开云体育


Re: Updating DMKRIO on VM370 CE #VMCE

 

开云体育

You can generate define these devices to VM and they can be dedicated (or attached) to guests.

Early releases of VM/370 (with SEPP which we don’t have) and VM/SP did 3270 networking via the PASSTHRU program (which again we don’t have)

Later support was added so that terminals connected to VTAM running in DOS could act as VM terminals. I think this might have been in VM/SP3.

?

Dave

?

?

From: [email protected] <[email protected]> On Behalf Of alejandro.olivan.alvarez@...
Sent: Tuesday, April 25, 2023 1:24 PM
To: [email protected]
Subject: Re: [h390-vm] Updating DMKRIO on VM370 CE #VMCE

?

aha... so, does this means that VM/370 is unable to DEDICATE (passthrough) to a guess, any device that it is unable to handle on its own?
Or does that mean that, since no VTAM in VM370, entries for related devices, since simply unnecessary for VM370? itself, where simply never considered to be added?

In any of the cases, a question regarding the second kind of device, 2703s, comes to my mind ... they're around in DMKRIO... so I especulate with the possibility to pass some to tk4- at the right CUUs


Re: Updating DMKRIO on VM370 CE #VMCE

 

aha... so, does this means that VM/370 is unable to DEDICATE (passthrough) to a guess, any device that it is unable to handle on its own?
Or does that mean that, since no VTAM in VM370, entries for related devices, since simply unnecessary for VM370? itself, where simply never considered to be added?

In any of the cases, a question regarding the second kind of device, 2703s, comes to my mind ... they're around in DMKRIO... so I especulate with the possibility to pass some to tk4- at the right CUUs


Re: Updating DMKRIO on VM370 CE #VMCE

 

开云体育

Ok so just in case despite there being no VTAM, TCAM, OR PASSTHRU in vm/370 ce you do want to mess lets look at this.

?

So DMKRIO is assembler source, so it needs to be assembled and then re-linked into the nucleus which is fiddly and then re-ipl to pick up new nucleus

So yes “VMSETUP CP” leaves 094 accessed as a read mode disk.

?

ACCESS 094 F

?

Makes it read/write

?

COPY DMKRIO ASSEMBLE F = ASMORG =

?

Backs up DMKRIO SO YOU CAN EDIT SAFELY.

VMFASM DMKRIO DMKHRC

?

Assembles it using the correct MACLIBs but leaves the output DMKRIO TEXT deck on the “A” disk.

?

COPY DMKRIO TEXT F = TXTORG =

?

Backs this up.

?

COPY DMKRIO TEXT A = = F (REPL

?

Puts the new one on the 094

?

ERASE DMKRIO TEXT A

?

Cleans up A disk…

?

Then run VMSETUP CP to put the disk access modes back as expected.

Then re-build the nuclease as from step 4 in MAINT MEMO…

?

Dave

?

From: [email protected] <[email protected]> On Behalf Of alejandro.olivan.alvarez@...
Sent: Tuesday, April 25, 2023 12:06 PM
To: [email protected]
Subject: Re: [h390-vm] Updating DMKRIO on VM370 CE #VMCE

?

Hi!

mmmm... this route looks familiar to me since, in fact, I tried my first backup and then EE edition of DMKRIO (I simply did ACCESS 094 C) ... just to find that the thing does not have any efect per-se, but rather the reassemble procedure was needed.

But still, I guess that doing VMFASM DMKRIO DMKHRC is not 'that easy'... I mean

In order to do VMFASM, I need to do perform an EXEC VMSETUP CP isn'it?
Or does DMKHRC look for a DMKRIO ASSEMBLE file at maint's A?
In fact, I recall that upon doing the VMSETUP CP command, disk 094 gets 'mounted' (Linux jargon here, sorry, don't know how to call that) on E/F Read/Only... are changes expected therefore to be done on copies at A?

Thank you for your help!
Cheers


Re: Updating DMKRIO on VM370 CE #VMCE

 

On Tue, Apr 25, 2023 at 12:32 PM, <alejandro.olivan.alvarez@...> wrote:


Hi!

I'm using VM/370 Community Edition. I think that, more precisely is CE v1
R1.2.

