Re: [Nsi-wg] Issue 2 in ogf-nsi-project: GenericAcknowledgmentType
Comment #11 on issue 2 by imo...@gmail.com: GenericAcknowledgmentType http://code.google.com/p/ogf-nsi-project/issues/detail?id=2 This was re-discussed on the Wed. call, with the following results: 1. Message transport layer takes care of ack retransmission in case of a packet loss. 2. Confirm/Fail message responses should be sent out without waiting on the receipt of the message ack. [comment on error recovery: if message transport layer is broken, say unidirectionally, the ack sending will eventually timeout and generate a message transport error that the NSA will need to handle] 3. From the receiving state machine perspective, receiving a confirm/fail message from a particular NSA should cancel out the need to receive an ACK. So the state machine should not only continue to process the confirm/fail message, but not gate on or wait for the ACK message. Hope this resolves this issue
Comment #12 on issue 2 by jmacau...@gmail.com: GenericAcknowledgmentType http://code.google.com/p/ogf-nsi-project/issues/detail?id=2 Added additional text to interface WSDL description. Will leave open until CS protocol documentation is fixed. ------------------------------------------------------------------------ r3 | jmacauley@gmail.com | 2011-10-14 14:17:55 -0400 (Fri, 14 Oct 2011) | 1 line Added additional documentation around Acknowledgment handling as per issue #2. Updating issue #2. ------------------------------------------------------------------------
Updates: Status: Done Comment #13 on issue 2 by guy.robe...@gmail.com: GenericAcknowledgmentType http://code.google.com/p/ogf-nsi-project/issues/detail?id=2 NSI connection service protocol version 1.1 updated and issue closed.
participants (1)
-
ogf-nsi-project@googlecode.com