]: abort on (NO CARRIER)
Sep 20 16:48:44 cigar chat[1094]: abort on (NO DIALTONE)
Sep 20 16:48:44 cigar chat[1094]: send (ATZ&F^M)
Sep 20 16:48:45 cigar chat[1094]: expect (OK)
Sep 20 16:48:45 cigar chat[1094]: ATZ&F^M^M
Sep 20 16:48:45 cigar chat[1094]: OK
Sep 20 16:48:45 cigar chat[1094]: -- got it
Sep 20 16:48:45 cigar chat[1094]: send (atdt0,33655000^M)
Sep 20 16:48:45 cigar chat[1094]: expect (CONNECT)
Sep 20 16:48:45 cigar chat[1094]: ^M
Sep 20 16:49:09 cigar chat[1094]: atdt0,33655000^M^M
Sep 20 16:49:09 cigar chat[1094]: CONNECT
Sep 20 16:49:09 cigar chat[1094]: -- got it
Sep 20 16:49:09 cigar chat[1094]: send (\d)
Sep 20 16:49:10 cigar pppd[1092]: Serial connection established.
Sep 20 16:49:10 cigar pppd[1092]: using channel 5
Sep 20 16:49:10 cigar pppd[1092]: Using interface ppp0
Sep 20 16:49:10 cigar pppd[1092]: Connect: ppp0 <--> /dev/ttyS1
Sep 20 16:49:11 cigar pppd[1092]: sent [LCP ConfReq id=0x1 <asyncmap
0x0> <magic 0x877fbdd9> <pcomp> <accomp>]
Sep 20 16:49:38 cigar last message repeated 9 times
Sep 20 16:49:41 cigar pppd[1092]: LCP: timeout sending Config-Requests
Sep 20 16:49:54 cigar pppd[1092]: Terminating on signal 2.
Sep 20 16:49:54 cigar pppd[1092]: Connection terminated.
Sep 20 16:49:55 cigar pppd[1092]: Exit.


> daemon.*;local2.* /var/log/ppplog
> into /etc/syslog.conf
> run the command
> killall -1 syslogd

I did that and actually rebooted (for other reasons).

This is what my syslog.conf looks like (the last 2 lines):

local2.* /var/log/ppp
daemon.* /var/log/ppp

> Does /dev/modem exist and point to the proper serial line?

I now use /dev/ttyS1

> No debug and wrong file.

Do you mean I'm supposed to get more info in /var/log/ppp than I
posted? All I see is what I posted.

Some other points about this problem. When I use minicom, I don't
always get the login prompt. But when I do I get this:

UQKT2login:

What I noticed is that I get this sometimes when the modem connects at
28800 and never at 31200 (may be a conicidence). So I think that the
tests I did with the 'ogin: assword:' scriting may be failing becuase
it does not request the login that time.

This modem uses AT&F1 for hardware flow control (which is it's
default). If I put that into the script I can't get the modem to dial.
It picks up the line and drops immediately and finally times out with
NO DIALTONE.

Any ideas are much appreciated.

Miguel

P.S. My typo in the script about the addt instead of atdt is because
of the late hour! Just a reminder not to do these things late at
night.

miguelk <(E-Mail Removed)> wrote:

> I now have this chat script:

> root@cigar:/etc/ppp# cat pppscript
> TIMEOUT 30
> ABORT ERROR
> ABORT BUSY
> ABORT "NO CARRIER"
> ABORT "NO DIALTONE"
> "" ATZ&F OK
> "atdt0,33550168"
> CONNECT \d\c

> and this options file:

> root@cigar:/etc/ppp# cat options
> lock
> defaultroute
> noipdefault
> modem
> /dev/ttyS1
> 57600

With a 16550A UART this should probably be 115200. The data coming from
the serial device to your host is faster than the serial connection speed
(the modem output is decompressed data).

> crtscts
> debug
> noauth
> passive
> asyncmap 0
> name "catbert_ktex"
....

> I now stopped using "ppp-go" and start the pppd like this:

> /usr/sbin/pppd -detach file /etc/ppp/options connect "/usr/sbin/chat
> -v -f /etc/ppp/pppscript"

You don't need or want to specify "file /etc/ppp/options" since the
options file is automatically read by pppd.

