[hercules-390] random I/O error

[[e-mail] [hercules-390]]

While doing an automated build of
an MVS product on a slightly modified
Hercules 3.07, I encountered this:

13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

I've never seen this before, and
simply rerunning the exact same
automated job made the problem
go away.

Is it likely that this is in fact triggered
by a random I/O error in Hercules?

I tried trimming the above job down
so that it stops assembling after
MVSSUPA, so that I can get quicker
test results (about 12 seconds total,
including IPL) and then ran it 1000
times (automatically) and there were
no further S106 abends.

I am going to run thousands more
test runs to see if I can get the problem
to reappear. When I know the frequency
I will be in a position to test on other
Hercules versions.

Unless someone knows about this
problem already such that no further
test runs are required.

Thanks. Paul.

[Jeremy Nicoll [e-mail] [hercules-390]]

Post by ***@yahoo.com.au [hercules-390]
While doing an automated build of
an MVS product on a slightly modified
13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

Does your version of MVS have EREP?

[[e-mail] [hercules-390]]

Post by Jeremy Nicoll ***@letterboxes.org [hercules-390]

13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

Does your version of MVS have EREP?

I don't have a SYS1.LINKLIB(EREP) but I
do have some datasets starting with "EREP".

I've now done 2000 runs without error.

BFN. Paul.

[Jeremy Nicoll [e-mail] [hercules-390]]

Post by ***@yahoo.com.au [hercules-390]

Post by Jeremy Nicoll ***@letterboxes.org [hercules-390]

13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

Does your version of MVS have EREP?

I don't have a SYS1.LINKLIB(EREP)

Well, I'd not expect there to be. I think EREPs module(s) start "IFC",
certainly in more recent versions of MVS they do, in particular
EXEC PGM=IFCEREP1...

It's a program that dumps the error records from SYS1.LOGREC.

On a real system, you'd get things like hardware errors reported there.
I can
barely remember the details.

I can't remember if logging has to be turned on somehow.

[Steve Smith [e-mail] [hercules-390]]

Post by ***@yahoo.com.au [hercules-390]
While doing an automated build of
an MVS product on a slightly modified
13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000
I've never seen this before, and
simply rerunning the exact same
automated job made the problem
go away.
Thanks. Paul.

Surely you know the drill by now... post the complete output... or at least
the 3 LOG datasets from the top..

There are classical commonish reasons for a 106-F but I'm not about to go
through them on the offchance.

Steve


.

[[e-mail] [hercules-390]]

Hint:


help abend operand(S106-0F)
S106-0F - AN UNCORRECTABLE INPUT/OUTPUT ERROR OCCURRED, THE LOAD
MODULE HAD AN ERROR WHICH CAUSED THE CHANNEL PROGRAM TO FAIL,
OR A LINKLIST LIBRARY EXPANDED INTO ANOTHER EXTENT SINCE IPL.
CHECK REGISTER 0:
40 - I/O ERROR ON A PDS.
44 - ERROR ON A VIRTUAL DS.
48 - SEEK ADDRESS OUTSIDE EXTENT.
READY

Post by ***@yahoo.com.au [hercules-390]
While doing an automated build of
an MVS product on a slightly modified
13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000
I've never seen this before, and
simply rerunning the exact same
automated job made the problem
go away.
Thanks. Paul.

Surely you know the drill by now... post the complete output... or at least
the 3 LOG datasets from the top..

There are classical commonish reasons for a 106-F but I'm not about to go
through them on the offchance.

Steve


.

[Dave Kreiss [e-mail] [hercules-390]]

I have seen the S106-F abends quite frequently on my prior hardware running Windows Vista (that system is now dead). Also I've seen command rejects (CCW command was X'00') in IEBCOPY jobs. On my current Windows 10 system neither has never occurred.

It is my belief both errors are related to timing and PCI but just one of those gut feelings trying to research cause in source code. I suspected the underlying Windows OS was partially responsible.

