I doubt anyone except z/VM Development uses S/A Dump procedures often - if
even they do.
Since I seldom (read: have NEVER had to use S/A Dump since HPO days) use
S/A Dump procedures, I'm not likely to remember how when I need to (a time
which should certainly classify as a crisis),, nor explain to Operators
how to do so.
Knowing at least this one of a litany of my failures, I wrote the
following EXEC, which is stored on the OPERATOR 191 disk of **both** our
z/VM systems. If one is not responding, then the Operator can issue
"SADUMP ?" on the other system to get an explanation. It is supplied with
the usual warranties, expressed or implied: none
Anyone wanting to use it will obviously need to change it for their
site(s), but it's a good starting point and may even help provide the "leg
up" when you need it.
Mike Walter
Hewitt Associates
Any opinions expressed herein are mine alone and do not necessarily
represent the opinions or policies of Hewitt Associates.
/* Prolog; See Epilog for additional information ********************
* Exec Name - SADUMP EXEC *
* Unit Support - OSS/VM *
* Status - Version 1, Release 1.0 *
********************************************************************/
address 'COMMAND'
parse source xos xct xfn xft xfm xcmd xenvir .
parse upper arg parms 1 operands '(' options ')' parmrest
Signal ON Syntax
Signal ON NoValue
/* Signal ON ERROR */
parse value diag(08,'QUERY USERID') with ,
self . ConfigSysID . '15'x .
DispSys=ConfigSysID /* Defaults 'R' us */
wkops=operands
Do while wkops<>''
parse var wkops op wkops
parse var op L6op 7 .
Select
When op='?' then nop
When L6op='HALINV' then DispSys=op
Otherwise
Do
say '-------------------------'
say xfn'; Unsupported operand: "'op'".'
say '-------------------------'
say
End
End
End
Signal Explain
Call Exit rc
/********************************************************************/
/* Sub-Routines below this point */
/********************************************************************/
Exit:
parse arg exitrc .
If verify(exitrc,'-0123456789')=0 then Exit exitrc
else Exit 999999
Return /* Bracket subrtn only, never get here */
Explain:
Select
When DispSys='HALINVA1' then
'PIPE (NAME Explain)' ,
'| <' xfn xft xfm ,
'| INSIDE /ExplainBeginHALINVA1/ /ExplainEndHALINVA1/' ,
'| CONSOLE'
When DispSys='HALINVB1' then
'PIPE (NAME Explain)' ,
'| <' xfn xft xfm ,
'| INSIDE /ExplainBeginHALINVB1/ /ExplainEndHALINVB1/' ,
'| CONSOLE'
Otherwise
'PIPE (NAME Explain)' ,
'| <' xfn xft xfm ,
'| INSIDE /ExplainBeginUnknown/ /ExplainEndUnknown/' ,
'| CONSOLE'
End /* Select */
Call Exit 0
/*
ExplainBeginHALINVA1:
S/A Dump procedures for: VMCMS system (HALINVA1)
There is no SADUMP (StandAlone Dump) command for VM. If the system is
responding, then from OPERATOR you may use the command:
SNAPDUMP
You may see more information about the SNAPDUMP command by entering:
HELP CP SNAPDUMP
If the system is not responding (then you probably cannot read this),
a Stand Alone Dump my be performed by executing the following procedure.
If commands are not being accepted at the OPERATOR console in 4OD:
a problem with the TCPIP stack may be locking up the consoles in the 4OD
campus. Have someone in the 98-3S Command Center go to the "wall" where
the z/VM Emulator Session for VMCMS runs, go to a VMCMS "z/VM" logo
session (or start one), and try to enter the command:
LOGON OPERATOR HERE
If that works, and you still need a VMCMS StandAlone DUMP (perhaps you
only need to FORCE and XAUTOLOG TCPIP ?) you can enter: SNAPDUMP
-----
ONLY IF THE SNAPDUMP COMMAND CANNOT BE ENTERED should you continue with
the remainder of these instructions...