I took a look at DMKRIO, and I didn't find (or failed to) any entry for 3705
devices/units... so any reserve on trying to avoid messing with that, just
vanished to me.
Also, by reading the Sysgen Book, I realized that (due to my lack of
knowledge) trial and error will be unavoidable, since different ADAPTER
settings are available... and several are the scenarios that guest tk4- is
configured to emulate through 3705 and 2703 hercules devices... it would be
awesome if only one of those 3270 SNA TCAC/VTAM methods or JES2/RJE via 2703
'passthrough's do fineally work.

Cheers.
No VTAM in VM/370 R6 so no 37xx except in EP or attached to MVS guests...

Dave


Re: Updating DMKRIO on VM370 CE #VMCE

 

Hi!

mmmm... this route looks familiar to me since, in fact, I tried my first backup and then EE edition of DMKRIO (I simply did ACCESS 094 C) ... just to find that the thing does not have any efect per-se, but rather the reassemble procedure was needed.

But still, I guess that doing VMFASM DMKRIO DMKHRC is not 'that easy'... I mean

In order to do VMFASM, I need to do perform an EXEC VMSETUP CP isn'it?
Or does DMKHRC look for a DMKRIO ASSEMBLE file at maint's A?
In fact, I recall that upon doing the VMSETUP CP command, disk 094 gets 'mounted' (Linux jargon here, sorry, don't know how to call that) on E/F Read/Only... are changes expected therefore to be done on copies at A?

Thank you for your help!
Cheers


Re: Updating DMKRIO on VM370 CE #VMCE

 

Hi!

I'm using VM/370 Community Edition. I think that, more precisely is CE v1 R1.2.

I took a look at DMKRIO, and I didn't find (or failed to) any entry for 3705 devices/units... so any reserve on trying to avoid messing with that, just vanished to me.
Also, by reading the Sysgen Book, I realized that (due to my lack of knowledge) trial and error will be unavoidable, since different ADAPTER settings are available... and several are the scenarios that guest tk4- is configured to emulate through 3705 and 2703 hercules devices... it would be awesome if only one of those 3270 SNA TCAC/VTAM methods or JES2/RJE via 2703 'passthrough's do fineally work.

Cheers.

??


Re: Updating DMKRIO on VM370 CE #VMCE

 

Alejandro,

In order to respond to you with any accuracy, we need to know what
"build" of VM/370 you have installed.

3pack, 4 pack, 5 pack, 6 pack, 6 pack beta, and CE are all different in
this area, and if you are using CE, you might find that DMKRIO already
has entries to suit your needs.

Chris
--
<cjar1950@...>



----------------------------------------------------------------------------------------------------------------------------------
On Tue, 25 Apr 2023 02:52:05 -0700
alejandro.olivan.alvarez@... wrote:
Hi Folks... first message here... noob level (Linux sys admin trying the mainframe experience out of curiosity)

After reading around at the VM/370 Sysgen Book, there're a couple of experiments I would like to test (as an exercice) under VM/370.
Basically, I would like to lab on its 'passthrough' (sorry...I'm a QEMU/KVM guy and I'm still not much familiarized with mainframe terms) capabilities. More precissely, I would like to repeat my experiments on tk4- different access methods (3705, 2703, 3335...) but, under VM/370.

At VM/370 Sysgen Book I found enough references and documentation on 3705 and 2703 support to think that, under my (very limited) understanding, there're technically chances to get some access methods to work.
The problem is that, beyond, hercules config and USER DIRECT, file (which I dare to edit and apply much like Moshix does on his videos) I need to update DMKRIO file.

While MAINT MEMO file is written quiete friendly, I still have too many gaps to get an action path.
Trying to read the often referred 'What Mother never tell you...' PDF file, however, was far from helpful, but rather pretty scaring.
Overall, i'm quite confused.

I'm asuming that I cannot ee DMKRIO file, but, rather, create some kind of local (LCL) changes file right? Could you please help me on a starting point to follow the bread crumbs?

Thanks.





Re: Updating DMKRIO on VM370 CE #VMCE

 

开云体育

Personally I would make a backup copy of DMKRIO and then edit it directly. (Or just copy it to your “A Disk)

You will need to use

VMFASM DMKRIO DMKHRC

To assemble it. You get four warnings which you can ignore.

