Re: SCTP Heartbeat
Michael Tüxen <[email protected]>
| Newsgroups | gmane.ietf.sigtran |
|---|---|
| Message-ID | <[email protected]> |
On Mar 19, 2010, at 7:44 AM, Srinivas Murthy wrote: > Hi, > > I have a confusion in Section 8.3 of RFC 4960. > As per the RFC it says, > “Only one heartbeat should be sent each time the heartbeat timer > expires (if multiple destinations are idle). It is an implementation > decision on how to choose which of the candidate idle destinations > to heartbeat to (if more than one destination is idle).” > > Does that mean if we have an association with say 2 destination addresses and both are idle, we have to send heartbeat to only one of them per heartbeat interval? and as the selection is implementation decision we can choose any one of them to send heartbeat to. > In this case if my implementation always chooses the same destination address then how will we know about the health of the other destination address? Hi Srinivas, first of all: SCTP issues should be discussed on [email protected]. The intention of RFC 4960 is that all idle paths are checked. So always choosing the same idle path is a bad implementation decision. I'm also wondering if it really is a good choice to send a HB every HB.timeout+RTO+jitter. It make much more sense to do this for every idle path every HB.timeout+RTO+jitter. Best regards Michael > > Regards, > Srinivas Murthy > > _______________________________________________ > Sigtran mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/sigtran