Taking a RESTART DUMP of the z/VM system (usually sufficient).
---------------------------------------- ====================
A quick, easier alternative to a VM stand-alone dump is a RESTART DUMP.
This is performed from the HMC, by clicking on the CPC4LP2 icon, and
then on the 'RESTART" icon. This will cause an SVC002 ABEND, followed
by an automatic IPL of the VM system.
Taking a Stand Alone Dump of the z/VM system.
---------------------------------------------
a) Ensure that the MagStar tape drive 04F0 (98 building) is not in use
by any system.
b) Ensure that tape drive 04F0 is "OFFLINE" on all z/VM and z/OS systems
Use the command on ALL OTHER z/VM systems: CP VARY OFF 4F0
and the command on any z/OS system: RO *all,V 4F0,OFFLINE
c) Mount "VM SADUMP nn" 3590 cartridge on 04F0 and make it READY.
(There may be one or more scratch tapes marked "VM SADUMP nn" on the
edge in the Tape Operations Center on 98-1S, in the rack labeled
"VM Software Tapes" at floor coordinates B-25).
d) From the HMC on the CPC4LP2 processor, perform the following
procedures taken from the "z/VM: CP Planning and Administration"
manual:
1. Click on the icon: "CPC4 CP4LP2 HALINVA1"
2. Click on the icon: "STOP"
2. Display locations X'0' to X'18' at the HMC console. The stand-alone
dump IPL sequence overlays these bytes, so they cannot be
recovered.
3. Click on the icon: "LOAD"
a. Checkmark the Load Type box: "Normal"
b. Checkmark the box: "Store Status"
c: Fill in the Load Address box: 0928 (which is volser VMPG01)
4. Ensure that a "VM SADUMP nn" tape is loaded and READY on 04F0.
5. When ready to begin the stand-alone dump, click on: "OK"
6. When the stand-alone dump is complete, the CPU will enter a
"WAIT STATE" with a PSW ending with: 8200
If the stand-alone dump utility is unsuccessful,
a. Correct the error
b. Invoke the HMC Hardware "RESTART" to restart the stand-alone
dump
ExplainEndHALINVA1:
*/
/*
ExplainBeginHALINVB1:
S/A Dump procedures for: VMSTE system (HALINVB1)
There is no SADUMP (StandAlone Dump) command for VM. If the system is
responding, then from OPERATOR you may use the command:
SNAPDUMP
You may see more information about the SNAPDUMP command by entering:
HELP CP SNAPDUMP
If the system is not responding (then you probably cannot read this),
a Stand Alone Dump my be performed by executing the following procedure.
If commands are not being accepted at the OPERATOR console in 4OD:
a problem with the TCPIP stack may be locking up the consoles in the 4OD
campus. Have someone in the 98-3S Command Center go to the "wall" where
the z/VM Emulator Session for SYSG runs, go to a VMSTE "z/VM" logo
session (or start one), and try to enter the command:
LOGON OPERATOR HERE
If that works, and you still need a VMSTE StandAlone DUMP (perhaps you
only need to FORCE and XAUTOLOG TCPIP ?) you can enter: SNAPDUMP
-----
ONLY IF THE SNAPDUMP COMMAND CANNOT BE ENTERED should you continue with
the remainder of these instructions...
Taking a RESTART DUMP of the z/VM system.
-----------------------------------------
A quick, easier alternative to a VM stand-alone dump is a RESTART DUMP.
This is performed from the HMC, by clicking on the CPC8LP2 icon, and
then on the 'RESTART" icon. This will cause an SVC002 ABEND, followed
by an automatic IPL of the VM system.
Taking a Stand Alone Dump of the z/VM system.
