开云体育

If you guys want anyone to test Confer out on 6.0, I can do so if someone can copy the files into an .AWS tape file.

?

From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Sunday, February 26, 2023 16:09
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of serveral things

?

Most of these issues would be addressed by a release of the 1996 system (maybe D7.0). You're more or less right about the status of various pieces of software. PascalJB would be easy to install, but Confer requires a bit more work since it requires various userIDs and files in obscure locations. I know that Jeff Ogden has looked into this and thinks it is possible, but it hasn't been done.

There is no interest in MTS at UM and no one is running or studying it officially. Most people aren't aware that it existed. I don't know if there is any interest in Software Archaeology here. A couple of years ago I gave a short talk in Newcastle about the history of MTS. I know the CS department there has a significant interest in the history of IT. II was impressed that the Mayor of Newcastle attended, complete with his regalia, and said he remembered using MTS when he aws a student at Newcastle.

Thanks for nudging me about the 1996 system. I really want to get that done but I have a few higher priority things to clear up first.

Mike

On 26 Feb 2023, at 15:15, John Palmer wrote:

I saw an e-mail posted on the MTS archive site from Bon Parnes (Confer creator) that seemed to indicate that he was OK with distributing Confer to people. Is there any way we could get an .AWS file with the Confer program components?
?
I was also wondering about Fortran 77 (*FTNVS) and Pascal VS. I imagine that those are probably still protected IBM property and are probably in the same boat as *ASMH. ??If Pascal VS or *FTNVS are available, it would be cool if we could get those somehow as well.
?
I also understand that the creator of *PascalJB has indicated that it can be made available as well.
?
I was wondering, in general, what is the status of MTS at U of M? Does U of M have any official instances of MTS running under Hercules?
?
I know that there is a new sub-field of CS called “Software Archaeology” that studies older software systems (for various reasons) – someone did a PHD on this recently. I was wondering if maybe the CS department had it running.
?
Sill anxiously awaiting the release of “d7.0”. I know a lot of people are doing this work on a voluntary basis, and we all appreciate that a lot.
?
Thanks
John

#124

sir,

I would very interested, if possible.

Thank you

tom c

Show quoted text

On Mon, Mar 6, 2023 at 11:15?AM John Palmer <jpalmer@...> wrote:

If you guys want anyone to test Confer out on 6.0, I can do so if someone can copy the files into an .AWS tape file.
?
From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Sunday, February 26, 2023 16:09
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of serveral things
?
Most of these issues would be addressed by a release of the 1996 system (maybe D7.0). You're more or less right about the status of various pieces of software. PascalJB would be easy to install, but Confer requires a bit more work since it requires various userIDs and files in obscure locations. I know that Jeff Ogden has looked into this and thinks it is possible, but it hasn't been done.
There is no interest in MTS at UM and no one is running or studying it officially. Most people aren't aware that it existed. I don't know if there is any interest in Software Archaeology here. A couple of years ago I gave a short talk in Newcastle about the history of MTS. I know the CS department there has a significant interest in the history of IT. II was impressed that the Mayor of Newcastle attended, complete with his regalia, and said he remembered using MTS when he aws a student at Newcastle.
Thanks for nudging me about the 1996 system. I really want to get that done but I have a few higher priority things to clear up first.
Mike
On 26 Feb 2023, at 15:15, John Palmer wrote:
I saw an e-mail posted on the MTS archive site from Bon Parnes (Confer creator) that seemed to indicate that he was OK with distributing Confer to people. Is there any way we could get an .AWS file with the Confer program components?
?
I was also wondering about Fortran 77 (*FTNVS) and Pascal VS. I imagine that those are probably still protected IBM property and are probably in the same boat as *ASMH. ??If Pascal VS or *FTNVS are available, it would be cool if we could get those somehow as well.
?
I also understand that the creator of *PascalJB has indicated that it can be made available as well.
?
I was wondering, in general, what is the status of MTS at U of M? Does U of M have any official instances of MTS running under Hercules?
?
I know that there is a new sub-field of CS called “Software Archaeology” that studies older software systems (for various reasons) – someone did a PHD on this recently. I was wondering if maybe the CS department had it running.
?
Sill anxiously awaiting the release of “d7.0”. I know a lot of people are doing this work on a voluntary basis, and we all appreciate that a lot.
?
Thanks
John

#151

开云体育

I think we will probably have to wait till d7.0 comes out, but it would be nice if someone could get us the Confer files to try out (I know, I keep pestering about this, but Confer is a centerpiece of MTS).

