jf@ap.co.umist.ac.uk (John Forrest) (04/06/91)
Thanks to all who send help on trying to get Async Appletalk to work. The solution was to get hold of the ethertalk installation disk - there is a copy on apple.com [ hope I'm not supposed to say that ... ]. Having got this to work, I'm interested in working out how far it will go - ie. Appleshare to CAP server works, but is *very* slow. The next question is "can I get MacTCP to work?". That would give telnet and eudora - sure it would be slow, but there may well be less of the other funnies we get, and the speed would be an inducement for anyone to try to get a proper Localtalk or Ethertalk connection. Anyway, MacTCP is quite happy in selecting the Async network icon. It does ask for which zone though - what affect does this have? Anyway, the system doesn't fall over, but I get no response from anywhere. I suspect I have to fit some routing up. Has anyone got this to work, or could someone please tell me how this is likely to appear to the network - ie. would the Mac be logically connected to the Webster Multigate (we are using for comms) or to the Unix box the serial line is connected to? Any help would be appreciated! John Forrest Dept of Computation UMIST
kre@cs.mu.OZ.AU (Robert Elz) (04/07/91)
jf@ap.co.umist.ac.uk (John Forrest) writes: >Anyway, MacTCP is quite happy in selecting the Async >network icon. It does ask for which zone though - what affect does this have? >Anyway, the system doesn't fall over, but I get no response from anywhere. I >suspect I have to fit some routing up. No - the problem is that there's no way to get the multigate to assign an IP address for an async node (yet, this may happen some day), as the async node appears to be on a remote network connected over IPtalk, which isn't something that an IPGATEWAY will normally ever assign addresses to. MacTCP doesn't know this, so tries to do the "right thing", but just doesn't get an answer (you should really get a message about not being able to assign an IP address, or "host or gateway not responding" or something similar). At the minute I don't think there's a solution to this one. kre
tom@wcc.oz.au (Tom Evans) (04/11/91)
Please ignore any previous non-Cancelled versions of this posting. In article <kre.670972368@mundamutti.cs.mu.OZ.AU>, kre@cs.mu.OZ.AU (Robert Elz) writes: > jf@ap.co.umist.ac.uk (John Forrest) writes: > > >Anyway, MacTCP is quite happy in selecting the Async > >network icon. It does ask for which zone though - what affect does this have? The zone you have to select is the zone that MacTCP looks in to find its MacIP gateway - so you can have one centralised MacIP gateway performaing the address assignment and gateway functions. > >Anyway, the system doesn't fall over, but I get no response from anywhere. I > >suspect I have to fit some routing up. > > No - the problem is that there's no way to get the multigate to assign > an IP address for an async node (yet, this may happen some day), as the > async node appears to be on a remote network connected over IPtalk, which > isn't something that an IPGATEWAY will normally ever assign addresses to. I have managed to do this OK. You need one MPG (MultiPort Gateway) performing the Async services, and a second one providing MacIP services out one of its EtherTalk ports, but with IPTalk disabled. You then select the the appropriate EtherTalk zone (that the second one is in) in MacTCP. Now the bad news - it still doesn't work. The combination of MacTCP and NCSA Telnet expects to be on a fast, rapid, LOCAL network, and the timeouts that NCSA selects can't handle the Async speed (or lack thereof). It nearly works at 9600, but not quite. Maybe not even at 19,200 - I gave up there. If you can run 19,200, you can usually run LocalTalk over the same physical path (so why bother with async?). You want it to work at 2400. This is a general problem with NCSA and MacTCP, as the TCP timeouts won't even let it work across the Pacific on a fast net. I would expect it to fail over slower SLIP and Ether half-bridges too. A posting from jackb@MDI.COM (Jack Brindle) on comp.sys.mac.comm dated 5 Apr 91 17:31:28 GMT gives better details on the timeouts. ======================== Tom Evans tom@wcc.oz.au ** ADD ".au" MANUALLY (don't trust "reply") ** Webster Computer Corp P/L, 1270 Ferntree Gully Rd Scoresby, Melbourne 3179 Victoria, Australia 61-3-764-1100 FAX ...764-1179 A.C.N. 004 818 455