Then re-build the CP Nucleus as per the memo

Dave

?

?

From: [email protected] <[email protected]> On Behalf Of alejandro.olivan.alvarez@...
Sent: Tuesday, April 25, 2023 10:52 AM
To: [email protected]
Subject: [h390-vm] Updating DMKRIO on VM370 CE #VMCE

?

Hi Folks... first message here... noob level (Linux sys admin trying the mainframe experience out of curiosity)

After reading around at the VM/370 Sysgen Book, there're a couple of experiments I would like to test (as an exercice) under VM/370.
Basically, I would like to lab on its 'passthrough' (sorry...I'm a QEMU/KVM guy and I'm still not much familiarized with mainframe terms) capabilities. More precissely, I would like to repeat my experiments on tk4- different access methods (3705, 2703, 3335...) but, under VM/370.

At VM/370 Sysgen Book I found enough references and documentation on 3705 and 2703 support to think that, under my (very limited) understanding, there're technically chances to get some access methods to work.
The problem is that, beyond, hercules config and USER DIRECT, file (which I dare to edit and apply much like Moshix does on his videos) I need to update DMKRIO file.

While MAINT MEMO file is written quiete friendly, I still have too many gaps to get an action path.
Trying to read the often referred 'What Mother never tell you...' PDF file, however, was far from helpful, but rather pretty scaring.
Overall, i'm quite confused.

I'm asuming that I cannot ee DMKRIO file, but, rather, create some kind of local (LCL) changes file right? Could you please help me on a starting point to follow the bread crumbs?

Thanks.
?


Updating DMKRIO on VM370 CE #VMCE

 

Hi Folks... first message here... noob level (Linux sys admin trying the mainframe experience out of curiosity)

After reading around at the VM/370 Sysgen Book, there're a couple of experiments I would like to test (as an exercice) under VM/370.
Basically, I would like to lab on its 'passthrough' (sorry...I'm a QEMU/KVM guy and I'm still not much familiarized with mainframe terms) capabilities. More precissely, I would like to repeat my experiments on tk4- different access methods (3705, 2703, 3335...) but, under VM/370.

At VM/370 Sysgen Book I found enough references and documentation on 3705 and 2703 support to think that, under my (very limited) understanding, there're technically chances to get some access methods to work.
The problem is that, beyond, hercules config and USER DIRECT, file (which I dare to edit and apply much like Moshix does on his videos) I need to update DMKRIO file.

While MAINT MEMO file is written quiete friendly, I still have too many gaps to get an action path.
Trying to read the often referred 'What Mother never tell you...' PDF file, however, was far from helpful, but rather pretty scaring.
Overall, i'm quite confused.

I'm asuming that I cannot ee DMKRIO file, but, rather, create some kind of local (LCL) changes file right? Could you please help me on a starting point to follow the bread crumbs?

Thanks.
?


Re: Wakeup userid - what is it?

 

The WAKEUP server REXX code calls several?functions defined in the RXUSERFN REXX function package.
The main functionused?by the WAKEUP server is $WAIT. The source?code for the package is in the
DMSLFN?ASSEMBLE file (plus various associated?update files. That code is in the VMSETUP?CMS search order.

I could not find any existing help file that matches the version of DMSLFN?that I have, so I don't have
anything immediately available. I will open an issue to make one for the next VMCE release, which I am
trying to release fairly soon.

Bob Bolch?

On Thu, Apr 20, 2023 at 7:39?AM Bertram Moshier <herc370390vm@...> wrote:
Hi!

Where is the source code for WAKEUP?? I apologize, in advance, should you already answer this question.?

Bert

On Thu, Apr 20, 2023, 00:39 Dennis Boone <drb@...> wrote:
The current-ish IBM documentation for WAKEUP includes some sample exec
snippets that show how you approach using the beast:



De






Re: Wakeup userid - what is it?

 

Am 21.04.2023 um 00:41 schrieb Fish Fish:
I would like to also point out that the password for any user can be easily discovered by simply examining the MAINT userid's "USER DIRECT" file.
If you know the password for the MAINT userid, you know the password for ALL userids.
In my VM 6.0 installation on Hercules, which is working since 2011, I sometimes forgot the MAINT password :-(

What then helped me:

from another user (which I was working with regularly):

LINK MAINT 191 AS 196 RR

the READ passwort for the minidisk has (luckily) been left on the default value, which is: READ

ACC 196 D

and then I could easily examine the DIRECT file, which in my case is called VM60 DIRECT (not USER DIRECT).

