[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Sip] Retransmission of INVITEs
- To: Jonathan Rosenberg <jdrosen@dynamicsoft.com>
- Subject: Re: [Sip] Retransmission of INVITEs
- From: Gonzalo Camarillo <Gonzalo.Camarillo@lmf.ericsson.se>
- Date: Thu, 24 Oct 2002 15:25:45 +0300
- Cc: Jiri Kuthan <jiri.kuthan@fokus.fraunhofer.de>, Igor Slepchin <ISlepchin@dynamicsoft.com>, "'Jiri Kuthan'" <jiri.kuthan@fokus.gmd.de>, "'Jiri Kuthan'" <jiri@iptel.org>, Medhavi Bhatia <mbhatia@nextone.com>, James Undery <jundery@ubiquity.net>, "SIP (E-mail)" <sip@ietf.org>, Henning Schulzrinne <hgs@cs.columbia.edu>
- List-help: <mailto:sip-request@ietf.org?subject=help>
- List-id: Session Initiation Protocol <sip.ietf.org>
- List-post: <mailto:sip@ietf.org>
- List-subscribe: <https://www1.ietf.org/mailman/listinfo/sip>,<mailto:sip-request@ietf.org?subject=subscribe>
- List-unsubscribe: <https://www1.ietf.org/mailman/listinfo/sip>,<mailto:sip-request@ietf.org?subject=unsubscribe>
- References: <5.1.0.14.0.20021004221532.03e50f20@mailhost.fokus.gmd.de> <3DA3C2DE.60404@dynamicsoft.com>
- Sender: sip-admin@ietf.org
Jonathan,
> I wonder what is the
> issue with TCP keepalives that make them infrequently used?
As Henning pointed out, they are seldom turned on. The problem is that
TCP keep-alives can terminate an alright TCP connection when there are
transient network failures.
For example, if we are playing a chess game over TCP, our TCP connection
may be open during days. If at the moment your TCP stack is sending me a
TCP keep alive my access router has crashed, we will lose the connection
because your TCP stack will time out and release it. However, 5 minutes
later my access router is OK again, and we could have gone on playing
for days...
Note, however, that TCP keep-alive time outs are in the order or hours
(two hours is a common value). SIP timeouts at establishment time are
much smaller.
Folks against TCP keepalives usually claim that this type of keepalive
mechanism belongs to the application layer rather than to the tranport
layer.
Regards,
Gonzalo
_______________________________________________
Sip mailing list https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementors@cs.columbia.edu for questions on current sip
Use sipping@ietf.org for new developments on the application of sip