PPP server problems.

PPP server problems.

Post by Richard Béney » Wed, 01 Apr 1998 04:00:00



Hi there,

I'm currently setting up a PPP dialin server for remote access based on a
RedHat 5.0 distrib. Till now everything worked fine and that's pretty cool.

Now, here are the two problems :

1) When the PPP link is established and running, the routes are established
at
both sides... I try to ping the PPP server and that works fine, I try to
ping a computer which is behind  it and that works fine to. Then I try
to telnet a computer behind the PPP server and that's ok, I mean the login
prompt appears immediatly;
BUT when I telnet the IP address of the PPP server it waits several minutes
before sending the
login prompt, WHY ?

2) This one worries me more. I set up a getty 19200 on /dev/ttyS1, because
my uart is a 16450 and I
want to limit speed. Good, but now the getty really expects to have a speed
of 19200 on ttyS1 so if
that's not true I got garbage at the other side instead of the prompt when
I connect. Is there a command to force the speed of the uart at exactly
19200 bps. I use minicom to reset this speed but I'd rather put a command
in rc.local.

Can you help.

Thanks.

--
Richard Bneyt

 
 
 

PPP server problems.

Post by bill davids » Wed, 01 Apr 1998 04:00:00



| I'm currently setting up a PPP dialin server for remote access based on a
| RedHat 5.0 distrib. Till now everything worked fine and that's pretty cool.
|
| Now, here are the two problems :
|
| 1) When the PPP link is established and running, the routes are established
| at
| both sides... I try to ping the PPP server and that works fine, I try to
| ping a computer which is behind  it and that works fine to. Then I try
| to telnet a computer behind the PPP server and that's ok, I mean the login
| prompt appears immediatly;
| BUT when I telnet the IP address of the PPP server it waits several minutes
| before sending the
| login prompt, WHY ?

One possible reason is that the server is running wrappers and is
trying to do address resolution on the incoming telnet before
allowing you a connect. The time is either a timeout or the time to
open a connect to a DNS which can do the lookup.

Try running DNS locally.

| 2) This one worries me more. I set up a getty 19200 on /dev/ttyS1, because
| my uart is a 16450 and I
| want to limit speed. Good, but now the getty really expects to have a speed
| of 19200 on ttyS1 so if
| that's not true I got garbage at the other side instead of the prompt when
| I connect. Is there a command to force the speed of the uart at exactly
| 19200 bps. I use minicom to reset this speed but I'd rather put a command
| in rc.local.

See your modem manual, you should be able to lock the speed, just
connect at the desired speed, issue the command, save the mode, and
exit. Or have getty issue the lock speed at the right speed.
Requires using hardware flow control.
--

    CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

 
 
 

1. PPP server problem (LCP problem?)

Hello,

I have problems configuring a PPP server with FreeBSD.
I have a ppp client with user space ppp and FreeBSD 2.2.1
and a ppp server with kernel ppp and FreeBSD 2.2.2
Both machines use a Zyxel Omni TA 128 Modem at /dev/cuaa0

The client works ok and can dial in a Win95 server without
problems. To dial in the FreeBSD server causes some problems.
The problems seems to be with LCP or the modem configuration.

Server:
I use the ppp/options file from the BSD handbook, changed with my domain
and addresses and debug added (crtscts, netmask ..., <ownip>:<remoteip>,
domain ..., passive, modem, debug)
I used kermit with the /etc/ppp/kermit.ans script
(with line /dev/ttyd0 and speed 19200)

I got following messages from the server:
pppd 2.2.0 started by root, uid 0
using interface ppp0
connect:ppp0 <--> /dev/ttyd0
sent [LCP ConfReq id=0x01 <mru 1500> <magic 0xdbb08dd5> <pcomp> <accomp>]
last message repeated 4 times
hangup (SIGHUP)
Modem hangup
Exit

Client:
I have set device /dev/cuaa0, speed 19200 and openmode active.
The AT commands for the modem are ATE1Q0B40
I got following messages from the client:
...
*Connected
LCP: state change Initial --> Closed
LCP: sendConfigReq
  ACFCOMP
  PROTOCOMP
  ACCMAP[6] 00000000
  MRU [4] 1500
  MAGICNUM [6] 2df3efd5
LCP: state change Closed --> Req-Sent
LCP: Received Configure Reject (1) state = Req-Sent (6)
LCP: RecvConfigReq
  ACCMAP[6] 00000000
  MRU [4] 1500
  MAGICNUM [6] 2df3efd5
<The last for lines are 4 times repeated>
LCP: state change Req-Sent --> Stopped
LCP: Layer Finish
Disconnected!
...

If I use user space ppp at the server side I got the same
behaviour at the client side but no debug info at the server
side. For a short time the getty V19200 ttyd0 process changes
to login -r ?#{?{ or a similar process and then getty is
started again.
Can anybody tell me what's going wrong or give me any hints,
what I can do?

Many thanks
        Elmar

--
====================================================
Elmar Haas, Siemens AG, ZT IK 2, Munich, Germany

====================================================

2. modem problem

3. PPP Server problem or modem problem

4. New flaky behavior

5. PPP server problem

6. Q: Why aren't users showing up in utmp?

7. ppp server problem

8. Help: comparing of mod_ntlm mod_auth_ntdom mod_auth_pam AuthenSmb

9. PPP Server problem

10. PPP-Server Problem

11. Dial up PPP Server problem

12. ppp server problem with Redhat 7.2 Kernel 2.4.7-10

13. ppp server problem