I’d be willing to do the work to try to get it working.? Is Jeff Ogden on this list? I heard that he may have some idea on what would be needed to get Confer working. I’m guessing that we need all of the files on user CNFR as well as several hooks related to surcharges.? I seem to recall that, at WSU, there was some file under CCID “BILL” that contained some sort of surcharge related data.? I noticed that the BILL user was not included with d6.0, probably because it contained U of M related billing software.

?

On the subject of Hercules changes,? Mike, I think you said that you still have to find out how to integrate your Hercules changes for the HIM. I’m going to upgrade to CentOS 9 from 7 because, for some reason, large data transfers hang. I’d like to upgrade Hercules at the same time.

?

Thanks

?

Show quoted text

From: [email protected] <[email protected]> On Behalf Of Tom Chandler
Sent: Monday, March 6, 2023 13:04
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

?

sir,

I would very interested, if possible.

?

Thank you

tom c

?

On Mon, Mar 6, 2023 at 11:15?AM John Palmer <jpalmer@...> wrote:

If you guys want anyone to test Confer out on 6.0, I can do so if someone can copy the files into an .AWS tape file.
?
From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Sunday, February 26, 2023 16:09
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of serveral things
?
Most of these issues would be addressed by a release of the 1996 system (maybe D7.0). You're more or less right about the status of various pieces of software. PascalJB would be easy to install, but Confer requires a bit more work since it requires various userIDs and files in obscure locations. I know that Jeff Ogden has looked into this and thinks it is possible, but it hasn't been done.
There is no interest in MTS at UM and no one is running or studying it officially. Most people aren't aware that it existed. I don't know if there is any interest in Software Archaeology here. A couple of years ago I gave a short talk in Newcastle about the history of MTS. I know the CS department there has a significant interest in the history of IT. II was impressed that the Mayor of Newcastle attended, complete with his regalia, and said he remembered using MTS when he aws a student at Newcastle.
Thanks for nudging me about the 1996 system. I really want to get that done but I have a few higher priority things to clear up first.
Mike
On 26 Feb 2023, at 15:15, John Palmer wrote:
I saw an e-mail posted on the MTS archive site from Bon Parnes (Confer creator) that seemed to indicate that he was OK with distributing Confer to people. Is there any way we could get an .AWS file with the Confer program components?
?
I was also wondering about Fortran 77 (*FTNVS) and Pascal VS. I imagine that those are probably still protected IBM property and are probably in the same boat as *ASMH. ??If Pascal VS or *FTNVS are available, it would be cool if we could get those somehow as well.
?
I also understand that the creator of *PascalJB has indicated that it can be made available as well.
?
I was wondering, in general, what is the status of MTS at U of M? Does U of M have any official instances of MTS running under Hercules?
?
I know that there is a new sub-field of CS called “Software Archaeology” that studies older software systems (for various reasons) – someone did a PHD on this recently. I was wondering if maybe the CS department had it running.
?
Sill anxiously awaiting the release of “d7.0”. I know a lot of people are doing this work on a voluntary basis, and we all appreciate that a lot.
?
Thanks
John

#152

I don't think that Jeff is on this list, but I know he has the Confer files and permission to distribute them.  Getting it running in MTS is not as simple as copying the files and running it.  You also need to create at least one User ID and several configuration files.  I think Jeff has instructions for doing this, but I'm not sure if he has tried doing it.  The plan (such as it is) is to include Confer installed in the next version of MTS.

As far as HIM changes and related stuff, I really intend to work on this over the next few weeks.  This may or may not happen before your upgrade.

Mike

Show quoted text

On 18 Jul 2023, at 12:57, John Palmer wrote:

I think we will probably have to wait till d7.0 comes out, but it would be nice if someone could get us the Confer files to try out (I know, I keep pestering about this, but Confer is a centerpiece of MTS).

I’d be willing to do the work to try to get it working.? Is Jeff Ogden on this list? I heard that he may have some idea on what would be needed to get Confer working. I’m guessing that we need all of the files on user CNFR as well as several hooks related to surcharges.? I seem to recall that, at WSU, there was some file under CCID “BILL” that contained some sort of surcharge related data.? I noticed that the BILL user was not included with d6.0, probably because it contained U of M related billing software.

On the subject of Hercules changes,? Mike, I think you said that you still have to find out how to integrate your Hercules changes for the HIM. I’m going to upgrade to CentOS 9 from 7 because, for some reason, large data transfers hang. I’d like to upgrade Hercules at the same time.

Thanks

#153

Thanks for the reply.

