Re: Local interface failure detection by SCTP.
"Brian F. G. Bidulock" <[email protected]>
| Newsgroups | gmane.ietf.sigtran |
|---|---|
| Organization | http://www.openss7.org/ |
| Message-ID | <[email protected]> |
Prathap, See the SCTP I-G discussed on TSVWG. It includes some advise on controlling the source address for user-space implementations. --brian Prathap T wrote: (Sat, 10 Dec 2005 00:38:05) > > How should SCTP detect Local interface failure assuming SCTP is > implemented in the user space. > > > > For ex: > > Assume there is Endpoint A, Endpoint B > > Endpoint A has interface A1, A2. > > > > Endpoint B has interface B1, B2. > > > > > > Endpoint A, communicates to B using it's local interface A1. Now > when A1 goes down, > > how does A, get to know this?? > > > > This can be a problem, when A sends out heartbeat with source as A1 > and destination as B1. > > ( I think IP sends this out on A2 with source as A1) > > and the endpoint B responds with an HBeatAck with destination as A1, > which is no more existing. > > > > Now since A did not receive HBeatAck from B, it will consider B1 not > reachable after some retries. > > This can be a real problem when HBeat Interval time of B is very > high and HBeat interval of A is low. This can lead to breakdown of > association. > > > > So, what is the mechanism SCTP has to employ to detect Local > failure. > > > Regards, > > Prathap. > _______________________________________________ > Sigtran mailing list > [email protected] > https://www1.ietf.org/mailman/listinfo/sigtran -- Brian F. G. Bidulock [email protected] http://www.openss7.org/