> The output I get in the log file (/var/log/ppp) is:

> Sep 20 16:12:25 cigar pppd[1008]: pppd 2.4.1 started by root, uid 0
> Sep 20 16:12:26 cigar chat[1010]: timeout set to 30 seconds
> Sep 20 16:12:26 cigar chat[1010]: abort on (ERROR)
> Sep 20 16:12:26 cigar chat[1010]: abort on (BUSY)
> Sep 20 16:12:26 cigar chat[1010]: abort on (NO CARRIER)
> Sep 20 16:12:26 cigar chat[1010]: abort on (NO DIALTONE)
> Sep 20 16:12:26 cigar chat[1010]: send (ATZ&F^M)
> Sep 20 16:12:26 cigar chat[1010]: expect (OK)
> Sep 20 16:12:26 cigar chat[1010]: ATZ&F^M^M
> Sep 20 16:12:26 cigar chat[1010]: OK
> Sep 20 16:12:26 cigar chat[1010]: -- got it
> Sep 20 16:12:26 cigar chat[1010]: send (atdt0,33550168^M)
> Sep 20 16:12:27 cigar chat[1010]: expect (CONNECT)
> Sep 20 16:12:27 cigar chat[1010]: ^M
> Sep 20 16:12:49 cigar chat[1010]: atdt0,33550168^M^M
> Sep 20 16:12:49 cigar chat[1010]: CONNECT
> Sep 20 16:12:49 cigar chat[1010]: -- got it
> Sep 20 16:12:49 cigar chat[1010]: send (\d)

This is what you should see, so that part is okay and I may have been
wrong in thinking the chat script was totally responsible.

> Sep 20 16:12:50 cigar pppd[1008]: Serial connection established.
> Sep 20 16:12:50 cigar pppd[1008]: using channel 3
> Sep 20 16:12:50 cigar pppd[1008]: Using interface ppp0
> Sep 20 16:12:50 cigar pppd[1008]: Connect: ppp0 <--> /dev/ttyS1
> Sep 20 16:12:51 cigar pppd[1008]: sent [LCP ConfReq id=0x1 <asyncmap
> 0x0> <magic 0xb75677cd> <pcomp> <accomp>]
> Sep 20 16:13:19 cigar last message repeated 9 times
> Sep 20 16:13:22 cigar pppd[1008]: LCP: timeout sending Config-Requests
> Sep 20 16:13:26 cigar pppd[1008]: Terminating on signal 2.

Keyboard interrupt.

> Sep 20 16:13:26 cigar pppd[1008]: Connection terminated.
> Sep 20 16:13:27 cigar pppd[1008]: Exit.

The other side not responding to anything sent at this point, when it
is reasonably sure that the ISP is not presenting a prompt, is typical
of a mis-configured UART type for serial device file. The UART should
be a 16550A (not a 16550 even if the device documentation says that)
for most PC serial devices, although an old serial device could have a
16450 UART; you can check the configured UART for the device file with

setserial /dev/ttyS1 -a

> Where it says "Terminating..." is where I gave the console a
> [crtl][c].
> ]

> So it looks like the remote host is not recognising or refusing to
> send information.

Yes, and pppd is not receiving any LCP requests from it.

> I also did another test like this:

> /usr/sbin/pppd /dev/ttyS1 57600 debug connect "/usr/sbin/chat -v ''
> ATDT0,33655000 CONNECT '\d\c'"

I should have read further before staring my response. :/ Nothing of
real significance changed in the log though.

--
Clifford Kite Email: "echo xvgr_yvahk-(E-Mail Removed)|rot13"
PPP-Q&A links, downloads: http://ckite.no-ip.net/

After the suggestions and tests on Oct 1, 2004, I took the modem to
another server with Fedora 1 (updated last week with yum) and tried it
there since it has a GUI to let me try out kppp. I set up kppp and
connected with this message in /var/log/messages:

