Hi Bruno,
You was looking for a log file with the 31 syntax error.
2001-08-10 12:49:16 [6] DEBUG: emi2 sending packet:
<00/00027/O/31/pass/0539/21>
2001-08-10 12:49:16 [6] DEBUG: Got packet from the main socket
2001-08-10 12:49:16 [6] DEBUG: emi2 parsing packet:
<00/00035/R/31/N/02/Syntax error/F8>
2001-08-10 12:49:16 [6] ERROR: Got negative ack. op:31, trn:0, error:2
(Syntax error), message: Syntax error
I used only username and no password. I do not have one at the SMSC (CMG
EMI), so I guess thats why it fails.
regards
Steffen Lindemann
> -----Original Message-----
> From: Bruno David Sim�es Rodrigues [SMTP:[EMAIL PROTECTED]]
> Sent: Sunday, August 12, 2001 3:36 PM
> To: [EMAIL PROTECTED]
> Subject: Re: emi2: main connection closed by SMSC
>
> I'll remember that my SMSC administrators insist that our-port equals to
> receive-port, and then
> we have two options:
>
> 1. Change the code to only open the listen port when smsc time-out and our
> connection is closed.
> I've almost done this code, but I had one problem, because when we
> connect,
> we can send and
> receive through that connection. when smsc times out, we close the
> connection and open listen
> port. If some new sms is received through smsc, it'll open the connection
> and kannel could, again,
> send and receive through it. My problem was when we are in listen mode
> (waiting for the smsc) and
> kannel have some message to send, I couldn't make kannel code to close the
> listen and restablish
> a new connection to smsc. Then I saw that I must have some kind of state
> (disconnected, connected,
> listen, connect_listen) shared between some threads and then given up (was
> then that I talked with
> SMSC admins and agree on sending the keepalive packet).
>
> 2. As someone told, his smsc doesn't accept -31 packet. why is that? I yet
> don't know. Maybe it's
> the same thing, that he doesn't use username. Then I can change the code
> to
> use something like -61
> to keep the connection alive (as I told you, one vantage of waiting for
> the
> 31-ACK in the login
> stage is that if it is NACKED, kannel could switch to other kind of
> command
> automatically).
>
>
> BTW: could you send me the logs for the 31 syntax error ? do you have or
> not
> username and
> password ? (I've coded that if you don't have username kannel doesn't
> send -31 packets, even
> if keepalive is given -> I think it won't start either if you give
> keepalive
> and don't give username,
> but I'll check)
>
> --
> Bruno Rodrigues
>
>
> ----- Original Message -----
> From: "Andreas Fink" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Friday, August 10, 2001 11:56 AM
> Subject: RE: emi2: main connection closed by SMSC
>
>
> > >I tried now.
> > >The newest CVS version sends a 31 with the username every 45 secs and
> the
> > >smsc make a "syntax error" on it. So I think that is not the way.
> > >It is not sending a 61 messages?
> >
> > Hmm wait, 31 is an alerting message and it needs an ID. if your SMSC
> > does reply with Syntax error then it is not conforming to the EMI/UCP
> > standard 3.3 and above. but if they dont support type 31, that means
> > they send you incoming messages directly to your port. In other
> > words, there is no problem when your link goes out of service since
> > you will repoen it automatically when you have something to send. For
> > receiving you would get the messages sent to you (the session gets
> > opened from the SMSC side) instead of polling for them. If you would
> > poll for messages you would need a 31 message and thus a username.
> >
> > So there isnt any problem. Its only looking like one :-).
> >
> > --
> >
> > Andreas Fink
> > Fink-Consulting
> >
> > ------------------------------------------------------------------
> > Tel: +41-61-6932730 Fax: +41-61-6932729 Mobile: +41-79-2457333
> > Address: A. Fink, Schwarzwaldallee 16, 4058 Basel, Switzerland
> > E-Mail: [EMAIL PROTECTED] Homepage: http://www.finkconsulting.com
> > ------------------------------------------------------------------
> > Something urgent? Try http://www.smsrelay.com/ Nickname afink
> >
> >
>