HTH, kind regards

Bernd


Re: Wakeup userid - what is it?

 

(piggyback)

Bertram Moshier wrote:
Larry S wrote:


Yes, WAKEUP.
I would like to also point out that the password for any user can be easily discovered by simply examining the MAINT userid's "USER DIRECT" file. If you know the password for the MAINT userid, you know the password for ALL userids.

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...


Re: Wakeup userid - what is it?

 

Yes, WAKEUP.


On Thu, Apr 20, 2023, 17:04 Larry S <larryschacher@...> wrote:
Is there a password for the WAKEUP user?


Re: Wakeup userid - what is it?

 

Is there a password for the WAKEUP user?


Re: Wakeup userid - what is it?

 

Hello Dave,

Thank you for your additional and extremely information.? It makes accessing and using the VM easier.

Bert

On Thu, Apr 20, 2023, 06:42 Dave Wade <dave.g4ugm@...> wrote:

Bert,

You can shut the machine down from MAINT or OPERATOR by “SMSG WAKEUP SHUTDOWN” => yes I looked at the source to figure this out.

If you then logon to the wakeup VM it starts as a normal VM and you can edit the files.

The whole machine is implemented as REXX EXECs so all the source is on your “A” disk.

The format of the WAKEUP file is documented in the end of the $WAITF EXEC

To restart type “wakeprof” and then “#cp disc”

Dave

?

From: [email protected] <[email protected]> On Behalf Of Bertram Moshier
Sent: Thursday, April 20, 2023 4:21 AM
To: [email protected]
Subject: Re: [h390-vm] Wakeup userid - what is it?

?

Hi Bob,

?

Thank you for your reply.? I'm looking forward to the HELP file.

?

I hope you'll consider including any commands someone can use after logging onto the userid.? After all, there must be someway to exit to CMS so one can modify the WAKEUP TIMES file.? I'm wondering is there a way to look at the entries from the command line (or other stuff) without? exiting to CMS.

?

Is the source code for WAKEUP on user's minidisks?

?

Bert

?

On Wed, Apr 19, 2023, 10:36 Bob Bolch <Bob@...> wrote:

WAKEUP is one of the many variants, commonly in use starting in VM/370, of a server

machine which starts?various tasks according to a schedule. It uses a control file named?WAKEUP TIMES.

Each line in the file contains a command to be executed, and when to execute it.

?

I will try to cobble a HELP file entry together, in the next few days, describing the format

of each entry.??

?

Bob Bolch

?

?

On Wed, Apr 19, 2023 at 10:47?AM Bertram Moshier <herc370390vm@...> wrote:

Hello,

?

I'm trying to find information about the WAKEUP user.? After logging into WAKEUP and trying the "help" command, I'm left not understanding it.

?

I looked through the text files for VM/CE and found:

?

19. Build WAKEUP server ID and add entry to autolog it in AUTOLOG1's PROFILE EXEC.
? ? This server issues commands at specified times during the day, week, month, or year.??

?

This isn't helpful in using it (at least for me).

?

I would appreciate help in finding information on this user / program.

?

Thank you.

?

?


Re: Wakeup userid - what is it?

 

开云体育

Bert,

You can shut the machine down from MAINT or OPERATOR by “SMSG WAKEUP SHUTDOWN” => yes I looked at the source to figure this out.

If you then logon to the wakeup VM it starts as a normal VM and you can edit the files.

The whole machine is implemented as REXX EXECs so all the source is on your “A” disk.

The format of the WAKEUP file is documented in the end of the $WAITF EXEC

To restart type “wakeprof” and then “#cp disc”

Dave

?

From: [email protected] <[email protected]> On Behalf Of Bertram Moshier
Sent: Thursday, April 20, 2023 4:21 AM
To: [email protected]
Subject: Re: [h390-vm] Wakeup userid - what is it?

?

Hi Bob,

?

Thank you for your reply.? I'm looking forward to the HELP file.

?