Oct 1 17:17:15 bombadil modprobe: modprobe: Can't locate module ppp0
Oct 1 17:17:15 bombadil modprobe: modprobe: Can't locate module ppp0
Oct 1 17:17:15 bombadil pppd[2198]: pppd 2.4.1 started by root, uid 0
Oct 1 17:17:15 bombadil pppd[2198]: Using interface ppp0
Oct 1 17:17:15 bombadil pppd[2198]: Connect: ppp0 <--> /dev/ttyS0
Oct 1 17:17:19 bombadil pppd[2198]: Remote message: Enter PPP Mode...
Oct 1 17:17:19 bombadil modprobe: modprobe: Can't locate module
ppp-compress-21
Oct 1 17:17:19 bombadil modprobe: modprobe: Can't locate module
ppp-compress-21
Oct 1 17:17:19 bombadil pppd[2198]: not replacing existing default
route to eth1 [10.1.10.215]
Oct 1 17:17:19 bombadil pppd[2198]: local IP address 200.98.10.196
Oct 1 17:17:19 bombadil pppd[2198]: remote IP address 200.221.25.34
Oct 1 17:17:34 bombadil pppd[2198]: Terminating on signal 15.
Oct 1 17:17:34 bombadil pppd[2198]: Connection terminated.
Oct 1 17:17:34 bombadil pppd[2198]: Connect time 0.4 minutes.
Oct 1 17:17:34 bombadil pppd[2198]: Sent 58 bytes, received 40 bytes.
Oct 1 17:17:34 bombadil pppd[2198]: Exit.

In order to connect I had to configure kppp with CR/LF line
termination and 38400 baud.

Then I tried the command line pppd:

pppd /dev/ttyS0 38400 -detach debug connect "/usr/sbin/chat -v ''
ATDT0,33550168 CONNECT '\d\c'"

I got this output:

Serial connection established.
using channel 3
Using interface ppp0
Connect: ppp0 <--> /dev/ttyS0
sent [LCP ConfReq id=0x1 <asyncmap 0x0> <magic 0x2888118a> <pcomp>
<accomp>]
rcvd [LCP ConfReq id=0x1 <mru 1500> <asyncmap 0xa0000> <auth pap>
<magic 0x1cab> <pcomp> <accomp>]
sent [LCP ConfRej id=0x1 <auth pap>]
rcvd [LCP ConfReq id=0x2 <mru 1500> <asyncmap 0xa0000> <magic 0x1cab>
<pcomp> <accomp>]
sent [LCP ConfAck id=0x2 <mru 1500> <asyncmap 0xa0000> <magic 0x1cab>
<pcomp> <accomp>]
rcvd [LCP ConfReq id=0x2 <mru 1500> <asyncmap 0xa0000> <magic 0x1cab>
<pcomp> <accomp>]
sent [LCP ConfAck id=0x2 <mru 1500> <asyncmap 0xa0000> <magic 0x1cab>
<pcomp> <accomp>]
sent [LCP ConfReq id=0x1 <asyncmap 0x0> <magic 0x2888118a> <pcomp>
<accomp>]
rcvd [LCP ConfAck id=0x1 <asyncmap 0x0> <magic 0x2888118a> <pcomp>
<accomp>]
sent [IPCP ConfReq id=0x1 <addr 10.1.9.170> <compress VJ 0f 01>]
sent [CCP ConfReq id=0x1 <deflate 15> <deflate(old#) 15>]
rcvd [LCP TermReq id=0x3]
LCP terminated by peer
sent [LCP TermAck id=0x3]
Modem hangup
Connection terminated.

So it looks like pppd seems to work on this server.

Aftre trying it a few times I got this:

[snip]
rcvd [LCP TermReq id=0x3]
LCP terminated by peer
sent [LCP TermAck id=0x3]
rcvd [LCP ProtRej id=0x1 80 fd 01 01 00 0c 1a 04 78 00 18 04 78 00]
Modem hangup
Connection terminated.

I believe my questions now have to be:

a) since it looks like pppd sent the pap authentication but the other
end refused it, how would I know if it sent the right pap strings? In
other words, where would pppd look for the pap-secrets file by
default?

b) how do I duplicate the CR/LF line terminator with pppd?

b) would "LCP ProtRe" mean that there is a protocol problem? Since the
remote end takes a variable time to cut off, I think I have to send
the remote end "something" which it is not getting. If so, how should
this be approached?

Miguel

I have solved the problem, albeit on Fedora for now. I'm sure it will
also work on Slack 9.1 and if it doesn't I'll report back. Many thanks
to all that helped!

