Ed Sometimes inquiring minds want to know what is going on.
The rest of this post talks about TSO as the primary LU application but I guess similar principles will apply to your "VTAM application switcher". I wouldn't be at all surprised if your application responded to the equivalent of the "Attn" key. - What the "TSO/E User's Guide" has to say on the matter is the following: <quote> 1.1.3.2 Interrupting a Process You can interrupt processing at any time by pressing the key assigned as the attention interrupt key on your terminal. On some terminals such as the IBM 3270 display terminal, the attention interrupt key is labeled "PA1". You can cause an attention interrupt to: - Terminate processing - Interrupt processing and resume it again by pressing the Enter key after pressing the attention interrupt key - Issue the TIME or TEST command during interrupted processing </quote> In essence these TSO folk are lost in the mists of prehistory here still pining for the 2741 and like devices. I expect the key designed to cause an interruption was the "Break" key or something similar. It appears they are not too familiar with this new-fangled 3270 thing[1] and imagine there may be other TSO-driving devices with PA1 keys somewhere on the keyboard. [1] It's important to appreciate that what they have in mind in a 3270 display coax-attached to a BSC 3271 or channel-attached 3272 controller (or the BSC 3275), in other words, before the age of SNA enlightenment. If we are dealing with the type of 3270 which relies on VTAM to convert the pre-SNA data traffic into SNA data traffic (LU type 0) then the flow is full-duplex and the end user can always press a - misleadingly named - "attention-generating" key[2]. These keys include ENTER, CLEAR, PAx and PFx. TSO has decided to use PA1 as the key to "interrupt processing" as described above. [2] If the keyboard appears locked, it suffices to press the RESET key in order to unlock the keyboard before optionally entering data and pressing an attention-generating key. However, SNA came along - and muddied TSO's clear water. If the end user now employs a 3270 - or, more importantly these days, a 3270 emulator (some say simulator) - session using LU type 2, the data flow is now - in my opinion - the better regulated half-duplex and it will now be very likely that, in order to "interrupt" the current "process", the PA1 key will no longer cause TSO to register the "attention" condition. The reason is that an ongoing "process" in the TSO address space is likely to have been initiated by a command and the "change direction" (CD) bit - or "baton" as I used to describe it to students using the analogy of a relay race - will have been transferred to the primary LU, TSO, from the secondary LU, the 3270 display, along with the data comprising the command text which initiated the "process". No amount of pounding on the RESET key is allowed to violate the half-duplex protocol which dictates that the "next speaker" is the LU which has the CD bit in its grasp, namely TSO. The only way that the CD bit can be wrested from the fist of the owning LU is to startle the LU with a special request, the SIGNAL request. The SIGNAL request is allowed to flow from an LU even if it doesn't possess the CD bit and is actually a polite request to the receiving LU to release the CD bit by sending a message to the sending LU with the CD bit in the header. The receiving LU application could decide not to accede to the request but TSO will. In addition TSO regards receiving a SIGNAL as an "attention" condition. The way that the end user of a 3270 display device causes the SIGNAL request to be sent is to press the "Attn" key - finally we get to the alternative to the PA1 key. Of course, if you are running a 3270 emulator, you will need to know which of your PC keyboard keys corresponds to the 3270 keyboard "Attn" key. Only the emulator documentation can cover this - or the "Help" selection. I seem to recall a right-click panel with the emulator I used to use, PCOMM I expect. This panel enabled the logical use of 3270 keys that didn't have a obvious counterpart on the PC keyboard and one of these was the "Attn" key. It hadn't occurred to me that use of the "Attn" key when not "stuck" in a "process" would be equally effective as a means to create the "attention" condition in TSO - as long as it was an LU type 2 session. However the SIGNAL request can flow even when the CD bit is in the possession of the LU sending the SIGNAL request, in this case, the secondary LU. Since receiving the SIGNAL request creates the "attention" condition in TSO, use of the "Attn" key can be a substitute for the PA1 key even when the keyboard is not locked. Warning: I've composed this largely from memory but I'm sure I'll be corrected where I have strayed too far from full accuracy. Chris Mason ----- Original Message ----- From: "Ed Gould" <[EMAIL PROTECTED]> Newsgroups: bit.listserv.ibm-main To: <[email protected]> Sent: Saturday, 23 September, 2006 5:39 AM Subject: Re: SDSF Question > 10+ years ago we had a vtam appl switcher (sorry forgot the name but > it could have been TPX). > > It worked great on real 3270's but they made us switch over pc's. The > 3270 emulator would not let us hit reset and pa1 . It essentially > made the vtam appl switcher unusable. grumble grumble I complained > but no one would listen. I suggested dropping the product as it was > useless now. The VP of the PC department got p-oed. We had a meeting > and when I showed up with VTAM traces to prove his emulator wasn't > working correctly he got even madder. > > Ed > > On Sep 22, 2006, at 10:18 AM, Greg Shirey wrote: > > > -----Original Message----- > > From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] > > On Behalf > > Of Shmuel Metz (Seymour J.) > > Sent: Friday, September 22, 2006 7:31 AM > > > >> In <[EMAIL PROTECTED]>, on 09/21/2006 > > at 04:25 PM, Greg Shirey <[EMAIL PROTECTED]> said: > > > >>> PA1. > > > >> That won't work if the keyboard is locked. Attention should always > >> work. > > > > Shmuel, > > > > Thanks for that helpful information. I never knew that. > > And it works with MacKinney's VTAM/Switch (which uses the PA1 key > > to switch > > between sessions) - that is, pressing Attention returns control to > > Switch > > where you can type "SIM PA1" (simulate pressing PA1) to end the SDSF > > auto-update. > > > > Greg Shirey > > Ben E. Keith Company ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