I hope you'll consider including any commands someone can use after logging onto the userid.? After all, there must be someway to exit to CMS so one can modify the WAKEUP TIMES file.? I'm wondering is there a way to look at the entries from the command line (or other stuff) without? exiting to CMS.

?

Is the source code for WAKEUP on user's minidisks?

?

Bert

?

On Wed, Apr 19, 2023, 10:36 Bob Bolch <Bob@...> wrote:

WAKEUP is one of the many variants, commonly in use starting in VM/370, of a server

machine which starts?various tasks according to a schedule. It uses a control file named?WAKEUP TIMES.

Each line in the file contains a command to be executed, and when to execute it.

?

I will try to cobble a HELP file entry together, in the next few days, describing the format

of each entry.??

?

Bob Bolch

?

?

On Wed, Apr 19, 2023 at 10:47?AM Bertram Moshier <herc370390vm@...> wrote:

Hello,

?

I'm trying to find information about the WAKEUP user.? After logging into WAKEUP and trying the "help" command, I'm left not understanding it.

?

I looked through the text files for VM/CE and found:

?

19. Build WAKEUP server ID and add entry to autolog it in AUTOLOG1's PROFILE EXEC.
? ? This server issues commands at specified times during the day, week, month, or year.??

?

This isn't helpful in using it (at least for me).

?

I would appreciate help in finding information on this user / program.

?

Thank you.

?

?


Re: Wakeup userid - what is it?

 

Hello Dennis!

Thank you so very much for your reply!

Bert


On Thu, Apr 20, 2023, 00:20 Dennis Boone <drb@...> wrote:
?> Thank you for your reply.? I'm looking forward to the HELP file.

I've included a WAKEUP HELP file below.

I have no idea what version of WAKEUP provided this file, and it's
nearly guaranteed to be different from whatever is in the CE.? That
said, it may be a source of clues.

?> I hope you'll consider including any commands someone can use after
?> logging onto the userid.? After all, there must be someway to exit to
?> CMS so one can modify the WAKEUP TIMES file.? I'm wondering is there
?> a way to look at the entries from the command line (or other stuff)
?> without exiting to CMS.

If you want to make your life sane, you rig your REXX or EXEC2 script to
wrangle WAKEUP to recognize a reload command sent by SMSG, and have it
reload the WAKEUP TIMES file.? In the service machine, put the file on a
read only minidisk.? That way you can link it to place the new file
there, release it, then tell the service machine to reaccess and reload.

The procedure for putting the updated file onto the service machine
mdisk is:

0. Do sane error handling as you...

1. Get an MR link to that mdisk

2. Access the mdisk

3. Copy the new file onto the target mdisk as UPD CMSUT1 or some such

4. Rename the old file out of the way; we always prepended a digit to
the filemode, and rolled older copies on down (1EXEC -> 2EXEC, EXEC ->
1EXEC, etc.)

5. Rename UPD CMSUT1 to the target fn fm

6. Release/detach the mdisk

7. Tell the service machine to refresh itself (reaccess mdisk, reread
file)

Bonus points: log files

Obviously this is easier to do if it's scripted up in REXX.

De


WAKEUP

WAKEUP is a? module designed to control? the operation of event? driven virtual
machines.? WAKEUP? may be? used to "wake? up" a virtual? machine at? a specific
time, or when some event occurs (e.g., receiving a file).

The format of WAKEUP is:
+------------+---------------------------------------------------------+
|? ? ? ? ? ? |? ? ? ?< at >? ? < HH:MM<:SS>? >? ? ?< ( options > <)>? ?|
|? ?WAKEUP? ?|? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?|
|? ? ? ? ? ? |? ? ? ? ? ? ? ? ?<+<HH:>MM<:ss>>? ? ?< ( options > <)>? ?|
|? ? ? ? ? ? |? ? ? ? ? ? ? ? ?<RESET>? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?|
+------------+---------------------------------------------------------+

Options:

? ? ?CONS? ? ? -? Wake up on a console attention or line in the stack
? ? ?EXT? ? ? ?-? Wake up on an external interrupt
? ? ?FILE? ? ? -? Wake up at the next time in the 'wakeup times' file
? ? ?IO? ? ? ? -? Wake up on an I/O interrupt
? ? ?IUCVMSG? ?-? Wake up when a message arrives (use IUCV to receive it)
? ? ?RDR? ? ? ?-? Wake up when a file is in the reader
? ? ?SPM? ? ? ?-? Wake up when a message arrives (use SPM to receive it)
? ? ?SMSG? ? ? -? Wake up when an SMSG or VMCF SENDX transmission arrives
? ? ?VMCF? ? ? -? Wake up when a VMCF SENDX message arrives

? ? ?NOEXT? ? ?-? Prevent DEBUG entry on external interrupts
? ? ?RESEXT? ? -? Reset a previous NOEXT option
? ? ?TIME? ? ? -? Stack the time and date on exit
? ? ?QUIET? ? ?-? Don't show any informational messages
? ? ?CC? ? ? ? -? Use Clock Comparator for timing (instead of STIMER)
? ? ?NOREAD? ? -? Do not issue a read on console attention interrupt

If more than? one option is specified,? WAKEUP will return control? when any of
the? specified? events? occur.? ?Thus,? "WAKEUP +30 (RDR IUCVMSG"? will? return
control in 30? minutes, as soon as a? file arrives, or when? a message arrives,
whichever occurs first.

Normal return codes:

? ? ?0 - All TIME file records for today have been executed
? ? ?1 - VMCF or SMSG message arrived
? ? ?2 - Time (HH:MM:SS or +MM) expired
? ? ?3 - Time from timer file expired
? ? ?4 - File in the reader
? ? ?5 - Message arrived (via IUCVMSG or SPM)
? ? ?6 - Console attention
? ? ?7 - External Interrupt
? ? ?8 - I/O interrupt

An interrupt? from the? console always? causes control? to be? returned with? a
return code of 6, regardless of the options given when WAKEUP was called.? This
usually means someone has? logged on the virtual machine and? hit the ENTER (or
carriage return) key to cause the interrupt.

Interrupts from? the FILE,? IUCVMSG, VMCF,? SMSG, EXT,? and I/O? options causes
WAKEUP to put details of the interrupt on the stack.

Some error return codes

? ? ?28? - Timer file not found ("STATE WAKEUP TIMES" failed)
? ? ?100 - Explanation complete (result of 'wakeup ?')
? ? ?1xx - Error from RDCARD macro
? ? ?101 - Empty file in card reader
? ? ?103 - Unknown error from card reader
? ? ?200 - Invalid option (For example,? WAKEUP (XXX ) )
? ? ?201 - Virtual reader not operational (rdcard rc=100)
? ? ?202 - Error from spool rdr hold
? ? ?203 - Invalid hours field
? ? ?204 - Invalid minutes field
? ? ?205 - Invalid seconds field
? ? ?206 - Invalid parameter
? ? ?207 - Error reading timer file
? ? ?208 - Error from vmcf authorise
? ? ?209 - Sendx data length > 100
? ? ?210 - VMCF data transfer error
? ? ?211? ?Error writing time/datestamp to timer file
? ? ?212? ?Time parm error in timer file
? ? ?213? ?Wakeup times file not on r/w disk
? ? ?214? ?I/O or ext new psw = mine (serious internal wakeup error)
? ? ?215? ?Relative time specified with monthly/yearly
? ? ?3xx? ?IUCV error






Re: Wakeup userid - what is it?

 

Hi!

Where is the source code for WAKEUP?? I apologize, in advance, should you already answer this question.?

Bert

On Thu, Apr 20, 2023, 00:39 Dennis Boone <drb@...> wrote:
The current-ish IBM documentation for WAKEUP includes some sample exec
snippets that show how you approach using the beast:



De






Re: Wakeup userid - what is it?

 

The current-ish IBM documentation for WAKEUP includes some sample exec
snippets that show how you approach using the beast:



De


Re: Wakeup userid - what is it?

 

Thank you for your reply. I'm looking forward to the HELP file.
I've included a WAKEUP HELP file below.

I have no idea what version of WAKEUP provided this file, and it's
nearly guaranteed to be different from whatever is in the CE. That
said, it may be a source of clues.

> I hope you'll consider including any commands someone can use after
> logging onto the userid. After all, there must be someway to exit to
> CMS so one can modify the WAKEUP TIMES file. I'm wondering is there
> a way to look at the entries from the command line (or other stuff)
> without exiting to CMS.

