[comp.sys.ibm.pc.rt] RT Ethernet problem revisited

gors@well.UUCP (Gordon Stewart) (07/21/89)

As a follow-up to my recent posting about TCP/IP problems...

It took IBM Product Support a week of working on this problem to 
discover that it is a common one, and that there are several levels
of "corrective services" (bug "fixes", you wascawwy wabbit!) for
TCP/IP.  After wasting my time for a week, they gave me another number
to call. Arrgh!

I DID receive the communications cookbook in the mail -- which is, of
course, loaded with erroneous info -- several command options which it lists
for "ping" don't exist, for example!

If I ever get the net up and running, I will post a recipe.  I still don't
know why the X on DOS software sets the Ethernet Address of the local
host to 00 00 00 00 00 !!! Is that a bug, or is it related to the ARP??

again, any e-mail replies appreciated.

-- 
				{apple, pacbell, hplabs, ucbvax}!well!gors
							gors@well.sf.ca.us
(Doolan) | (Meyer) | (Sierchio) | (Stewart)

brunner@bullhead.uucp (08/09/89)

In article <12794@well.UUCP> gors@well.UUCP (Gordon Stewart) writes:
>As a follow-up to my recent posting about TCP/IP problems...

>I DID receive the communications cookbook in the mail -- which is, of
>course, loaded with erroneous info -- several command options which it lists
>for "ping" don't exist, for example!

The source for ping has been in the public domain ever since Mike Meuss (BRL)
worte it. I don't know which options you refer to, source routing is a common
unimplimented flag. Could you specify which options are unimplimented and how
you would use any option to fix your net? I assume that at least you can send
an icmp echo packet and get in return an icmp echo-reply packet.


Eric Brunner, IBM AWD Palo Alto
inet: brunner@monet.berkeley.edu or brunner%ibmsupt@uunet.uu.net
uucp: uunet!ibmsupt!brunner		(415) 855-4486

gors@well.UUCP (Gordon Stewart) (08/10/89)

Unless your responses are timely and informative, they are of little
value.

The Communications cookbook has several errors which are not entirely
typographic in nature -- these include descriptions of the responses
to various commands, including 'netstat' and 'ping'.

The RT Diagnostic diskettes for AIX 2.2.1 DO contain baseband adapter
tests -- however, it required a service call from a CE (and another 3
days) before I had working copies of THOSE!

The entire support structure at IBM seems intended to frustrate the
poor programmer -- despite the fact that there are a few well-informed
people, whose work is invaluable.  But IBM's corporate policy in re
Software support is awful.  


-- 
				{apple, pacbell, hplabs, ucbvax}!well!gors
							gors@well.sf.ca.us
(Doolan) | (Meyer) | (Sierchio) | (Stewart)