**>To: [email protected]
**>From: Matthew Hixson <[EMAIL PROTECTED]>
**>Subject: message_id uniqueness?
**>Date: Tue, 15 Feb 2005 11:41:52 -0800
**>
**>As I understand things the SMSC is in charge of creating message_ids 
**>for each SMS.  Should there ever be a case where two consecutive SMS 
**>messages are assigned the same message_id?  I ask because this is what 
**>I see from a recent test while sending a ringtone that was chopped into 
**>two SMS messages.  The message_ids in bearerbox.log were identical.
**>  Is this normal?

I guess now you know why Kannel has a 'ts' field instead of a
'unique-id' field in its dlr storage for such a long time ;-).

Looks like your SMPP server does not return unique message-id's to
you.  You should ask them.  They might (as you speculated before)
have a SMPP server that using some variation of time in seconds
as their message-id.  Talk with them to verify this.

About the only recourse would be to tell the they are not in
specs with SMPP v3.4:

  5.2.23 messsage_id

  The unique message identifier reference assigned by the SMSC to
  each submitted short message. It is an opaque value and is set
  according to SMSC implementation. It is returned by the SMSC
  in the submit_sm_resp, submit_multi_resp, deliver_sm_resp and
  data_sm_resp PDUs and my be used by the ESME in subsequent SMPP
  operations relating to the short message, e.g., the ESME can use
  the query_sm operation to query a previously submitted message
  using the SMSC message_id as the message handle.

But, I'm know sure of v3.3. Anyone have the old Logica v3.3 SMPP
specs anywhere?

See ya...

d.c.

Reply via email to