Re: Throughput patches
Alvin Starr <[email protected]>
| Newsgroups | gmane.comp.mobile.kannel.devel |
|---|---|
| Message-ID | <[email protected]> |
I have patched the smpp code to throttle outbound messages. It seems to be working with our outbound provider and has passed there accptance test. One of the things I have not addressed in my patch( I am sure I have missed others also) it what do do if the input is greater than the output and a huge backlog of messages builds up. I can post the patch. It may be a little out of date and need some fixing. Vincent CHAVANIS wrote: > The main issue is located to the fact we are using the same thread in > order to send/receive(=ack) a message. > We need to think about a global fix which result in modifying the > actual structure of kannel (=> implicates compatibility breaker) > I will take a look when i will have some time. > The main idea to temporatily fix thjis is to check for each msg, if > they have already been received. > If yes, then drop it. If not, insert it into your db. > > Vincent > -- > Telemaque - 06560 SOPHIA-ANTIPOLIS - (FR) > Service Technique/Reseau - NOC > Developpement SMS/MMS/Kiosques > http://www.telemaque.fr/ > [email protected] > > ----- Original Message ----- From: "Juan Nin" <[email protected]> > To: <[email protected]> > Sent: Monday, November 26, 2007 4:10 PM > Subject: Re: Throughput patches > > >> Hi! >> >> Any feedback on this? >> >> Thanks in advance >> >> Juan >> >> On Nov 16, 2007 12:35 PM, Juan Nin <[email protected]> wrote: >>> Hi! >>> >>> I've been reading the different threads about the problem with >>> throughput, being this the last one (I think): >>> http://www.mail-archive.com/[email protected]/msg06847.html >>> >>> Apparently that fixes the outgoing traffic throughput, but impacts >>> incoming ACKs. >>> >>> Is there any new version of the patch which fixes this, or any plans >>> on fixing it soon? >>> Seems to me to be something important, since many SMSCs limit their >>> inbound traffic, and not being able to correctly set the throughput >>> parameters is a big problem. >>> >>> Also, that patch is EMI/UCP specific and Stuart Beck ported it to SMPP. >>> If there are intentions on fixing it on Kannel, will it be coded on >>> each smsc implementation, or will it be abstracted to some more global >>> place (don't know if it's possible) >>> >>> Thanks in advance, >>> >>> Juan >>> >> >> > > > -- Alvin Starr || voice: (416)585-9971 Interlink Connectivity || fax: (416)585-9974 [email protected] ||