I'm actually moving back to NetBSD, which is what I usually use. Just set up a NetBSD 9.3 VM and installed Hercules 4.5 (they have a pre-built pkgsrc install for it).

Problem is, MTS superdumps during the IPL process when its trying to print out a message about this IPL time being more than 12 hours than the last one. It can print the current date, but not the date of the last IPL.

I'll try to get the super-dump to a tape file, it keeps rejecting the tape drive designation. Do I need to initialize it somehow other than the hercules devint command? 

John

Show quoted text

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Friday, July 21, 2023 18:48
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

I don't think that Jeff is on this list, but I know he has the Confer files and permission to distribute them.  Getting it running in MTS is not as simple as copying the files and running it.  You also need to create at least one User ID and several configuration files.  I think Jeff has instructions for doing this, but I'm not sure if he has tried doing it.  The plan (such as it is) is to include Confer installed in the next version of MTS.

As far as HIM changes and related stuff, I really intend to work on this over the next few weeks.  This may or may not happen before your upgrade.

Mike

On 18 Jul 2023, at 12:57, John Palmer wrote:

I think we will probably have to wait till d7.0 comes out, but it 
would be nice if someone could get us the Confer files to try out (I 
know, I keep pestering about this, but Confer is a centerpiece of 
MTS).

I’d be willing to do the work to try to get it working.  Is Jeff Ogden 
on this list? I heard that he may have some idea on what would be 
needed to get Confer working. I’m guessing that we need all of the 
files on user CNFR as well as several hooks related to surcharges.  I 
seem to recall that, at WSU, there was some file under CCID “BILL” 
that contained some sort of surcharge related data.
I noticed that the BILL user was not included with d6.0, probably 
because it contained U of M related billing software.

On the subject of Hercules changes,  Mike, I think you said that you 
still have to find out how to integrate your Hercules changes for the 
HIM. I’m going to upgrade to CentOS 9 from 7 because, for some reason, 
large data transfers hang. I’d like to upgrade Hercules at the same 
time.

Thanks

#154

开云体育

On 21 Jul 2023, at 22:23, John Palmer wrote:

I'll try to get the super-dump to a tape file, it keeps rejecting the tape drive designation. Do I need to initialize it somehow other than the hercules devint command?

Yes, dump tapes must be labeled tapes with a volume name starting with DUMP. You can label the tape using MTS or some other operating system or you can use the lbltp command line program. See . It should work in most Unix-like systems or in Windows.

Mike

#155

开云体育

On 21 Jul 2023, at 22:23, John Palmer wrote:

I'll try to get the super-dump to a tape file, it keeps rejecting the tape drive designation. Do I need to initialize it somehow other than the hercules devint command?

Yes, dump tapes must be labeled tapes with a volume name starting with DUMP. You can label the tape using MTS or some other operating system or you can use the lbltp command line program. See . It should work in most Unix-like systems or in Windows.

Mike

PS: I tried to send this earlier but it didn't seem to get sent. Sorry if you get two copies.

#156

开云体育

I’ll try this evening, but just actually got some info on the hercules console:

?

/00000 16:41:16??????????? ** UMMPS/XA ** assembled 04/19/88

/00000 16:41:16?? CONFIG? CPU 0000 online

/00001 16:41:16???? INIT? Time and date have been set to? 16:41:16 on 07-25-23.

/00002 16:41:16 OPERATOR??? Started at 16:41:16 on 07-25-23? Model 3090 Serial 010611

/00002 16:41:16 OPERATOR??? Using display DS01 & printer CON1 (1052)

HHC00107I Starting thread cckd_writer(), active=0, started=0, max=2

HHC00100I Thread id 00006faf0aaee000, prio 1, name 'cckd_writer thread 1' started

/00004 16:41:18????? MTS?? Current time of 16:41 Tue Jul 25 is more than 12 hours later than

HHC00801I Processor CP00: Protection exception interruption code 0004 ilc 4

HHC02324I PSW=070D100000164324 INST=42609011???? STC?? 6,17(0,9)????????????? store_character

HHC02326I V:00000011:R:0000000000000011:K:0E=000000 00000000 033D2000 0001CE2A 07 ................

HHC02269I GR00=00000018 GR01=00166780 GR02=00166254 GR03=F560F2F3

HHC02269I GR04=00000000 GR05=00000000 GR06=FFFFFFF2 GR07=00000000

HHC02269I GR08=00036A60 GR09=00000000 GR10=00162D18 GR11=00163D18

HHC02269I GR12=00165EC8 GR13=8016421E GR14=80164302 GR15=00000000

HHC02271I CR00=18B0FE40 CR01=0012D000 CR02=00000000 CR03=FFFF0000

HHC02271I CR04=00000000 CR05=00000000 CR06=70000000 CR07=00000000

HHC02271I CR08=000000FF CR09=00000000 CR10=00000000 CR11=00000000

HHC02271I CR12=000B3F1D CR13=00000000 CR14=1F000000 CR15=00000000

?

