I wanted to try to start into the Visualage for JAVA deal once more. So I opened up the HELP session on getting started. There is a version of KEYHOST which can be used to generate digital certificates under your own CA game in OS/2, I've discovered and I wanna play with it. So that wound me back up again, sigh, with the NETQ server error an unable to reach port 49213 again.

Long ago and very far away, I got all this thing working, FINALLY, with DHCP and setting all the hostname and host settings in the CONFIG.SYS file, plus modifying the DHCPSTRT operations to make it all work.

Since that time, I've gone to a fixed address. I'm guessing that the failure to start the NETQ server is now related to the difference between DHCP and what

all that dog drags in the door in sequence that satisfies the master of the house for NETQ.

But where do I start now to unravel all this now that the fixed addressing scheme is upon us?

A cursory look back at all this notes that my "servername" is different than "localhost". It also notes in the NETQOS2 directory, that there was a previous

jockeying of switching the "localhost" to "servername" that was involved with all this mess before.

What would be the effect, now that I'm looking at this all over again from a different view, of naming my machine "localhost" for a name? Does anyone know if this would magically solve the whole thing? I never considered that before.

Also!

This box is not on 32 bit TCPIP now. I have the paid-in copy of 32 bit TCP/IP that I am considering installing on this FP 16 box. Would I be better off to try to perhaps install the 32 Bit TCP/IP package on in BEFORE I start fooling around with NETQ and all?

Mikey Dog is sitting on the edge of this huge canyon and looking at
the other side ... sort of like Evil Knevile (sic?), but with a
velocipede built for one and pedal power...

That haunting arpegio is playing in one ear as well! ;)

Advice?

Sleep well; OS/2's still awake! ;)

Mike @ 1:117/3001

--- Maximus/2 3.01
# Origin: Ziplog Public Port (1:117/3001)
* Origin: Baddog BBS (1:218/903)

Hi Mike,

Replying to a message of Mike Luther to All:

[snip]

Advice?

I went through the same problems (and perhaps a few more) with the same help server for Visual Age C++ 4.0.

If you have access to one of the eComStation newservers (e.g. news.ecomstation.nl) then you will find the complete and rather sorry history of how to fix this sucker in messages from around early June [so they should still be on the server]. If you cannot find them then tell me and I will post a

precis of them here.

Do not expect this server to work with newer kernels or TCP/IP stacks. It is a bucket of bolts and is being left to rust by IBM. I got it to run to a point where I could prove there was a bug in the code. That bug will never be fixed. The final result is that I must either search the HTML-esque documentation by hand using a Web browser or code without documentation.

<SARCASM>
From the quality of IBM's code here, I concluded that IBM's programmers chose the latter course of action.
</SARCASM>

Regards

Dave
<Team PL/I>

--- FleetStreet 1.25.1
# Origin: My other computer is an IBM S/390 (2:257/609.5)
* Origin: Baddog BBS (1:218/903)