Most often the S106-F occurred fetching an IFOX module. Reruns were always eventually successful.

The environment is many jobs (SMP installation) run on the Volker turnkey MVS 3.8 and is actually quite heavy with thousands of assemblies and many IEBCOPY invocations with run times of several hours. I've saw the 106-F and command rejects consistently with various Hercules 3 and 4 versions.

Dave

Post by ***@id.ethz.ch [hercules-390]
help abend operand(S106-0F)
S106-0F - AN UNCORRECTABLE INPUT/OUTPUT ERROR OCCURRED, THE LOAD
MODULE HAD AN ERROR WHICH CAUSED THE CHANNEL PROGRAM TO FAIL,
OR A LINKLIST LIBRARY EXPANDED INTO ANOTHER EXTENT SINCE IPL.
40 - I/O ERROR ON A PDS.
44 - ERROR ON A VIRTUAL DS.
48 - SEEK ADDRESS OUTSIDE EXTENT.
READY

Post by ***@yahoo.com.au [hercules-390]
While doing an automated build of
an MVS product on a slightly modified
13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000
I've never seen this before, and
simply rerunning the exact same
automated job made the problem
go away.
Thanks. Paul.

Surely you know the drill by now... post the complete output... or at least
the 3 LOG datasets from the top..
There are classical commonish reasons for a 106-F but I'm not about to go
through them on the offchance.
Steve
.

[[e-mail] [hercules-390]]

I have seen the S106-F abends quite frequently on my prior hardware running Windows Vista (that system is now dead). Also I've seen command rejects (CCW command was X'00') in IEBCOPY jobs. On my current Windows 10 system neither has ever occurred. 

Most often the S106-F occurred fetching an IFOX module. Reruns were always eventually successful though sometimes multiple runs were necessary.

What was being run is many jobs (SMP installation) and is actually quite heavy with thousands of assemblies and many IEBCOPY invocations with run times of several hours. The environment is the Volker turnkey MVS 3.8. I believe I've seen these errors on the tk4- as well though it has been some time since I ran on tk4-. I've seen the 106-F and command rejects consistently with various Hercules 3 and 4 versions.

So it is my belief both errors are related to timing and PCI but just one of those gut feelings after spending some time trying to research cause in MVS source code available on the Volker turnkey system.  I suspected the underlying Windows OS was partially responsible and not Hercules itself.

Dave

[[e-mail] [hercules-390]]

Hi Dave


that's an interesting coincidence, thanks a lot for reporting.


In this case I remove the ironic touch of my previous reply, which was only intended to remind Paul that even geniuses should follow some down to earth systematics when analyzing common problems...


Sorry about that!



Cheers
JÃŒrgen

---In hercules-***@yahoogroups.com, <***@...> wrote :


I have seen the S106-F abends quite frequently on my prior hardware running Windows Vista (that system is now dead). Also I've seen command rejects (CCW command was X'00') in IEBCOPY jobs. On my current Windows 10 system neither has ever occurred.

Most often the S106-F occurred fetching an IFOX module. Reruns were always eventually successful though sometimes multiple runs were necessary.

What was being run is many jobs (SMP installation) and is actually quite heavy with thousands of assemblies and many IEBCOPY invocations with run times of several hours. The environment is the Volker turnkey MVS 3.8. I believe I've seen these errors on the tk4- as well though it has been some time since I ran on tk4-. I've seen the 106-F and command rejects consistently with various Hercules 3 and 4 versions.

So it is my belief both errors are related to timing and PCI but just one of those gut feelings after spending some time trying to research cause in MVS source code available on the Volker turnkey system. I suspected the underlying Windows OS was partially responsible and not Hercules itself.

Dave

[[e-mail] [hercules-390]]

This post might be inappropriate. Click to display it.

[Jeremy Nicoll [e-mail] [hercules-390]]

Why does your log show Hercules shutting down at the exact
same time as your mysterious abend occurred?

