LinuxSA Mailing list archives

  From: Alan Kennington <akenning@dog.topology.org>
  To  : darryl@bachblue.com.au
  Date: Wed, 3 Mar 1999 23:51:01 +1030

Re: Modem Problems

Darryl,

Further to tht last message, I had the same problem identically
with the redback external modem, as you know.
That's what made me think the problem was at the
telstra end (a cisco router with 2 Mbps connections to the PSTN).
The guy at Telstra said the problem should go away if I turned
off compression and error control.
But I'm not sure I agree with this.

Here's another snippet of info:
When I had the problem with the redback external modem,
the engineer who looked into it for me
said that he could see the modem at his end _retraining_
for 7 seconds out of every 8.5.
But it was _not_ doing this at my side.
This seems to indicate that either:
-	may signal on the line was not correctly generated, or
-	the cisco router was not correctly locking onto my signal.

Here's an interesting test to do on the loop data content:

Increase the ping packet size with the -s parameter.
E.g. ping -s 1000 gives you a 1000 byte packet.
At some point, the sequence of returend pings starts
to become disjoint. That is, packets are lost.
If you now multiply the packet size by 8, you get the
number of bytes "in the loop", i.e. buffered at the
IP layer at the other end.
I foudn this to be 32 kBytes.
The modem buffer is about 600 bytes -- i know this because
I phoned the manufacturers of the redback modem.
There are certain design reasons for this to be true too.

Anyway, lets look at what features are in common,
and the owest common denominator way indicate the cause.
It may be:
-	linux
-	33.6 kbit/sec modems in general
-	cisco routers

Cheers,
Alan Kennington.

-- 
Check out the LinuxSA web pages at http://www.linuxsa.org.au/
To unsubscribe from the LinuxSA list:
  mail linuxsa-request@linuxsa.org.au with "unsubscribe" as the subject