home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!kronos.arc.nasa.gov!iscnvx!news
- From: myoung@force.ssd.lmsc.lockheed.com
- Subject: SNMP and DIagnostics
- Message-ID: <1992Nov9.141013.6243@iscnvx.lmsc.lockheed.com>
- Sender: news@iscnvx.lmsc.lockheed.com (News)
- Reply-To: myoung@force.ssd.lmsc.lockheed.com
- Organization: LMSC, Sunnyvale, California
- Date: Mon, 9 Nov 92 14:10:13 GMT
- Lines: 38
-
- SNMP
-
- The SNMP community must have considered mechanisms for supporting SNMP
- objects over cell relay. Path tables, node parameters, and physical
- device parameters must be read as SNMP objects.
-
- Currently SNMP is designed for TCP/IP support, but I think some work
- is being done to make SNMP protocol independent at that layer. Have
- the SNMP vendors signed onto the SSCOP signaling layer as a carrier
- protocol?
-
- And Diagnostics:
-
- At a deeper level, what about sniffer protocols? Don't we need some
- capability to snoop through the virtual path tables, following some
- path down its multi-hop links, reading out the path tables in route,
- possibly splitting the snooper packet as the virtual path splits in a
- one to many circuit. Or how about finding a route around a particular
- physical link, then passing data through the link, bringing back the
- results for verification. Or how about building a branch from an existing
- path, and bring the cell stream to some embedded protocol analyzer.
-
- It seems that the sniffer protocols would need some more elementary
- access to the path tables, not wanting to rely on SCCOP which it
- might be testing.
-
- Why Not:
-
- SCCOP must have some exploratory "pinging" protocol which it uses to
- discover new links, check for new nodes and build nearest neighbor maps.
- Why isn't this pinging protocol enhanced and made generally available
- as a programmable layer?
-
- There is a lot of software and a lot of protocols available already,
- as well as companies with proprietary requirements. Why not allow this
- existing base of technology and distribution be ported directly to
- the switches, operating their own signaling systems? Why duplicate
- existing software?
-