--------------------------------------------
a) Ensure that the MagStar tape drive 04F0 (98 building) is not in use
by any system.
b) Ensure that tape drive 04F0 is "OFFLINE" on all z/VM and z/OS systems
Use the command on ALL OTHER z/VM systems: CP VARY OFF 4F0
and the command on any z/OS system: RO *all,V 4F0,OFFLINE
c) Mount "VM SADUMP nn" 3590 cartridge on 04F0 and make it READY.
(There may be one or more scratch tapes marked "VM SADUMP nn" on the
edge in the Tape Operations Center on 98-1S, in the rack labeled
"VM Software Tapes" at floor coordinates B-25).
d) From the HMC on the CPC8LP2 processor, perform the following
procedures taken from the "z/VM: CP Planning and Administration"
manual:
1. Click on the icon: "CPC8 CP8LP2 HALINVB1"
2. Click on the icon: "STOP"
2. Display locations X'0' to X'18' at the HMC console. The stand-alon
dump IPL sequence overlays these bytes, so they cannot be
recovered.
3. Click on the icon: "LOAD"
a. Checkmark the Load Type box: "Normal"
b. Checkmark the box: "Store Status"
c: Fill in the Load Address box: 5802 (which is volser VSPG01)
4. Ensure that a "VM SADUMP nn" tape is loaded and READY on 04F0.
5. When ready to begin the stand-alone dump, click on: "OK"
6. When the stand-alone dump is complete, the CPU will enter a
"WAIT STATE" with a PSW ending with: 8200
If the stand-alone dump utility is unsuccessful,
a. Correct the error
b. Invoke the HMC Hardware "RESTART" to restart the stand-alone
dump
ExplainEndHALINVB1:
*/
/*
ExplainBeginUnknown:
General S/A Dump procedures for z/VM systems.
There is no SADUMP (StandAlone Dump) command for VM. If the system is
responding, then from OPERATOR you may use the command:
SNAPDUMP
You may see more information about the SNAPDUMP command by entering:
HELP CP SNAPDUMP
If the system is not responding (then you probably cannot read this),
a Stand Alone Dump my be performed by executing the following procedure.
If commands are not being accepted at the OPERATOR console in 4OD:
a problem with the TCPIP stack may be locking up remote consoles.
Have someone at the data center running the z/VM system go to
the z/VM Emulator Session for that z/VM, get to a "z/VM" logo
session (or start one), and try to enter the command:
LOGON OPERATOR HERE
If that works, and you still need a VMCMS StandAlone DUMP (perhaps you
only need to FORCE and XAUTOLOG TCPIP ?) you can enter: SNAPDUMP
-----
ONLY IF THE SNAPDUMP COMMAND CANNOT BE ENTERED should you continue with
the remainder of these instructions...
Taking a RESTART DUMP of the z/VM system (usually sufficient).
---------------------------------------- ====================
A quick, easier alternative to a VM stand-alone dump is a RESTART DUMP.
This is performed from the HMC, by clicking on the CPCnLPn icon, and
then on the 'RESTART" icon. This will cause an SVC002 ABEND, followed
by an automatic IPL of the VM system.
Taking a Stand Alone Dump of the z/VM system.
---------------------------------------------
a) Ensure that the MagStar tape drive 04nn s not in use by any sustem.
b) Ensure that tape drive 04nn is "OFFLINE" on all z/VM and z/OS systems
Use the command on ALL OTHER z/VM systems: CP VARY OFF 4nn
and the command on any z/OS system: RO *all,V 4nn,OFFLINE
c) Mount "VM SADUMP nn" 3590 cartridge on 04nn and make it READY.
(There may be one or more scratch tapes marked "VM SADUMP nn" on the
edge in the Tape Operations Center on 98-1S, in the rack labeled
"VM Software Tapes" at floor coordinates B-25).
d) From the HMC on the CPCnLPn processor, perform the following
procedures taken from the "z/VM: CP Planning and Administration"
manual:
1. Click on the icon: "CPCn CPnLPn HALINVA1"
2. Click on the icon: "STOP"
2. Display locations X'0' to X'18' at the HMC console. The stand-alone
dump IPL sequence overlays these bytes, so they cannot be
recovered.