This is the /etc/ppp/options file:

[root@bombadil ppp]# cat options
/dev/ttyS0
38400
modem
lock
crtscts
debug
default-asyncmap
noauth
defaultroute
passive
connect-delay 500
ipcp-accept-local
ipcp-accept-remote
noipdefault
proxyarp
user my-user
connect "/usr/sbin/chat -v -f /etc/ppp/pppscript"

(this probably has more than it needs)

and this the /etc/ppp/pppscript file:

[root@bombadil ppp]# cat pppscript
TIMEOUT 60
ABORT ERROR
ABORT BUSY
ABORT "NO CARRIER"
ABORT "NO DIALTONE"
"" AT&F OK
"ATDT0,33550168"
CONNECT \d\c


This is how I call pppd:

pppd file /etc/ppp/options

And this is the result:

[root@bombadil etc]# pppd file /etc/ppp/options
Serial connection established.
using channel 41
Using interface ppp0
Connect: ppp0 <--> /dev/ttyS0
sent [LCP ConfReq id=0x1 <magic 0x87001b7> <pcomp> <accomp>]
rcvd [LCP ConfReq id=0x1 <mru 1500> <asyncmap 0xa0000> <auth pap>
<magic 0x9c84b7ee> <pcomp> <accomp>]
sent [LCP ConfRej id=0x1 <asyncmap 0xa0000>]
rcvd [LCP ConfReq id=0x2 <mru 1500> <auth pap> <magic 0x9c84b7ee>
<pcomp> <accomp>]
sent [LCP ConfAck id=0x2 <mru 1500> <auth pap> <magic 0x9c84b7ee>
<pcomp> <accomp>]
sent [LCP ConfReq id=0x1 <magic 0x87001b7> <pcomp> <accomp>]
rcvd [LCP ConfAck id=0x1 <magic 0x87001b7> <pcomp> <accomp>]
sent [PAP AuthReq id=0x1 user="(E-Mail Removed)"
password=<hidden>]
rcvd [PAP AuthAck id=0x1 "Enter PPP Mode..."]
Remote message: Enter PPP Mode...
sent [IPCP ConfReq id=0x1 <addr 0.0.0.0> <compress VJ 0f 01>]
sent [CCP ConfReq id=0x1 <deflate 15> <deflate(old#) 15>]
rcvd [IPCP ConfReq id=0x1 <addr 200.221.25.53>]
sent [IPCP ConfAck id=0x1 <addr 200.221.25.53>]
rcvd [IPCP ConfRej id=0x1 <compress VJ 0f 01>]
sent [IPCP ConfReq id=0x2 <addr 0.0.0.0>]
rcvd [LCP ProtRej id=0x1 80 fd 01 01 00 0c 1a 04 78 00 18 04 78 00]
rcvd [IPCP ConfNak id=0x2 <addr 200.98.37.237>]
sent [IPCP ConfReq id=0x3 <addr 200.98.37.237>]
rcvd [IPCP ConfAck id=0x3 <addr 200.98.37.237>]
not replacing existing default route to eth1 [10.1.10.215]
Cannot determine ethernet address for proxy ARP
local IP address 200.98.37.237
remote IP address 200.221.25.53
Script /etc/ppp/ip-up started (pid 3276)
Script /etc/ppp/ip-up finished (pid 3276), status = 0x0
Terminating on signal 2.
Script /etc/ppp/ip-down started (pid 3279)
sent [LCP TermReq id=0x2 "User request"]
Script /etc/ppp/ip-down finished (pid 3279), status = 0x0
rcvd [LCP TermAck id=0x2]
Connection terminated.
Connect time 1.4 minutes.
Sent 58 bytes, received 40 bytes.

[root@bombadil ppp]# ifconfig
[snip]
ppp0 Link encap:Point-to-Point Protocol
inet addr:200.98.37.237 P-t-P:200.221.25.53
Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
RX packets:4 errors:0 dropped:0 overruns:0 frame:0
TX packets:5 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3
RX bytes:40 (40.0 b) TX bytes:58 (58.0 b)

[root@bombadil ppp]# ping 200.221.25.38 -c 2
PING 200.221.25.38 (200.221.25.38) 56(84) bytes of data.
64 bytes from 200.221.25.38: icmp_seq=0 ttl=254 time=207 ms
64 bytes from 200.221.25.38: icmp_seq=1 ttl=254 time=219 ms

--- 200.221.25.38 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1017ms
rtt min/avg/max/mdev = 207.801/213.893/219.986/6.110 ms, pipe 2

Note: In my case I did not get the default route set uo beacuse I
already have on on this machine.

I had made several mistakes in order to get this going:

1) I did not have the user option
2) My speed was set too high (57600 will not connect)
3) typing errors in files (late at night....)
4) probably used bad modem AT sequences
5) did not have '\d\c' after CONNECT originally

Thanks.

Miguel