Post by ***@yahoo.com.au [hercules-390]
23:43:31 13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
23:43:31 13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

and

Post by ***@yahoo.com.au [hercules-390]
23:43:31 HHCIN900I Begin Hercules shutdown
23:43:31 HHCIN901I Releasing configuration
23:43:31 HHCCP008I CPU0000 thread ended: tid=00002EB0, pid=15292
23:43:31 HHCCF047I Subchannel 0:0000 detached
23:43:31 HHCCF047I Subchannel 0:0001 detached
23:43:31 HHCCF047I Subchannel 0:0002 detached
23:43:31 HHCCF047I Subchannel 0:0003 detached
23:43:31 HHCCF047I Subchannel 0:0004 detached

...

[[e-mail] [hercules-390]]

Post by Jeremy Nicoll ***@letterboxes.org [hercules-390]
Why does your log show Hercules shutting down at the exact
same time as your mysterious abend occurred?

These are all automated scripts, and I
have removed all delays except for one
1 second delay (which I will hopefully
solve when I install TK4-), so everything
happens very fast.

If you look at the log, you will see all
of this:

23:43:31 13.43.31 JOB 1 $HASP150 PDPMVS ON PRINTER1 10,037 LINES

23:43:31 13.43.31 JOB 2 $HASP250 TERMHERC IS PURGED

23:43:31 stopall
23:43:31 quit

happening between the two displays you
showed below.

There is no reason to add delays, although
that stopall is only there to provide a delay
to avoid the "quit hang bug", but I still
managed to get a "quit hang bug" once in
5000 runs, so I will probably add another
"stopall" in due course.

And like I say - this job normally works and
allows me to do my real work. I can do
thousands (literally) of runs without error.

BFN. Paul.

Post by Jeremy Nicoll ***@letterboxes.org [hercules-390]
23:43:31 13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE
ACCESSED IFOX62
23:43:31 13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

and

Post by Jeremy Nicoll ***@letterboxes.org [hercules-390]
23:43:31 HHCIN900I Begin Hercules shutdown
23:43:31 HHCIN901I Releasing configuration
23:43:31 HHCCP008I CPU0000 thread ended: tid=00002EB0, pid=15292
23:43:31 HHCCF047I Subchannel 0:0000 detached
23:43:31 HHCCF047I Subchannel 0:0001 detached
23:43:31 HHCCF047I Subchannel 0:0002 detached
23:43:31 HHCCF047I Subchannel 0:0003 detached
23:43:31 HHCCF047I Subchannel 0:0004 detached

...

[Steve Smith [e-mail] [hercules-390]]

Post by Steve Smith ***@yahoo.com [hercules-390]

13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000

Surely you know the drill by now... post the complete output... or at
least the 3 LOG datasets from the top..

3 logs? I only have the hercules log file. There was no sysabend dd (I'm
not expecting to debug the assembler) so I didn't capture that. Full log
below.

When you run a batch job there are 3 'LOG' datsets produced. I can't
remember their names but it's something like SYSLOG, JESJCL and SYSMSG. It's
the messages and JCL that the job produces.

If there is a problem with your job, it's what third parties use to find out
what you are trying to do with your job and so to have a slight chance of
being able to help with your problem. Any answer anyone provides to the
details you have given is a guess.

Note to Juergen - it was someone else who suggested that this was a
"longstanding I/O problem with Hercules", so "proper protocol" for me was
to see what the group thought about that claim. I'm not the genius who
made the claim.
Note to those mentioning LPA expansions. My automated runs are simple
assemblies, if anyone is expanding LPA or whatever, it's happening
without my knowledge. And my automated runs involve doing a fresh IPL, so
that's not an issue either. Also note that a successful assembly
immediately preceded the failing one, although the failing one is
different in that it produces a listing.