If you want to make your life sane, you rig your REXX or EXEC2 script to
wrangle WAKEUP to recognize a reload command sent by SMSG, and have it
reload the WAKEUP TIMES file. In the service machine, put the file on a
read only minidisk. That way you can link it to place the new file
there, release it, then tell the service machine to reaccess and reload.

The procedure for putting the updated file onto the service machine
mdisk is:

0. Do sane error handling as you...

1. Get an MR link to that mdisk

2. Access the mdisk

3. Copy the new file onto the target mdisk as UPD CMSUT1 or some such

4. Rename the old file out of the way; we always prepended a digit to
the filemode, and rolled older copies on down (1EXEC -> 2EXEC, EXEC ->
1EXEC, etc.)

5. Rename UPD CMSUT1 to the target fn fm

6. Release/detach the mdisk

7. Tell the service machine to refresh itself (reaccess mdisk, reread
file)

Bonus points: log files

Obviously this is easier to do if it's scripted up in REXX.

De


WAKEUP

WAKEUP is a module designed to control the operation of event driven virtual
machines. WAKEUP may be used to "wake up" a virtual machine at a specific
time, or when some event occurs (e.g., receiving a file).

The format of WAKEUP is:
+------------+---------------------------------------------------------+
| | < at > < HH:MM<:SS> > < ( options > <)> |
| WAKEUP | |
| | <+<HH:>MM<:ss>> < ( options > <)> |
| | <RESET> |
+------------+---------------------------------------------------------+

Options:

CONS - Wake up on a console attention or line in the stack
EXT - Wake up on an external interrupt
FILE - Wake up at the next time in the 'wakeup times' file
IO - Wake up on an I/O interrupt
IUCVMSG - Wake up when a message arrives (use IUCV to receive it)
RDR - Wake up when a file is in the reader
SPM - Wake up when a message arrives (use SPM to receive it)
SMSG - Wake up when an SMSG or VMCF SENDX transmission arrives
VMCF - Wake up when a VMCF SENDX message arrives

NOEXT - Prevent DEBUG entry on external interrupts
RESEXT - Reset a previous NOEXT option
TIME - Stack the time and date on exit
QUIET - Don't show any informational messages
CC - Use Clock Comparator for timing (instead of STIMER)
NOREAD - Do not issue a read on console attention interrupt

If more than one option is specified, WAKEUP will return control when any of
the specified events occur. Thus, "WAKEUP +30 (RDR IUCVMSG" will return
control in 30 minutes, as soon as a file arrives, or when a message arrives,
whichever occurs first.

Normal return codes:

0 - All TIME file records for today have been executed
1 - VMCF or SMSG message arrived
2 - Time (HH:MM:SS or +MM) expired
3 - Time from timer file expired
4 - File in the reader
5 - Message arrived (via IUCVMSG or SPM)
6 - Console attention
7 - External Interrupt
8 - I/O interrupt

An interrupt from the console always causes control to be returned with a
return code of 6, regardless of the options given when WAKEUP was called. This
usually means someone has logged on the virtual machine and hit the ENTER (or
carriage return) key to cause the interrupt.

Interrupts from the FILE, IUCVMSG, VMCF, SMSG, EXT, and I/O options causes
WAKEUP to put details of the interrupt on the stack.

Some error return codes

28 - Timer file not found ("STATE WAKEUP TIMES" failed)
100 - Explanation complete (result of 'wakeup ?')
1xx - Error from RDCARD macro
101 - Empty file in card reader
103 - Unknown error from card reader
200 - Invalid option (For example, WAKEUP (XXX ) )
201 - Virtual reader not operational (rdcard rc=100)
202 - Error from spool rdr hold
203 - Invalid hours field
204 - Invalid minutes field
205 - Invalid seconds field
206 - Invalid parameter
207 - Error reading timer file
208 - Error from vmcf authorise
209 - Sendx data length > 100
210 - VMCF data transfer error
211 Error writing time/datestamp to timer file
212 Time parm error in timer file
213 Wakeup times file not on r/w disk
214 I/O or ext new psw = mine (serious internal wakeup error)
215 Relative time specified with monthly/yearly
3xx IUCV error