Re: 0.15.x -> 0.16.x -- LINUX5250

Hi there...

"Jason M. Felice" <jasonf@Baldwingroup.COM> wrote:
>
> This means that help in QA testing and submitting bug reports is
>  greatly appreciated at this point, and (unlike some development
>  releases) I don't have any excuses about how the emulator *should*
>  *should* handle that :)   It also means you get many karma points
>  for downloading 0.15.6 and using it by default when connecting to a
>  host :)
>
> This also means you should resubmit any outstanding bugs which have
>  not been resolved.  For the most part, I consider all prior issues
>  closed due to the extent of work in 0.15.x, but I certainly expect
>  there to be a few more issues.


I've been a bit pressed for time, lately, so I haven't been testing
this stuff as extensively as I used to (sorry).

I'm still getting the "assertion failed" error in UPDDTA, that I
first reported in the 0.13.x series.  However, I found some new
information about it that may help you.

I can't get this error to occur when I'm on the same LAN as the
AS/400!  It only happens when I'm at home, and I've dialed into
the company's network with a modem.  I've got a FreeBSD machine on
the LAN at work that acts as a PPP server, so what I do is call in
to the LAN at work with PPP, and then use tn5250 over that 33.6 PPP
connection to access the AS/400. (I hope I'm explaining this clearly)

Its only over this (slower) connection that I have the problem, so I'd
guess that this is a "race condition" of sorts.

In both the LAN scenario, and the PPP scenario, the cursor doesn't
position itself "correctly", as well.

Here's a recap of the problem itself:

I go into UPDDTA mylib/myfile.   When the "WORK WITH DATA IN A FILE"
screen comes up, tn5250 positions the cursor in the "Format:" field,
(whereas RUMBA puts it in the *RECNBR field, or the first key field
if the file is keyed -- this problem happens on the LAN, too)
If I hit F10, followed by ENTER (as quickly as I can type them) I'll
get: "Assertion 0 failed at session.c, line 354." (only when via PPP)
(And yes, this happens on both RedHat 6.1 and FreeBSD)

If you pause for a little bit between pressing F10 and pressing
ENTER, it does not crash.  (Tho the cursor still repositions in a
different place than RUMBA puts it)

I sent a tracefile in the 0.13.x series, and I may have sent another
one when 0.14.x and 0.15.x came out, I don't recall.  If you'd like
another tracefile, I'd be happy to send one.

It appears that the other bugs that I've submitted are fixed, and
the 0.15.6 release (at least at first glance) does look like the
most stable one to date!  kudos!

+---
| This is the LINUX5250 Mailing List!
| To submit a new message, send your mail to LINUX5250@midrange.com.
| To subscribe to this list send email to LINUX5250-SUB@midrange.com.
| To unsubscribe from this list send email to LINUX5250-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---