I don't recall anyone mentioning LPA expansions, someone mentioned F
LLA,REFRESH and LNKLST expansions because that is the classical cause of
S106-F abends; You are writing to a LNKLST datset and causing additional
extents to be allocated; subsequently you are reading from the same LNKLST
datset and the LNKLST doesn't know about the extension.

I have now done 5000 runs without having a repeat of this problem.

Steve

.

[[e-mail] [hercules-390]]

Post by Steve Smith ***@yahoo.com [hercules-390]
When you run a batch job there are 3 'LOG' datsets produced. I can't
remember their names but it's something like SYSLOG, JESJCL and SYSMSG. It's
the messages and JCL that the job produces.

Oh, the printer output. That's 1 MB in
size so I've trimmed it down to some
somewhat-relevant things below.

Post by Steve Smith ***@yahoo.com [hercules-390]
S106-F abends; You are writing to a LNKLST datset and causing additional
extents to be allocated; subsequently you are reading from the same LNKLST
datset and the LNKLST doesn't know about the extension.

Ok, but as far as I know, the IFOX modules
are being used, not rewritten and extended.
Nothing should be changing SYS1.LINKLIB.

I guess if I can reproduce the problem I
can try putting it into a steplib that no-one
besides me knows about, and ensure
that there is only one extent too.

BFN. Paul.




J E S 2 J O B L O G


13.43.26 JOB 1 $HASP373 PDPMVS STARTED - INIT 3 - CLASS C - SYS BSP1
13.43.26 JOB 1 IEFACTRT - Stepname Procstep Program Retcode
13.43.26 JOB 1 PDPMVS CREATE IEFBR14 RC= 0
13.43.26 JOB 1 PDPMVS PDPTOP IEBGENER RC= 0
13.43.26 JOB 1 PDPMVS PDPMAIN IEBGENER RC= 0
13.43.27 JOB 1 PDPMVS PDPPRLG IEBGENER RC= 0
13.43.27 JOB 1 PDPMVS PDPEPIL IEBGENER RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART ASM ASMA90 RC= 0
13.43.27 JOB 1 PDPMVS MVSSTART LKED IEWL RC 4
13.43.31 JOB 1 IEA703I 106- F PDPMVS ASM MODULE ACCESSED IFOX62
13.43.31 JOB 1 IEF450I PDPMVS ASM MVSSUPA - ABEND S106 U0000
13.43.31 JOB 1 PDPMVS MVSSUPA ASM ASMA90 AB S106
13.43.31 JOB 1 PDPMVS MVSSUPA LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS START ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS START LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS STDIO ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS STDIO LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS STDLIB ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS STDLIB LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS CTYPE ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS CTYPE LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS STRING ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS STRING LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS TIME ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS TIME LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS ERRNO ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS ERRNO LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS ASSERT ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS ASSERT LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS LOCALE ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS LOCALE LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS MATH ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS MATH LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS SETJMP ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS SETJMP LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS SIGNAL ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS SIGNAL LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS @@MEMMGR ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS @@MEMMGR LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS PDPTEST ASM ASMA90 *FLUSH*
13.43.31 JOB 1 PDPMVS PDPTEST LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS LKED IEWL *FLUSH*
13.43.31 JOB 1 PDPMVS PDPTEST PDPTEST *FLUSH*
13.43.31 JOB 1 PDPMVS DELETE IEFBR14 RC= 0
13.43.31 JOB 1 $HASP395 PDPMVS ENDED