3. Click on the icon: "LOAD"
a. Checkmark the Load Type box: "Normal"
b. Checkmark the box: "Store Status"
c: Fill in the Load Address box: 0928 (which is volser VMPG01)
4. Ensure that a "VM SADUMP nn" tape is loaded and READY on 04nn.
5. When ready to begin the stand-alone dump, click on: "OK"
6. When the stand-alone dump is complete, the CPU will enter a
"WAIT STATE" with a PSW ending with: 8200
If the stand-alone dump utility is unsuccessful,
a. Correct the error
b. Invoke the HMC Hardware "RESTART" to restart the stand-alone
dump
ExplainEndUnknown:
*/
Error:
say '+++ "ERROR" error routine entered in:' xfn xft xfm', rc='rc
say '+++ from line:' sigl', which reads:'
say '+++'sourceline(sigl)
Call Exit 20
Syntax:
say '+++ "SYNTAX" error routine entered in:' xfn xft xfm', rc='rc
say '+++ from line:' sigl', which reads:'
say '+++'sourceline(sigl)
Call Exit 20
NoValue:
say '+++ "NoValue" error routine entered in:' xfn xft xfm', rc='rc
say '+++ from line:' sigl', which reads:'
say '+++'sourceline(sigl)
say '+++ Variable with no value is:' condition('Description')
Call Exit 24
/* Epilog ***********************************************************
* Function - Tutorial *
* Component of - z/VM Sysprog's Toolbox. *
* +-?-+ *
* Command format- >-+------+-SADUMP-+---+-+-------------+------->< *
* +-EXEC-+ +-ConfigSysID-+ *
* Called by - *
* Dependencies - VM/SP 3 or higher. *
* - e.g. LISTFILE, COPYFILE, etc. *
* Program Lang. - CMS REXX *
* Date Written - 20050620 *
* Author - Michael R. Walter *
* Changed | By | Description of Change *
* --------+-----+------------------------------------------------- *
* 20060808 mrw -Update with latest addresses. *
* *
********************************************************************/
"Alan Ackerman" <[EMAIL PROTECTED]>
Sent by: "The IBM z/VM Operating System" <IBMVM@LISTSERV.UARK.EDU>
08/08/2006 11:56 AM
Please respond to
"The IBM z/VM Operating System" <IBMVM@LISTSERV.UARK.EDU>
To
IBMVM@LISTSERV.UARK.EDU
cc
Subject
Stand Alone Dump to Disk
We were asked to update our stand-alone dump procedures. We don't use
stand-alone dump very often, and the list of tape drives to use had gotte
n
badly out of date.
We discovered that z/OS can take a stand-alone dump to disk. We get more
involved in DASD address changes than in tape address changes (we have to
move the data), so we would find it easier to keep a list of DASD
addresses up-to-date. We run "lights-out" data centers, so operations
would prefer not to have to mount a tape half-way across the country.
Is there some way to take a stand-alone dump to DASD for VM? If not, woul
d
this be worth a SHARE requirement?
z/OS apparantly requires stand-alone dumps more often that VM. When a z/O
S
system fails, it "varies itself out of the plex" -- at which point our
operations is instructed to take a stand-alone dump.
I'm not sure when we really need a VM stand-alone dump. Usually a PSW
restaret will get us a dump. How often do you use stand-alone dumps?
The information contained in this e-mail and any accompanying documents
may contain information that is confidential or otherwise protected
from disclosure. If you are not the intended recipient of this message,
or if this message has been addressed to you in error, please
immediately alert the sender by reply e-mail and then delete this message,
including any attachments. Any dissemination, distribution or other use of
the contents of this message by anyone other than the intended recipient
is strictly prohibited.
