Hi,
What you are telling is not flow control or windowing.

Real flow control is  the concept of a receiver informing the sender that it
can’t accept any more data.

Not that configuring statically before the process begins.

Regards
Jyoti Ranjan Panda

-----Original Message-----
From: Nikos Balkanas [mailto:nbalka...@gmail.com]
Sent: Monday, February 02, 2009 2:08 PM
To: jyoti; devel@kannel.org
Subject: Re: Real Windoing in kannel for smpp ?


You could use throughput.

BR,
Nikos
----- Original Message -----
From: "jyoti" <jyoti.ran...@cellnext.com>
To: "Nikos Balkanas" <nbalka...@gmail.com>; <devel@kannel.org>
Sent: Monday, February 02, 2009 10:37 AM
Subject: RE: Real Windoing in kannel for smpp ?


> Hi,
> I want to know the solution for
> : suppose if an ESME  submits messages at a rate that exceeds the
> capabilities of its peer, congestion may occur.
> The ESME will continue to top up its window of unacknowledged requests,
> keeping the SMSC under load to process these requests.
>
> By configuring Wait-ack I think we are making the process static, By using
> congestion_state TLV  we can control windowing at run time.
>
> If I am wrong please suggest.
>
> Thanks & Regards
> Jyoti Ranjan Panda
>
> -----Original Message-----
> From: Nikos Balkanas [mailto:nbalka...@gmail.com]
> Sent: Monday, February 02, 2009 1:35 PM
> To: jyoti; devel@kannel.org
> Subject: Re: Real Windoing in kannel for smpp ?
>
>
> Hi,
>
> Please check parameters wait-ack and wait-ack-expire in User's Guide.
>
> BR,
> Nikos
> ----- Original Message -----
> From: "jyoti" <jyoti.ran...@cellnext.com>
> To: <devel@kannel.org>
> Sent: Monday, February 02, 2009 8:31 AM
> Subject: Real Windoing in kannel for smpp ?
>
>
>> Hi All,
>>
>> Does there any configurable parameter for windowing in smpp(i.e if I have
>> connected to smsc through smpp)?
>>
>> Does congestion_state TLV is used in kannel for smpp? which could be sent
>> in
>> response as parameter to provide real windowing
>> like TCP.
>>
>> Waiting for reply
>>
>> Thanks & Regards
>> Jyoti Ranjan Panda
>>
>>
>
>



Reply via email to