------ JES2 JOB STATISTICS ------


30 SEP 17 JOB EXECUTION DATE


22,221 CARDS READ


10,037 SYSOUT PRINT RECORDS


0 SYSOUT PUNCH RECORDS


0.07 MINUTES EXECUTION TIME

1 //PDPMVS JOB CLASS=C,REGION=0K JOB 1
***
//PDPASM PROC MEMBER='',LIST='NOLIST'
//ASM EXEC PGM=ASMA90,
// PARM='DECK,&LIST,TERM'
//SYSLIB DD DSN=SYS1.MACLIB,DISP=SHR,DCB=BLKSIZE=32720
// DD DSN=&&MACLIB,DISP=(OLD,PASS)
// DD DSN=SYS1.MODGEN,DISP=SHR
//SYSUT1 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
//SYSUT2 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
//SYSUT3 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
//SYSTERM DD SYSOUT=*
//SYSPRINT DD SYSOUT=*
//SYSLIN DD DUMMY
//SYSGO DD DUMMY
//SYSPUNCH DD DSN=&&OBJSET,UNIT=SYSALLDA,SPACE=(80,(9000,9000)),
// DISP=(,PASS)
//*
//LKED EXEC PGM=IEWL,PARM='NCAL',
// COND=(0,LT,ASM)
//SYSLIN DD DSN=&&OBJSET,DISP=(OLD,DELETE)
//SYSLMOD DD DSN=&&NCALIB(&MEMBER),DISP=(OLD,PASS)
//SYSUT1 DD UNIT=SYSALLDA,SPACE=(CYL,(2,1))
//SYSPRINT DD SYSOUT=*
// PEND
2 //CREATE EXEC PGM=IEFBR14
3 //DD12 DD DSN=&&NCALIB,DISP=(,PASS),
// DCB=(RECFM=U,LRECL=0,BLKSIZE=6144),
// SPACE=(CYL,(10,10,20)),UNIT=SYSALLDA
4 //DD13 DD DSN=&&LOADLIB,DISP=(,PASS),
// DCB=(RECFM=U,LRECL=0,BLKSIZE=6144),
// SPACE=(CYL,(10,10,20)),UNIT=SYSALLDA
5 //DD14 DD DSN=&&MACLIB,DISP=(,PASS),
// DCB=(RECFM=FB,LRECL=80,BLKSIZE=6160),
// SPACE=(CYL,(10,10,20)),UNIT=SYSALLDA
***
6 //PDPTOP EXEC PGM=IEBGENER
7 //SYSUT2 DD DSN=&&MACLIB(PDPTOP),DISP=(OLD,PASS)
8 //SYSUT1 DD *
9 //SYSPRINT DD SYSOUT=*
10 //SYSIN DD DUMMY
***
11 //PDPMAIN EXEC PGM=IEBGENER
12 //SYSUT2 DD DSN=&&MACLIB(PDPMAIN),DISP=(OLD,PASS)
13 //SYSUT1 DD *
14 //SYSPRINT DD SYSOUT=*
15 //SYSIN DD DUMMY
***
16 //PDPPRLG EXEC PGM=IEBGENER
17 //SYSUT2 DD DSN=&&MACLIB(PDPPRLG),DISP=(OLD,PASS)
18 //SYSUT1 DD *
19 //SYSPRINT DD SYSOUT=*
20 //SYSIN DD DUMMY
***
21 //PDPEPIL EXEC PGM=IEBGENER
22 //SYSUT2 DD DSN=&&MACLIB(PDPEPIL),DISP=(OLD,PASS)
23 //SYSUT1 DD *
24 //SYSPRINT DD SYSOUT=*
25 //SYSIN DD DUMMY
***
26 //MVSSTART EXEC PDPASM,MEMBER=MVSSTART
27 ++PDPASM PROC MEMBER='',LIST='NOLIST'
28 ++ASM EXEC PGM=ASMA90,
++ PARM='DECK,&LIST,TERM'
29 ++SYSLIB DD DSN=SYS1.MACLIB,DISP=SHR,DCB=BLKSIZE=32720
30 ++ DD DSN=&&MACLIB,DISP=(OLD,PASS)
31 ++ DD DSN=SYS1.MODGEN,DISP=SHR
32 ++SYSUT1 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
33 ++SYSUT2 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
34 ++SYSUT3 DD UNIT=SYSALLDA,SPACE=(CYL,(20,10))
35 ++SYSTERM DD SYSOUT=*
36 ++SYSPRINT DD SYSOUT=*
37 ++SYSLIN DD DUMMY
38 ++SYSGO DD DUMMY
39 ++SYSPUNCH DD DSN=&&OBJSET,UNIT=SYSALLDA,SPACE=(80,(9000,9000)),
++ DISP=(,PASS)


...


