[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: Re: minutes 21-Dec-2000 tr&p con-call
Mr. Doug Bunting, On Mon, 01 Jan 2001 21:45:43 -0800 Doug Bunting <Doug@ariba.com> wrote: ... > Unfortunately, your description doesn't help with the issue I intended to > raise. I was speaking about your proposed modification (thinking it was > "current"), which added a Reset condition. > > Say the sender sent messages 99, 100 and 101, then reset the sequence number > series. The receiver MSH may receive messages 99, 100 and 0 and think > that's correct for in-order delivery to the application layers. The sender > may be retrying message 101 at about the same time. But, it doesn't help > matters now that the messages have been passed "up". I understand your question. Yes, it is an issue in the SequenceNumber method. I'd like to propose a solution to resolve the issue. Solution: Add following condition to reset the SequenceNumber in the specification: Before the Sender reset the SequenceNumber, the Sender MUST wait for receiving of all the Acknowledgement Messages for all the sent Normal Messages. Only when all the sent NormalMessages are acknowledged, the Sender may reset the SequenceNumber. If the condition is added, the Sender will reset the SequenceNumber and send message 0 after the messages 99, 100 and 101 reached the Receiver in your example. What do you think? Regards, -- SHIMAMURA Masayoshi <shima.masa@jp.fujitsu.com> TEL:+81-45-476-4590(ext.7128-4241) FAX:+81-45-476-4726(ext.7128-6783) Planning Dep., Strategic Planning Div., Software Group, FUJITSU LIMITED
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC