I've done a lot of debugging with SNA over the years and majority of bind failures were due to RU size mismatch (for 3270 most of the rest due to device attributes). Hardware was memory constrained and the hardware engineers squeezed the maximum out of every byte ... leading to software having to support literally a dozen different sizes for similar function. I have never used 3192g (thankfully) but went through worse with 3290 (the original tiger terminals). GDDM does let you exploit quite a few hardware capabilities that most people never knew existed on these 35 year old terminals that had better graphics than contemporary PCs.
On Tue, Dec 14, 2021 at 10:38 AM Alexander Huemer <[email protected]> wrote: > Thank you _very_ much. That did the trick. > After your hint I did some web search and then found [1] which for > whatever reason didn't show up in my previous research. > I then coded the MODEENT as > > 004210 > ********************************************************************** > 004211 ** Logmode entry for 3192G > 004212 > ********************************************************************** > 004213 FOO3192G MODEENT LOGMODE=FOO3192G, > * > 004214 TYPE=1, > * > 004215 FMPROF=X'03', > * > 004216 TSPROF=X'03', > * > 004217 PRIPROT=X'B1', > * > 004218 SECPROT=X'90', > * > 004219 COMPROT=X'3080', > * > 004220 RUSIZES=X'8687', > * > 004221 PSNDPAC=X'01', > * > 004222 SRCVPAC=X'01', > * > 004223 PSERVIC=X'028000000000000000000300' > > May I ask how you figured that the 3192G has particular requirements for > RUSIZE? > > [1] > https://www.ibm.com/docs/en/gddm?topic=network-instructions-checking-vtam > > -Alex > > On Tue, Dec 14, 2021 at 09:03:40AM +1100, Attila Fogarasi wrote: > > Your RUSIZE is too big for some 3192g configurations, also suggest adding > > pacing ... try the following > > RUSIZES=X'8687', PSNDPAC=X'01' ,SRCVPAC=X'01' C > > > > On Tue, Dec 14, 2021 at 2:37 AM Seymour J Metz <[email protected]> wrote: > > > > > I would not expect GDDM to work with logmode DYNAMICB. Have you looked > > > into Attila Fogarasi's suggestion about R/R unit sizes? > > > > > > > > > -- > > > Shmuel (Seymour J.) Metz > > > http://mason.gmu.edu/~smetz3 > > > > > > ________________________________________ > > > From: IBM Mainframe Discussion List [[email protected]] on > behalf > > > of Alexander Huemer [[email protected]] > > > Sent: Sunday, December 12, 2021 11:34 PM > > > To: [email protected] > > > Subject: Re: VTAM LU coding for 3192G display > > > > > > The DLOGMOD setting I had previously was D4C32XX3, which is defined as: > > > > > > D4C32XX3 MODEENT > LOGMODE=D4C32XX3,FMPROF=X'03',TSPROF=X'03',PRIPROT=X'B* > > > > 1',SECPROT=X'90',COMPROT=X'3080',RUSIZES=X'87F8',PSERVIC* > > > =X'028000000000000000000300',APPNCOS=#CONNECT *@KGC* > > > > > > EDS capability is set, according to the description of the bytes I > found > > > at [1]. > > > Terminal behavior is identical with D4C32XX3 and DYNAMICB. The latter > > > doesn't have X'80' set. > > > > > > MODEENT LOGMODE=DYNAMICB, > X > > > APPNCOS=#CONNECT, D409262 > X > > > COS=INTERACT, > X > > > FMPROF=X'02', > X > > > TSPROF=X'02', > X > > > PRIPROT=X'71', > X > > > SECPROT=X'40', > X > > > COMPROT=X'2000', > X > > > RUSIZES=X'0000', > X > > > PSERVIC=X'00C000000000000000000300' > > > > > > I am out of ideas. > > > > > > -Alex > > > > > > [1] > > > > https://www.ibm.com/docs/en/zos/2.1.0?topic=devices-pservic-operand-modeent-macroinstruction > > > > > > On Mon, Dec 13, 2021 at 02:49:26AM +0000, Seymour J Metz wrote: > > > > Take a look at D4A32XX3, D4B32XX3 and D4332XX3. What is important is > > > that they all have BINDFLAG set tp BINSEDS (X'80'), indicating support > for > > > extended data stream. > > > > > > > > > > > > -- > > > > Shmuel (Seymour J.) Metz > > > > http://mason.gmu.edu/~smetz3 > > > > > > > > ________________________________________ > > > > From: IBM Mainframe Discussion List [[email protected]] on > > > behalf of Alexander Huemer [[email protected]] > > > > Sent: Sunday, December 12, 2021 10:14 AM > > > > To: [email protected] > > > > Subject: Re: VTAM LU coding for 3192G display > > > > > > > > In VTAM there are several LUs defined, all equal. Only displays, no > > > > printers. On the 3174 there is no special provisioning either. > > > > I know of the existence of PSERVIC in the LOGMODE macros, but not > how to > > > > craft PSERVIC values. > > > > GDDM is working e.g. on the 3290. > > > > > > > > -Alex > > > > > > > > On Sun, Dec 12, 2021 at 09:57:29PM +1100, Attila Fogarasi wrote: > > > > > I just remembered that unrelated devices on the same controller can > > > affect > > > > > what are valid values for the bind for 3192g. For example if > there is > > > a > > > > > 3287 (printer) defined on the same controller, then the max RU > size is > > > > > reduced even for the 3192g (so what would have been a valid bind > with > > > no > > > > > 3287 configured becomes invalid). This is the reason the otherwise > > > very > > > > > capable 3270 architecture became abandoned :) All of this varies > with > > > > > specific device types, of course, once you get out of mainstream > > > devices. > > > > > Good luck getting GDDM working. > > > > > > > > > > On Sun, Dec 12, 2021 at 9:42 PM Attila Fogarasi < > [email protected]> > > > wrote: > > > > > > > > > > > Logmode is invalid for the physical device, sense 0821 means that > > > the bind > > > > > > was rejected due to having invalid parameters (which come from > the > > > logmode > > > > > > entry). You might want to check the PSERVIC value. > > > > > > > > > ---------------------------------------------------------------------- > > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > > send email to [email protected] with the message: INFO > IBM-MAIN > > > > > > > > > ---------------------------------------------------------------------- > > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > > send email to [email protected] with the message: INFO > IBM-MAIN > > > > > > ---------------------------------------------------------------------- > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > send email to [email protected] with the message: INFO IBM-MAIN > > > > > > ---------------------------------------------------------------------- > > > For IBM-MAIN subscribe / signoff / archive access instructions, > > > send email to [email protected] with the message: INFO IBM-MAIN > > > > > > > ---------------------------------------------------------------------- > > For IBM-MAIN subscribe / signoff / archive access instructions, > > send email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