IEF236I ALLOC. FOR PDPMVS CREATE
IEF237I 2A0 ALLOCATED TO DD12
IEF237I 1B2 ALLOCATED TO DD13
IEF237I 1B2 ALLOCATED TO DD14
IEF142I PDPMVS CREATE - STEP WAS EXECUTED - COND CODE 0000
IEF285I SYS17273.T134326.RA000.PDPMVS.NCALIB PASSED *--------0
IEF285I VOL SER NOS= PUB002.
IEF285I SYS17273.T134326.RA000.PDPMVS.LOADLIB PASSED *--------0
IEF285I VOL SER NOS= SEASIK.
IEF285I SYS17273.T134326.RA000.PDPMVS.MACLIB PASSED *--------0
IEF285I VOL SER NOS= SEASIK.
IEF373I STEP /CREATE / START 17273.1343
IEF374I STEP /CREATE / STOP 17273.1343 CPU 0MIN 00.00SEC SRB 0MIN 00.00SEC VIRT 4K SYS 316K
IEF236I ALLOC. FOR PDPMVS PDPTOP
IEF237I 1B2 ALLOCATED TO SYSUT2
IEF237I JES2 ALLOCATED TO SYSUT1
IEF237I JES2 ALLOCATED TO SYSPRINT
IEF237I DMY ALLOCATED TO SYSIN
IEF142I PDPMVS PDPTOP - STEP WAS EXECUTED - COND CODE 0000
IEF285I SYS17273.T134326.RA000.PDPMVS.MACLIB PASSED *--------4
IEF285I VOL SER NOS= SEASIK.
IEF285I JES2.JOB00001.SI0101 SYSIN
IEF285I JES2.JOB00001.SO0123 SYSOUT
IEF373I STEP /PDPTOP / START 17273.1343
IEF374I STEP /PDPTOP / STOP 17273.1343 CPU 0MIN 00.02SEC SRB 0MIN 00.00SEC VIRT 48K SYS 336K
************************************************************************************************************************************
* 2. Jobstep of job: PDPMVS Stepname: PDPTOP Program name: IEBGENER Executed on 30.09.17 from 13.43.26 to 13.43.26 *
* elapsed time 00:00:00,02 CPU-Identifier: BSP1 Page-in: 0 *


...


IEF373I STEP /ASM / START 17273.1343
IEF374I STEP /ASM / STOP 17273.1343 CPU 0MIN 02.85SEC SRB 0MIN 00.29SEC VIRT 2120K SYS 352K
************************************************************************************************************************************
* 8. Jobstep of job: PDPMVS Stepname: ASM Program name: ASMA90 Executed on 30.09.17 from 13.43.27 to 13.43.31 *
* elapsed time 00:00:03,57 CPU-Identifier: BSP1 Page-in: 0 *
* CPU time 00:00:03,14 Virtual Storage used: 2120K Page-out: 0 *
* I/O Operation *
* Number of records read via DD * or DD DATA: 4281 *
* 148.....301 1B2......18 248......50 2A0....2968 2A0.....183 2A0......99 DMY.......0 DMY.......0 DMY.......0 DMY.......0 *
* 1B2.....258 DMY.......0 *
* *
* Charge for step (w/o SYSOUT): 5,23 *
************************************************************************************************************************************
IEF272I PDPMVS LKED MVSSUPA - STEP WAS NOT EXECUTED.
IEF373I STEP /LKED / START 17273.1343
IEF374I STEP /LKED / STOP 17273.1343 CPU 0MIN 00.00SEC SRB 0MIN 00.00SEC VIRT 0K SYS 0K
IEF272I PDPMVS ASM START - STEP WAS NOT EXECUTED.