It seems to be trying to craft a message to display the last IPL time.?? I’m assuming that maybe register 9 should have a value in it, but maybe not.

Also, Hercules doesn’t like ARCHMODE ESA/390 and it has the wrong serial number (010611 rather than the 000611 in the cnf file).

?

I posted the cnf file on the Hercules 390 group. I’m guessing some settings changed between 3.13 and 4.5.

?

Thanks

John

?

Show quoted text

From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Saturday, July 22, 2023 15:06
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

?

On 21 Jul 2023, at 22:23, John Palmer wrote:

I'll try to get the super-dump to a tape file, it keeps rejecting the tape drive designation. Do I need to initialize it somehow other than the hercules devint command?

Yes, dump tapes must be labeled tapes with a volume name starting with DUMP. You can label the tape using MTS or some other operating system or you can use the lbltp command line program. See . It should work in most Unix-like systems or in Windows.

Mike

PS: I tried to send this earlier but it didn't seem to get sent. Sorry if you get two copies.

#157

John,

There are differences between the cnf file for hercules 4.5 and hercules
3.13

My 4.5 cnf file (windows) is as follows:-

====================================================================================================

#########################################
# Hercules emulator configuration file  #
#    Version AN302 (30 January 2012)    #
#   A template for use with MTS D6.0A   #
#########################################

# Adjust this template for use in your own environment

# You'll need to put the D6 files in the right places.

# For D6.0, but not for D6.0A you need to create a 
# 3380 disk image before you start using something 
# similar to:
#    dasdinit -z -a mts600.dsk 3380-1 MTS600

# In general the Hercules device configuration needs to
# match or at least be a subset of the D6.0 version
# of TABLES as described in the following list from
# the D6.0-NEWSYS.txt writeup:
#
#    Device Type     Address            Device name
#    -----------     ---------          -----------
#
#    3287            0000               PTR1  
#    3270            0001-0007          DS01-DS07
#    3270 (in D5.0)  0009               CON1 
#    1052 (in D6.0A) 0009               CON1 
#
#    2540R           000C               RDR1
#    2540P           000D               PCH1
#    1403            000E               PTR2
#
#    3270            001F               CON2
#
#    3287            0100               PTR3  
#    3270            0101-011F          DS21-DS3F
#    3420            0180-018F          T900-T90F
#
#    3330            0200-020F          D100-D10F
#    3350            0220-022F          D200-D20F
#    3370            0240-024F          D300-D30F
#    3380            0260-026F          D400-D40F
#
#    3420            0C70               T920
#
#    9335            0D00-0D03          D500-D503
#    9335            0E00-0E03          D600-D603

ARCHLVL   ESA/390  # includes 370-XA (MTS architecture code "MP")
CODEPAGE  819/037
SHCMDOPT             ENABLE NODIAG8
HTTP PORT             8041 NOAUTH
HTTP START
CPUSERIAL 000735
CPUMODEL  3090
CPUVERID  FF       # the value FF here causes MTS to bypass
                   # its reguests to set the TOD clock
MAINSIZE  128
XPNDSIZE 0 
PANTITLE "MTS 6.0A"

# Adjust the following two values for your local configuration
# MTS does support operation on multi-CPU configurations with
# (in theory) up to 32 processors, although 6 processors is likely
# the maximum number ever used in "real life".

NUMCPU   1 
#MAXCPU   1

SYSEPOCH 1900
##TZOFFSET -0000  # MTS uses GMT for its TOD clock
#TZOFFSET  +0100         #BST


# YROFFSET is not needed for the D6.0A or D1996.0 versions of MTS.
# For D6.0 uncomment the following parameter. 

# YROFFSET -23  # For D6.0, but not D6.0A set the year back to 
                # avoid time/date rollover issues.
                # For D6.0, needs to give dates after D6.0 was 
                # released in 1988, but before November 1989.

CNSLPORT 3241   # the port that tn3270 sessions will use to connect
OSTAILOR QUIET

###### Device definitions #####

# Display and console devices

# The definition for address 0009 below doesn't match the D6.0 
# definition in TABLES, but in D6.0A TABLES is patched to make 
# address 0009 a 1052 console/printer and CON1 will be used as 
# the MTS Operator's Console log printer with its output directed 
# to the Hercules console.

0000   3211    "h:\Hercules\MTS-D6-0A\Prt/PTR1x.txt" lpp=66 fcb=1:4,2:34,3:66,4:19,5:5,6:14,7:54,8:63,10:8,11:1 # PTR1, a 3270 data stream printer
#0000   3211    "h:\Hercules\MTS-D6-0A\Prt/PTR1x.txt" lpp=66 fcb=4:1,34:2,66:3,19:4,5:5,14:6,54:7,63:8,8:10,1:11 print # PTR1, a 3270 data stream printer
0001.7 3270    # DS01 through DS07
0009   1052-C  # CON1, comment this line out and
# 0009 3270    # CON1, uncomment this line if you want the
               #       definition to match the definition
               #       in the D6.0 version of TABLES 
001F   3270    # CON2
0100   3287    # PTR3, a 3270 data stream printer

0101.7 3270    # DS21 through DS27

# Unit record devices

000C   3505   Units/RDR1.txt ascii eof   # RDR1
000D   3525   Units/PCH1.txt ascii       # PCH1

# Diferent versions of Hercules support different printer options.
# Uncomment one of the following three device definitions for addr 000E.
# The three definitions are listed in order from newest to oldest
# and from best to worst in terms of functionality. 
# The fcb= option defines the MTS printer carriage control tape.
# The nofcbcheck option supresses errors (command reject) when
# a skip to an undefined printer carriage tape channel is used.
# If the third (last) option is used, errors will result when 
# HASP tries to print its head and tail sheets on the printer,
# effectively making HASP printing unusable.

#000E 1403   Units/PTR2.txt lpp=66 fcb=1:11,4:1,5:5,8:10,14:6,19:4,24:7,34:2,44:6,49:4,54:7,63:8,66:3 optprint # PTR2
#000E 1403   Units/PTR2.txt lpp=66 fcb=4:1,34:2,66:3,19:4,5:5,14:6,54:7,63:8,8:10,1:11 optprint # PTR2
#000E 1403   "h:\Hercules\MTS-D6-0A\Prt/PTR2x.txt" lpp=66 fcb=4:1,34:2,66:3,19:4,5:5,14:6,54:7,63:8,8:10,1:11 print # PTR2
000E 1403   "h:\Hercules\MTS-D6-0A\Prt/PTR2x.txt" lpp=66 cctape=(4=1,34=2,66=3,19=4,5=5,14=6,54=7,63=8,8=10,1=11) # PTR2
#000E 1403   Units/PTR2.txt lpp=60 fcb=1:11,4:1,5:5,8:10,14:6,19:4,24:7,34:2,44:6,49:4,54:7,63:8,66:3 optprint # PTR2
# 000E 1403   Units/PTR2.txt nofcbcheck  # PTR2
# 000E   1403   Units/PTR2.txt             # PTR2

# Tapes

# These ten tape definitions are not required for D6.0A
# and may be commented out.

0180   3420   Tapes/d6.0util.aws ro  # T900, D6.0 utilities tape
0181   3420   Tapes/d6.0dr1.aws  ro  # T901, D6.0 IPLable dump/restore tape #1
0182   3420   Tapes/d6.0dr2.aws  ro  # T902, D6.0 dump/restore tape #2
0183   3420   Tapes/d6.0dr3.aws  ro  # T903, D6.0 dump/restore tape #3
0184   3420   Tapes/d6.0t1.aws   ro  # T904, *FS D6.0 distribution tapes #1
0185   3420   Tapes/d6.0t2.aws   ro  # T905,  " #2
0186   3420   Tapes/d6.0t3.aws   ro  # T906,  " #3
0187   3420   Tapes/d6.0t4.aws   ro  # T907,  " #4
0188   3420   Tapes/d6.0t5.aws   ro  # T908,  " #5
0189   3420   Tapes/d6.0t6.aws   ro  # T909,  " #6
018A   3420   Tapes/cmd001.aws       # T90A, a CMDSTAT tape for use by *DWB

# Devices 018B through 018F and 0C70 (T90B to T90F and T920) are defined
# as 9TP (3420) tape drives in TABLES and so could be defined here

# Disk(s)

# Shadow files (sf=) won't be used until the Hercules command "sf+" 
# is issued, unless the shadow file already exists or the base and 
# shadow files are read-only in which case a new shadow file will 
# be created.

