nanog mailing list archives
Re: Large-datagram hop-by-hop delivery mechanism
From: Sean Donelan <SEAN () SDG DRA COM>
Date: Wed, 4 Aug 1999 14:54:27 -0500
mark () vielle datasys NET (Mark R. Lindsey) writes:
I am seeking a mechanism for reliable delivery of large blobs through a network of occasionally-connected hosts that can forward the blobs using available transport protocols.
The phrase "reliable delivery" throws a kink in things. The Inter-Library Loan protocol does this for some definition of "reliable." It is also amazingly complex and obtuse because it tries to deal with every possibility. And most people won't like it because it was originally an OSI protocol. A much simplier solution is to use "e-mail" as the transport layer and to encapsulate a simple method of detecting missing and duplicate messages (e.g. a message serial number). Your application will need to have some method of signaling when a message needs to be retransmitted. A simple return-receipt is usually not sufficient. I have seen several applications which did this, but I don't recall any general purpose standard or RFC. E-mail has a long history, and gateways exist for almost every network protocol in existance, including most store-and-forward protocols (BITNET, UUCP, FIDONET, X.25, DECNET, etc). The MIME folks did a great job coming up with a way to encapsulate almost anything in a mail message. The trick is decapsulating it :-) -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation
Current thread:
- Large-datagram hop-by-hop delivery mechanism Mark R. Lindsey (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Andrew Brown (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Andrew Bangs (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Faust (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Edward S. Marshall (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Forrest W. Christian (Aug 05)
- <Possible follow-ups>
- RE: Large-datagram hop-by-hop delivery mechanism Mohler, Jeff (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Sean Donelan (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Vadim Antonov (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Steve Uurtamo (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Peter Galbavy (Aug 05)
- Re: Large-datagram hop-by-hop delivery mechanism Steve Uurtamo (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Mark R. Lindsey (Aug 04)
- Re: Large-datagram hop-by-hop delivery mechanism Andrew Brown (Aug 04)