...

****MVSSTART DOES NOT EXIST BUT HAS BEEN ADDED TO DATA SET
AUTHORIZATION CODE IS 0.




DIAGNOSTIC MESSAGE DIRECTORY
IEW0461 WARNING - SYMBOL PRINTED IS AN UNRESOLVED EXTERNAL REFERENCE; NCAL WAS SPECIFIED, OR THE REFERENCE WAS
MARKED FOR RESTRICTED NO-CALL OR NEVERCALL.

MVSSUPA EXTERNAL SYMBOL DICTIONARY PAGE 1


SYMBOL TYPE ID ADDR LENGTH LDID ASM 0201 13.43 09/30/17

MVSSUPA SD 0001 000000 003319
@@BUGF LD 000938 0001
@@AOPEN LD 000000 0001
@@ALINE LD 00158C 0001
@@AREAD LD 001628 0001
@@AWRITE LD 001DF4 0001
@@ANOTE LD 002236 0001


...


@@SETM31 LD 003312 0001
MVSSUPA M V S S U P A *** MVS VERSION OF PDP CLIB SUPPORT PAGE 2


LOC OBJECT CODE ADDR1 ADDR2 STMT SOURCE STATEMENT ASM 0201 13.43 09/30/17

2 ***********************************************************************
3 * Updated 2017-09-21 *
4 * *
5 * This program written by Paul Edwards. *
6 * Released to the public domain *
7 * *
8 * Extensively modified by others *
9 * *
10 ***********************************************************************
11 * *
12 * MVSSUPA - Support routines for PDPCLIB under MVS *


...


0000E 14167 R14 EQU 14 NO STANDARD REGEQU MACRO GP15019
0000F 14168 R15 EQU 15 NO STANDARD REGEQU MACRO GP15019
14169 END ,
MVSSUPA RELOCATION DICTIONARY PAGE 140


POS.ID REL.ID FLAGS ADDRESS ASM 0201 13.43 09/30/17

0001 0001 0C 0008FC
0001 0001 0C 000900
0001 0001 0C 000930
0001 0001 0C 000934


...



=X'40' 00001 000030E4 05430 05413
=X'80000000'
00004 00003280 05667 05487
=A(@@BUGF)
00004 00003284 05668 05592
PPPPPPPPPPP DDDDDDDDD PPPPPPPPPPP MM MM VV VV SSSSSSSSSS
PPPPPPPPPPPP DDDDDDDDDD PPPPPPPPPPPP MMM MMM VV VV SSSSSSSSSSSS
PP PP DD DD PP PP MMMM MMMM VV VV SS SS


(end of job here)

[[e-mail] [hercules-390]]

The most common reason for a 106-F abend is that one of your link list datasets has been extended when it ran out of space. If you have a system that has LLA, the LLA region keeps open copies of all the link list datasets. If a new extent is added to one of them, the next time you try to load a module from that PDS, you may get a 106-F. If you don't have LLA, the link list in memory is built at IPL time and I don't know a way
to refresh it without an IPL.

The solution is simple. If you have LLA, issue the following command at the console:

F LLA,REFRESH

If you don't have LLA, just IPL the system. The in memory link list will be rebuilt including the new extent.

Roy Frederick

[[e-mail] [hercules-390]]

While LLA not being refreshed on newer MVS systems can cause these errors there is no LLA involved with the free turnkey systems I have used. But there is the DYNABLDL STC on these systems. One of the first things I did was turn off DYNABLDL but that had no noticeable effect - that is the S106-F abends continued. At one time I located the code where this abend occurred and as I remember it was related to handling PCI during program fetch. PCI is a facility available in channel programming used to dynamically modify a running channel program. I suspect the channel program (as executed by the Hercules emulator) out ran the MVS software instruction stream such that the updates to the channel program were to late.

As far as the suggestion about examining LOGREC I don't recall finding anything there and I am sure I checked.

Dave