0260   3380   "h:/hercules/MTS-D6-0A/Disks/mts600.dsk" ro sf="h:/hercules/MTS-D6-0A/Disks/Shadows/mts600_*.dsk" # D400

# D6.0 and D6.0A of MTS support single density (3380-1), but not 
# double density (3380-2) or triple density (3380-3), 3380 disks.

# New 3380 disks need to be initialized using dasdinit's -a option 
# so that the alternate tracks are included. For example:
#    dasdinit -z -a mts600.dsk 3380-1 MTS600

# Devices 0200-020F (D100-D10F), 0220-022F (D200-D20F),
# 0240-024F (D300-D30F) and 0261-026F (D401-D40F) are 
# defined as 3330, 3350, 3370, and 3380 disks and so
# copuld be defined here.

# Devices 0D00-0D03 (D500-D503) and 0E00-0E03 (D600-D603) are defined
# as 9335 disks in the D6.0 and D6.0A TABLES, but there are bugs in
# the D6.0 and D6.0A 9335 support, so these devices can't be used.

# END

====================================================================================================


Chris
-- 
 <cjar1950@...>



----------------------------------------------------------------------------------------------------------------------------------
On Tue, 25 Jul 2023 16:55:33 -0500
"John Palmer" <jpalmer@...> wrote:
I'll try this evening, but just actually got some info on the hercules
console:

 

/00000 16:41:16            ** UMMPS/XA ** assembled 04/19/88

/00000 16:41:16   CONFIG  CPU 0000 online

/00001 16:41:16     INIT  Time and date have been set to  16:41:16 on
07-25-23.

/00002 16:41:16 OPERATOR    Started at 16:41:16 on 07-25-23  Model 3090
Serial 010611

/00002 16:41:16 OPERATOR    Using display DS01 & printer CON1 (1052)

HHC00107I Starting thread cckd_writer(), active=0, started=0, max=2

HHC00100I Thread id 00006faf0aaee000, prio 1, name 'cckd_writer thread 1'
started

/00004 16:41:18      MTS   Current time of 16:41 Tue Jul 25 is more than 12
hours later than

HHC00801I Processor CP00: Protection exception interruption code 0004 ilc 4

HHC02324I PSW=070D100000164324 INST=42609011     STC   6,17(0,9)
store_character

HHC02326I V:00000011:R:0000000000000011:K:0E=000000 00000000 033D2000
0001CE2A 07 ................

HHC02269I GR00=00000018 GR01=00166780 GR02=00166254 GR03=F560F2F3

HHC02269I GR04=00000000 GR05=00000000 GR06=FFFFFFF2 GR07=00000000

HHC02269I GR08=00036A60 GR09=00000000 GR10=00162D18 GR11=00163D18

HHC02269I GR12=00165EC8 GR13=8016421E GR14=80164302 GR15=00000000

HHC02271I CR00=18B0FE40 CR01=0012D000 CR02=00000000 CR03=FFFF0000

HHC02271I CR04=00000000 CR05=00000000 CR06=70000000 CR07=00000000

HHC02271I CR08=000000FF CR09=00000000 CR10=00000000 CR11=00000000

HHC02271I CR12=000B3F1D CR13=00000000 CR14=1F000000 CR15=00000000

 

It seems to be trying to craft a message to display the last IPL time.   I'm
assuming that maybe register 9 should have a value in it, but maybe not.

Also, Hercules doesn't like ARCHMODE ESA/390 and it has the wrong serial
number (010611 rather than the 000611 in the cnf file).

 

I posted the cnf file on the Hercules 390 group. I'm guessing some settings
changed between 3.13 and 4.5.

 

Thanks

John

 

 

 

From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Saturday, July 22, 2023 15:06
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

 

On 21 Jul 2023, at 22:23, John Palmer wrote:

I'll try to get the super-dump to a tape file, it keeps rejecting the tape
drive designation. Do I need to initialize it somehow other than the
hercules devint command?

Yes, dump tapes must be labeled tapes with a volume name starting with DUMP.
You can label the tape using MTS or some other operating system or you can
use the lbltp command line program. See
<> lbltp. It should work in most
Unix-like systems or in Windows.

Mike

PS: I tried to send this earlier but it didn't seem to get sent. Sorry if
you get two copies.

#158

On Tue, Jul 25, 2023 at 04:55:33PM -0500, John Palmer wrote:
 
I posted the cnf file on the Hercules 390 group. I'm guessing some settings
changed between 3.13 and 4.5.There's no need to guess.  This page:

documents the deprecated and new parameters for Hercules 4.x configuration
files.

-- 

Kevin

Bruceville, TX

What's the definition of a legacy system? One that works!
Errare humanum est, ignoscere caninum.

#159

Kevin,

Or  for SDL
Hercules

Chris
-- 
 <cjar1950@...>

----------------------------------------------------------------------------------------------------------------------------------
On Tue, 25 Jul 2023 18:11:17 -0500
"Kevin Monceaux" <Kevin@...> wrote:
On Tue, Jul 25, 2023 at 04:55:33PM -0500, John Palmer wrote:
 
I posted the cnf file on the Hercules 390 group. I'm guessing some settings
changed between 3.13 and 4.5.There's no need to guess.  This page:

documents the deprecated and new parameters for Hercules 4.x configuration
files.

-- 

Kevin

Bruceville, TX

What's the definition of a legacy system? One that works!
Errare humanum est, ignoscere caninum.

#160

I updated the config based on another MTS user but am still getting that
error.

I'll keep working on it. Thanks for the suggestions.

John

Show quoted text

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Kevin Monceaux
Sent: Tuesday, July 25, 2023 18:11
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

On Tue, Jul 25, 2023 at 04:55:33PM -0500, John Palmer wrote:

I posted the cnf file on the Hercules 390 group. I'm guessing some 
settings changed between 3.13 and 4.5.There's no need to guess.  This page:

documents the deprecated and new parameters for Hercules 4.x configuration
files.

-- 

Kevin

Bruceville, TX

What's the definition of a legacy system? One that works!
Errare humanum est, ignoscere caninum.

#163

On 25 Jul 2023, at 17:55, John Palmer wrote:

HHC00801I Processor CP00: Protection exception interruption code 0004 ilc 4
HHC02324I PSW=070D100000164324 INST=42609011???? STC?? 6,17(0,9)????????????? store_characterI would need a map of the system you are running to see exactly where that PGNT is happening.  It should be in SEG2:S2MAP on that system if you could ever get it up long enough to see it.  If it's the standard D6.0 system you got from us we should have a copy of the map somewhere too.

However, the PGNT is a problem state interrupt which should not cause a SuperDump directly.  It may or may not be an expected interrupt.  MTS often tests an address by poking at it and seeing ig it gets a PGNT.  Such PGNTs are not unexpected.

Mike

#164

Mike - it dies exactly the same way on the standard d6.0 system as well as the one that I have been running for awhile, which started out as the standard d6.0 system.

Show quoted text

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Tuesday, August 1, 2023 21:42
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

On 25 Jul 2023, at 17:55, John Palmer wrote:

HHC00801I Processor CP00: Protection exception interruption code 0004 
ilc 4 HHC02324I PSW=070D100000164324 INST=42609011     STC
6,17(0,9)              store_characterI would need a map of the system you are running to see exactly where that PGNT is happening.  It should be in SEG2:S2MAP on that system if you could ever get it up long enough to see it.  If it's the standard
D6.0 system you got from us we should have a copy of the map somewhere too.

However, the PGNT is a problem state interrupt which should not cause a SuperDump directly.  It may or may not be an expected interrupt.  MTS often tests an address by poking at it and seeing ig it gets a PGNT.  
Such PGNTs are not unexpected.

Mike

#165

Just out of curiosity,  to which version of Hercules are you adding HIM support and what system do you develop on?

Have you tried SDL 4.x (any SDL version) with the standard d6.0a? I can get it to IPL on 4.5 on a CentOS Linux system, but it crashes anywhere from 10 mins to 2 hours.  I tried changing the IODELAY on David Trout's suggestion on the NetBSD system with 4.6 and that didn't do any good.

The old 3.13 installed on NetBSD 9.3 seems to work ok.

Show quoted text

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Mike Alexander
Sent: Tuesday, August 1, 2023 21:42
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

On 25 Jul 2023, at 17:55, John Palmer wrote:

HHC00801I Processor CP00: Protection exception interruption code 0004 
ilc 4 HHC02324I PSW=070D100000164324 INST=42609011     STC
6,17(0,9)              store_characterI would need a map of the system you are running to see exactly where that PGNT is happening.  It should be in SEG2:S2MAP on that system if you could ever get it up long enough to see it.  If it's the standard
D6.0 system you got from us we should have a copy of the map somewhere too.

However, the PGNT is a problem state interrupt which should not cause a SuperDump directly.  It may or may not be an expected interrupt.  MTS often tests an address by poking at it and seeing ig it gets a PGNT.  
Such PGNTs are not unexpected.

Mike

#166

开云体育

On Aug 1, 2023, at 8:56 PM, John Palmer <jpalmer@...> wrote:

Have you tried SDL 4.x (any SDL version) with the standard d6.0a? I can get it to IPL on 4.5 on a CentOS Linux system, but it crashes anywhere from 10 mins to 2 hours. ?I tried changing the IODELAY on David Trout's suggestion on the NetBSD system with 4.6 and that didn't do any good.

The old 3.13 installed on NetBSD 9.3 seems to work ok.

For what it’s worth, I’m able to run MTS 6.0a without a problem on a Mac Studio (M1 Ultra) using the SDL version of Hercules 4.7.0 built using Hercules-Helper. Here’s some more version info:

HHC01413I Hercules version 4.7.0.11007-SDL-DEV-gcddce71b

HHC01417I Built using Hercules-Helper (version: v0.9.14-43-g88ec36e-dirty)

HHC01415I Build date: Jul 25 2023 at 21:19:47

HHC01413I Hercules version 4.7.0.11007-SDL-DEV-gcddce71b

HHC01417I Built using Hercules-Helper (version: v0.9.14-43-g88ec36e-dirty)

HHC01415I Build date: Jul 25 2023 at 21:19:47

HHC01417I Built with: Apple Clang 14.0.3 (clang-1403.0.22.14.1)

HHC01417I Build type: Mac OS X aarch64 host architecture build

I highly recommend trying Hercules-Helper if you haven’t already. I tried building multiple versions of Hercules myself from source without success, despite having built Hercules 3.x successfully in the past. Once I grabbed Hercules-Helper, though, it not only automated the whole process but it built successfully on the first try with no need to tweak anything and the result seems to be rock-solid.

As a test, I left MTS running overnight, and I saw no crash.

--?
Ron Frederick
ronf@...

#167

开云体育

The 4.7 on the NetBSD system and the 4.5 on the CentoOS system were built with hercules-helper.

?

John

?

Show quoted text

From: [email protected] <[email protected]> On Behalf Of Ron Frederick
Sent: Wednesday, August 2, 2023 07:25
To: [email protected]
Subject: Re: [H390-MTS] Question about the availability of several things

?

On Aug 1, 2023, at 8:56 PM, John Palmer <jpalmer@...> wrote:

Have you tried SDL 4.x (any SDL version) with the standard d6.0a? I can get it to IPL on 4.5 on a CentOS Linux system, but it crashes anywhere from 10 mins to 2 hours. ?I tried changing the IODELAY on David Trout's suggestion on the NetBSD system with 4.6 and that didn't do any good.

The old 3.13 installed on NetBSD 9.3 seems to work ok.

?

For what it’s worth, I’m able to run MTS 6.0a without a problem on a Mac Studio (M1 Ultra) using the SDL version of Hercules 4.7.0 built using Hercules-Helper. Here’s some more version info:

?

HHC01413I Hercules version 4.7.0.11007-SDL-DEV-gcddce71b

HHC01417I Built using Hercules-Helper (version: v0.9.14-43-g88ec36e-dirty)

HHC01415I Build date: Jul 25 2023 at 21:19:47

HHC01413I Hercules version 4.7.0.11007-SDL-DEV-gcddce71b

HHC01417I Built using Hercules-Helper (version: v0.9.14-43-g88ec36e-dirty)

HHC01415I Build date: Jul 25 2023 at 21:19:47

HHC01417I Built with: Apple Clang 14.0.3 (clang-1403.0.22.14.1)

HHC01417I Build type: Mac OS X aarch64 host architecture build

?

I highly recommend trying Hercules-Helper if you haven’t already. I tried building multiple versions of Hercules myself from source without success, despite having built Hercules 3.x successfully in the past. Once I grabbed Hercules-Helper, though, it not only automated the whole process but it built successfully on the first try with no need to tweak anything and the result seems to be rock-solid.

?

As a test, I left MTS running overnight, and I saw no crash.

--?
Ron Frederick
ronf@...

?

#168

I'm using SDL Hercules built from GIT, but I haven't updated in a while. I'm not sure what version it corresponds to.  I'm running it on a Mac. I haven't tried plain vanilla D6 in this environment.

Mike

Show quoted text

On 1 Aug 2023, at 23:56, John Palmer wrote:

Just out of curiosity, to which version of Hercules are you adding HIM support and what system do you develop on?

Have you tried SDL 4.x (any SDL version) with the standard d6.0a? I can get it to IPL on 4.5 on a CentOS Linux system, but it crashes anywhere from 10 mins to 2 hours. I tried changing the IODELAY on David Trout's suggestion on the NetBSD system with 4.6 and that didn't do any good.

The old 3.13 installed on NetBSD 9.3 seems to work ok.