[pskmail] Re: Fwd: QRM, pskmail jamming itself

  • From: Rein Couperus PA0R <rein@xxxxxxxxxxxx>
  • To: pskmail@xxxxxxxxxxxxx
  • Date: Wed, 06 Sep 2006 10:34:57 +0200

Guess what I found on my pskmail display yesterday:

SM0RWO pskmail v.0.1.19
Hi G0HWW, you have No mail
That got me thinking, and in fact the solution is simple. I have still
to implement the 'session byte' in the protocol. Once this is done the
problem will go away. Until now we had only one station on a frequency,
and the 'error' did not occur. I know what I have to do to implement it,
but it is quite some work. All a matter of priorities, I wanted to get a
working version 'prototype' as quickly as possible, to prove the
feasibility of the project.

Now that has been settled we can take care of the 'nitty gritty':)

It would be a good idea to learn some Perl, so you can help...

I have meanwhile implemented:
* a separate mail headers list, which enables you to download only new
headers, so you can leave the mail on the server
* a possibility to update the user record in the server database over
the air via the client
* a 'technical ack' for asynchronous messages. When you send an unproto
message to the server it will respond with 'QSL' so you know the message
was received ok.

These features will be in the next release of server and client which
will be out in the next days.

Meanwhile I will tackle arq.pm to include the 'session byte'. Will take
some time.


Rein PA0R

PS: Keep on testing :)

On Tue, 2006-09-05 at 21:47 +0100, Darren Long wrote:
> Oops. Sent only to Per before...
> Begin forwarded message:
> > From: Darren Long <long.spike@xxxxxxxxxx>
> > Date: 5 September 2006 20:35:25 BDT
> > To: Per Crusefalk <per@xxxxxxxxxxx>
> > Subject: Re: QRM, pskmail jamming itself
> > 
> > 
> > Hi Per & others,
> > 
> > 
> > I'm glad you noticed that.  It looked a bit strange from this end.
> > Seemed to me that G0HWW (me) didn't correctly receive the 'k'
> > message and it all went downhill from there. I have a log
> > transcript, and was going to post it, but thought it may play havoc
> > if someone downloads this email using pskMail.  Any protection
> > against replay attacks?  I guess that the expansion of control codes
> > into <SOH> etc, should make it OK.
> > 
> > 
> > I noticed a few collisions between me and Per's server's
> > transmissions.  This isn't the first time I've noticed a number of
> > consecutive collisions when in a connection that's been perturbed.
> > Eventually the two ends settled into a ping-pong session, but I
> > couldn't get pskMail to actually send a QUIT, so I aborted and left
> > the server to poll until it gave up.
> > 
> > 
> > Another strange thing, which may have been down to finger trouble is
> > this.  One morning recently, I noticed my pskMail/gMFSK trying to
> > talk to Per's server when it was connected with PA0R.  My Signalink
> > rig interface was off, so no danger of transmission, but it was
> > hammering away (if you can call it that) responding to each
> > transmission from SM0RWO.  This had me quite perplexed but I
> > wondered if I had left pskMail  still trying to connect when I
> > switched off the Signalink and went to bed.  I'm not sure exactly
> > happened, but for a while I thought it may have responded to a 'k'
> > message sent to PA0R.
> > 
> > 
> > Strange, huh?  I may have to get more familiar with Perl...
> > 
> > 
> > Darren, G0HWW

Other related posts: