karl@naitc.naitc.com (Karl Denninger) (12/15/90)
Hello Usenet.Oracle... We have a real nice X-windows 3270-style emulator here, but it has one problem..... It is talking through a Mitek gateway in tn3270 mode. We have 95+% functionality, but there is one glaring hole -- there is no code for an ATTN key! Needless to say, this is a problem. We need this supported, as it's the only way to do some things. I can easily hack in the code to send the sequence(s) to the host, IF I know what they are! So far, I have been unable to determine this. Any help appreciated greatly.... If someone knows just what is generated when an ATTN key is pressed, that is probably enough. Thanks! -- Karl Denninger AC Nielsen kdenning@ksun.naitc.com (708) 317-3285 Disclaimer: Contents represent opinions of the author; I do not speak for AC Nielsen on Usenet.
karl@bbs.UUCP (Karl Denninger) (12/15/90)
Hello Usenet.Oracle...
We have a real nice X-windows 3270-style emulator here, but it has one
problem.....
It is talking through a Mitek gateway in tn3270 mode. We have 95+%
functionality, but there is one glaring hole -- there is no code for an ATTN
key!
Needless to say, this is a problem. We need this supported, as it's the
only way to do some things.
I can easily hack in the code to send the sequence(s) to the host, IF I know
what they are! So far, I have been unable to determine this.
Any help appreciated greatly.... If someone knows just what is generated
when an ATTN key is pressed, that is probably enough.
Thanks!
--
Karl Denninger AC Nielsen
kdenning@ksun.naitc.com
(708) 317-3285
Disclaimer: Contents represent opinions of the author; I do not speak for
AC Nielsen on Usenet.jmaynard@thesis1.hsch.utexas.edu (Jay Maynard) (12/18/90)
Karl Denninger asked about doing ATTN on his tn3270 X-windows client. (I want a copy!) Basically, that's a matter of what the gateway expects to see and turn into an ATTN signal - it's not an in-band control sequence, but an out- of band signal sent from the 3274, or, in this case, the gateway, straight to VTAM as SNA control data. There is no AID generated; in fact, the application program doesn't see it as part of the datastream. Your best bet is to ask Mitek (!) and see what they want to generate an ATTN signal on the emulated SNA session. I looked at RFC1041, which is the only one to address tn3270 at all, and it's silent on the issue as well. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@thesis1.hsch.utexas.edu | adequately be explained by stupidity. "...flames are a specific art form of Usenet..." -- Gregory C. Woodbury
thesis1.hsch.utexas.EDU@lib.UUCP (Jay Maynard) (12/18/90)
Karl Denninger asked about doing ATTN on his tn3270 X-windows client. (I want a copy!) Basically, that's a matter of what the gateway expects to see and turn into an ATTN signal - it's not an in-band control sequence, but an out- of band signal sent from the 3274, or, in this case, the gateway, straight to VTAM as SNA control data. There is no AID generated; in fact, the application program doesn't see it as part of the datastream. Your best bet is to ask Mitek (!) and see what they want to generate an ATTN signal on the emulated SNA session. I looked at RFC1041, which is the only one to address tn3270 at all, and it's silent on the issue as well. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@thesis1.hsch.utexas.edu | adequately be explained by stupidity. "...flames are a specific art form of Usenet..." -- Gregory C. Woodbury
drake@drake.almaden.ibm.com (12/18/90)
The ATTN key is, as was already mentioned, an out-of-band communications key. It's the moral equivalent of the "escape" key in many telnet implementations; you hit the escape key (often ctl-t) and the telnet client prompts you to enter various commands. When the command completes you're put back into your telnet session. ATTN is used similarly to allow 3270s to exit their host session and talk to the network. This allows users to forcibly log off from hosts (a la ctl-t Q) and other such things. Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, I'm not sure how you'd implement it on a telnet server. Sam Drake / IBM Almaden Research Center Internet: drake@ibm.com BITNET: DRAKE at ALMADEN Usenet: ...!uunet!ibmarc!drake Phone: (408) 927-1861
drake@drake.almaden.ibm.COM (12/18/90)
The ATTN key is, as was already mentioned, an out-of-band communications key. It's the moral equivalent of the "escape" key in many telnet implementations; you hit the escape key (often ctl-t) and the telnet client prompts you to enter various commands. When the command completes you're put back into your telnet session. ATTN is used similarly to allow 3270s to exit their host session and talk to the network. This allows users to forcibly log off from hosts (a la ctl-t Q) and other such things. Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, I'm not sure how you'd implement it on a telnet server. Sam Drake / IBM Almaden Research Center Internet: drake@ibm.com BITNET: DRAKE at ALMADEN Usenet: ...!uunet!ibmarc!drake Phone: (408) 927-1861
jmaynard@thesis1.hsch.utexas.edu (Jay Maynard) (12/18/90)
In article <374@rufus.UUCP> drake@drake.almaden.ibm.com writes: >Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, >I'm not sure how you'd implement it on a telnet server. Well...sort of. It's also used to interrupt applications in progress, especially TSO, and in some session managers (Legent's TPX springs to mind) to signal a desire to switch from the current session to the manager screen. As such, it is a useful (possibly, in some environments, necessary) thing to implement. I know that there's a canonical implementation, since FTP Software's tn3270 client does it in a manner that our Interlink telnet server understands; I just don't know what it is. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@thesis1.hsch.utexas.edu | adequately be explained by stupidity. "...flames are a specific art form of Usenet..." -- Gregory C. Woodbury
thesis1.hsch.utexas.EDU@lib.UUCP (Jay Maynard) (12/18/90)
In article <374@rufus.UUCP> drake@drake.almaden.ibm.com writes: >Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, >I'm not sure how you'd implement it on a telnet server. Well...sort of. It's also used to interrupt applications in progress, especially TSO, and in some session managers (Legent's TPX springs to mind) to signal a desire to switch from the current session to the manager screen. As such, it is a useful (possibly, in some environments, necessary) thing to implement. I know that there's a canonical implementation, since FTP Software's tn3270 client does it in a manner that our Interlink telnet server understands; I just don't know what it is. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jmaynard@thesis1.hsch.utexas.edu | adequately be explained by stupidity. "...flames are a specific art form of Usenet..." -- Gregory C. Woodbury
P85025@BARILVM.BITNET (Doron Shikmoni) (12/18/90)
In article <1990Dec15.014654.28204@naitc.naitc.com>, Karl Denninger <ddsw1!bbs!karl@GARGOYLE.UCHICAGO.EDU> says: >Any help appreciated greatly.... If someone knows just what is generated >when an ATTN key is pressed, that is probably enough. It's more complicated than that, unfortunately. ATTN is a feature of SNA controllers. 3270 emulation via TCP/IP emulates a non-SNA terminal (regular, local 3270). On such machines, ATTN is not defined at all (if you press ATTN on a 3270 attached to a local non-SNA controller, you get X-F - function not supported). In SNA controllers, ATTN generates a special interrupt RU - an SNA frame. Doron
PMW1@PSUVM.PSU.EDU ("Peter M. Weiss") (12/18/90)
Here is an NLDM trace of an SNA Attn within a TSO/E session using
the IBM WSP 1.12 over an IBM 3174-01L TRN Gw under VTAM 3.2 PUT
9003+
NLDM.PIUT SESSION TRACE DATA PAGE
----------- PRIMARY --------------+---------- SECONDARY --------------+- DOM -
NAME TSOK003 SA 00000007 EL 003A | NAME SHLMSA98 SA 00000007 EL 0269 | NCCF2
----------------------------------+-----------------------------------+-------
SEL# TIME SEQ# DIR TYPE ******** REQ/RESP HEADER ******** RULEN SENS T
09:13:12 0000 S-P SIG ....OC.DR........................ 5
09:13:12 0000 P-S (+)RSP ....OC.DR........................ 1
09:13:12 0000 P-S SIG ....OC.DR........................ 5
09:13:12 0000 S-P (+)RSP ....OC.DR........................ 1
09:13:12 0011 S-P DATA ....OC.ER............CD.......... 0
09:13:12 0035 P-S DATA ....OC.DR........................ 31
09:13:12 0035 S-P (+)RSP ....OC.DR........................ 0
09:13:12 0036 P-S DATA ....OC.DR............CD.......... 2
09:13:12 0036 S-P (+)RSP ....OC.DR........................ 0
The SNA SIG:
NLDM.PIUD SPECIFIC RU DETAIL DATA PAGE 1
----------- PRIMARY --------------+---------- SECONDARY --------------+- DOM -
NAME TSOK003 SA 00000007 EL 003A | NAME SHLMSA98 SA 00000007 EL 0269 | NCCF2
----------------------------------+-----------------------------------+-------
RU DATA: C900010000 *I..........*
/Pete
--
Peter M. Weiss | pmw1 @ PSUADMIN | vm.psu.edu | psuvm
31 Shields Bldg - PennState Univ.| not affiliated with VM.PSU.EDU | PSUVM
University Park, PA USA 16802 | Secrecy is the guardian of bureaucracyjkl@hpindwa.UUCP (Jon Lewis) (12/19/90)
I'll add some info as to what the ATTN key does when pressed on a real 3270.
I can't speak about tn3270, as I know nothing about that product.
Keyboard 3274 VTAM
ATTN key pressed
DFC Signal ---->>>>>
When convenient,
send change direction
(CDI). This turns
the line around, and
allows the 3270 to
enter data.
Note, that it is up to the application/presentation services layer on the
IBM side to handle the DFC (Data Flow Control) command and turn the line
around. For LU2(3270), when a signal is received, the application
should complete any pending I/O, e.g. if it is the middle of a chain, and
turn the line around as soon as possible.
Hope this helps, if not, and you want some more clarification, drop me
some email.
Jon Karl Lewis
Information Networks Division - Hewlett Packard Company
jkl@hpindzl.cup.hp.comjkl@hpindwa.cup.hp.com (Jon Lewis) (12/19/90)
I'll add some info as to what the ATTN key does when pressed on a real 3270.
I can't speak about tn3270, as I know nothing about that product.
Keyboard 3274 VTAM
ATTN key pressed
DFC Signal ---->>>>>
When convenient,
send change direction
(CDI). This turns
the line around, and
allows the 3270 to
enter data.
Note, that it is up to the application/presentation services layer on the
IBM side to handle the DFC (Data Flow Control) command and turn the line
around. For LU2(3270), when a signal is received, the application
should complete any pending I/O, e.g. if it is the middle of a chain, and
turn the line around as soon as possible.
Hope this helps, if not, and you want some more clarification, drop me
some email.
Jon Karl Lewis
Information Networks Division - Hewlett Packard Company
jkl@hpindzl.cup.hp.com
drake@drake.almaden.ibm.COM (12/20/90)
In article <9012190925.AA26026@lilac.berkeley.edu> Jon Lewis <hpda!hpcupt1!hpindwa!jkl@ucbvax.Berkeley.EDU> writes: >I'll add some info as to what the ATTN key does when pressed on a real 3270. [details deleted] Nice summary ... unfortunately it's only sometimes accurate. Specifically, that's what happens if you press ATTN on a real 3270 attached to a SNA 3x74. On a local (channel-attached) non-SNA controller, ATTN is a dead key. Sam Drake / IBM Almaden Research Center Internet: drake@ibm.com BITNET: DRAKE at ALMADEN Usenet: ...!uunet!ibmarc!drake Phone: (408) 927-1861
drake@drake.almaden.ibm.com (12/20/90)
In article <9012190925.AA26026@lilac.berkeley.edu> Jon Lewis <hpda!hpcupt1!hpindwa!jkl@ucbvax.Berkeley.EDU> writes: >I'll add some info as to what the ATTN key does when pressed on a real 3270. [details deleted] Nice summary ... unfortunately it's only sometimes accurate. Specifically, that's what happens if you press ATTN on a real 3270 attached to a SNA 3x74. On a local (channel-attached) non-SNA controller, ATTN is a dead key. Sam Drake / IBM Almaden Research Center Internet: drake@ibm.com BITNET: DRAKE at ALMADEN Usenet: ...!uunet!ibmarc!drake Phone: (408) 927-1861
mercer@npdiss1.StPaul.NCR.COM (Dan Mercer) (12/22/90)
In article <9012162110.AA17045@lilac.berkeley.edu> Karl Denninger <ddsw1!bbs!karl@GARGOYLE.UCHICAGO.EDU> writes:
:Hello Usenet.Oracle...
:
:We have a real nice X-windows 3270-style emulator here, but it has one
:problem.....
:
:It is talking through a Mitek gateway in tn3270 mode. We have 95+%
:functionality, but there is one glaring hole -- there is no code for an ATTN
:key!
:
:Needless to say, this is a problem. We need this supported, as it's the
:only way to do some things.
:
:I can easily hack in the code to send the sequence(s) to the host, IF I know
:what they are! So far, I have been unable to determine this.
:
:Any help appreciated greatly.... If someone knows just what is generated
:when an ATTN key is pressed, that is probably enough.
:
:Thanks!
:
:--
:Karl Denninger AC Nielsen
:kdenning@ksun.naitc.com
:(708) 317-3285
:Disclaimer: Contents represent opinions of the author; I do not speak for
: AC Nielsen on Usenet.
The ATTN key will send SIGNAL to the host to request the Change
Direction indicator (CD). The command is expedited but has no effect
on SLU states. Once Signal has been sent, pressing the ATTN or PA
keys will not cause a second ATTN until the 3274 has received a
response to the first. Normally, the PLU responds by sending a NULL
RU with the CD flag set. If I had my SNA FAP (Formats and Protocols)
on me I'd give you the actual Signal format, but I took it home.
--
Dan Mercer
NCR Network Products Division - Network Integration Services
Reply-To: mercer@npdiss1.StPaul.NCR.COM (Dan Mercer)
"MAN - the only one word oxymoron in the English Language"mercer@npdiss1.UUCP (Dan Mercer) (12/22/90)
In article <9012162110.AA17045@lilac.berkeley.edu> Karl Denninger <ddsw1!bbs!karl@GARGOYLE.UCHICAGO.EDU> writes: :Hello Usenet.Oracle... : :We have a real nice X-windows 3270-style emulator here, but it has one :problem..... : :It is talking through a Mitek gateway in tn3270 mode. We have 95+% :functionality, but there is one glaring hole -- there is no code for an ATTN :key! : :Needless to say, this is a problem. We need this supported, as it's the :only way to do some things. : :I can easily hack in the code to send the sequence(s) to the host, IF I know :what they are! So far, I have been unable to determine this. : :Any help appreciated greatly.... If someone knows just what is generated :when an ATTN key is pressed, that is probably enough. : :Thanks! : :-- :Karl Denninger AC Nielsen :kdenning@ksun.naitc.com :(708) 317-3285 :Disclaimer: Contents represent opinions of the author; I do not speak for : AC Nielsen on Usenet. The ATTN key will send SIGNAL to the host to request the Change Direction indicator (CD). The command is expedited but has no effect on SLU states. Once Signal has been sent, pressing the ATTN or PA keys will not cause a second ATTN until the 3274 has received a response to the first. Normally, the PLU responds by sending a NULL RU with the CD flag set. If I had my SNA FAP (Formats and Protocols) on me I'd give you the actual Signal format, but I took it home. -- Dan Mercer NCR Network Products Division - Network Integration Services Reply-To: mercer@npdiss1.StPaul.NCR.COM (Dan Mercer) "MAN - the only one word oxymoron in the English Language"
karl@naitc.naitc.com (Karl Denninger) (12/28/90)
In article <374@rufus.UUCP> drake@drake.almaden.ibm.com writes: >The ATTN key is, as was already mentioned, an out-of-band communications key. >It's the moral equivalent of the "escape" key in many telnet implementations; >you hit the escape key (often ctl-t) and the telnet client prompts you to >enter various commands. When the command completes you're put back into your >telnet session. > >ATTN is used similarly to allow 3270s to exit their host session and talk to >the network. This allows users to forcibly log off from hosts (a la ctl-t Q) >and other such things. > >Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, >I'm not sure how you'd implement it on a telnet server. The Mitek uses a IAC BREAK (ala out-of-band Telnet command) to do this. Let's hope IBM's TCP/IP for MVS supports this.... otherwise when we get real TCP/IP loaded on the 3090s around here, we'll not be happy anymore! Right now what we have works... -- Karl Denninger AC Nielsen kdenning@ksun.naitc.com (708) 317-3285 Disclaimer: Contents represent opinions of the author; I do not speak for AC Nielsen on Usenet.
karl@bbs.UUCP (Karl Denninger) (12/28/90)
In article <374@rufus.UUCP> drake@drake.almaden.ibm.com writes: >The ATTN key is, as was already mentioned, an out-of-band communications key. >It's the moral equivalent of the "escape" key in many telnet implementations; >you hit the escape key (often ctl-t) and the telnet client prompts you to >enter various commands. When the command completes you're put back into your >telnet session. > >ATTN is used similarly to allow 3270s to exit their host session and talk to >the network. This allows users to forcibly log off from hosts (a la ctl-t Q) >and other such things. > >Since ATTN's used to talk to the network (VTAM, the 3x74 et al), not the host, >I'm not sure how you'd implement it on a telnet server. The Mitek uses a IAC BREAK (ala out-of-band Telnet command) to do this. Let's hope IBM's TCP/IP for MVS supports this.... otherwise when we get real TCP/IP loaded on the 3090s around here, we'll not be happy anymore! Right now what we have works... -- Karl Denninger AC Nielsen kdenning@ksun.naitc.com (708) 317-3285 Disclaimer: Contents represent opinions of the author; I do not speak for AC Nielsen on Usenet.