My wife works for a pretty big Enterprise company. They have about 300 sites, all of them have IPX running. All WAN stuff is IPX EIGRP in addition to IP and some SNA. To me it would suggest that bigger Enterprise companies are still in need for network people with good IPX understanding, especially those who know how to control it :) This may be considered by some a legacy stuff, but legacy is what makes those who know it even more valuable...
""Priscilla Oppenheimer"" wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > Thanks. That sounds right to me. By default the router discards a packet if > the IPX hop count reaches 16. But I discovered that you can configure the > number of hops with the "ipx maximum-hops" command. There wouldn't be any > need in a RIP network, because RIP can't learn about a network with 16 or > more hops. (16 means infinity.) But routers running EIGRP and NLSP can > learn about paths that are more than 15 hops away, so it might make sense > in those cases. > > Does anyone care about IPX anymore? IPX RIP? EIGRP for IPX? NLSP for IPX? > > Any feedback would be appreciated. Thanks. > > Priscilla > > At 09:50 PM 10/18/01, Henry D. wrote: > >I'm no expert at this but from I was able to get from cisco's web site is > >that the router discards the packet if the control field is set to 16 or up > >for ipx rip. > >In mixed environment, with both NLSP and RIP running, the router might > >have routes of greater than 16 if it learnt those routes using NLSP,the > >important thing > >would be the servers' configuration. If the server supports only RIP, then > >obviously > >the hop count would still be an issue and the server would discard the RIP > >update > >with 16 and up. To take the full benefit from NLSP and its hop count > >enhancement > >I'd think one would have to run NLSP in the whole network, including the > >servers. > > > >Again, i'm not experienced with IPX... > >""Priscilla Oppenheimer"" wrote in message > >[EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > > > The IPX header has a "transport control" field which is really a "hop > > > count." The sender sets it to zero. Each router adds one to it. > > > > > > Novell documentation used to show it as a 4-bit field with 4 bits > reserved > > > before it. Recent documentation shows it as an 8-bit field. Older > document > > > ion said a router would trash a frame if it arrived with a transport > > > control field already at 15 (0xFFFF). Recently I read this weird thing on > > > Novell's site: > > > > > > A RIP router discards the packet if the value in this field is greater > >than > > > 15. > > > > > > An NLSP router discards the packet if the value in this field is greater > > > than the value of the Hop Count Limit parameter, which is 127 by default. > > > > > > Is this believable? From what we know about the router having two > separate > > > tasks (forwarding and learning the topology), I think the hop-count > limits > > > happen when installing routes. I could believe that RIP and NLSP are > > > different. But when a router goes to forward a frame, is it really going > >to > > > behave differently with respect to hop count if it's running NLSP versus > > > RIP? Does it even care which protocol installed the route. The FIB > >probably > > > wouldn't even say which protocol installed the route? > > > > > > Chuck likes to remind us about these differences so maybe he has some > > > comments. > > > > > > Thanks > > > > > > Priscilla > > > > > > > > > > > > > > > > > > ________________________ > > > > > > Priscilla Oppenheimer > > > http://www.priscilla.com > ________________________ > > Priscilla Oppenheimer > http://www.priscilla.com Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=23517&t=23389 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
