ca.portal.admin

[IDMS-L] XA Storage Type Terminal and Database

Discussion created by ca.portal.admin on May 19, 2005
Latest reply on May 20, 2005 by ca.portal.admin
Hi,



We have noticed something funny after going to 16.0 SP1 + apars over the
weekend dealing with XA storage type DB.



We have had our storage pools split since we moved to LE runtime a
number of years ago. We have an XA pool 131 for type DB/TR. We had set
this at 2500 and mostly seemed to use 1000 as a maximum most days. Since
we went to 16.0 on the weekend we have had to increase this to 3500 with
new highs around 3000 for each day. This figure is reached at the end of
day by batch CV jobs. We sit at an HWM of 520 for our online and then
this spikes during the evening.



Has anyone else noticed this behaviour running 16.0 SP1 under z/OS 1.4?



Thanks



Chris Wood

Alberta Department of Energy

CANADA



Note John Siraco has raised QI68061 to describe the output of DCMT D
SUBTASK NUM

This communication is intended for the use of the recipient to which it is
addressed, and may contain confidential, personal and or privileged
information. Please contact us immediately if you are not the intended
recipients of this communication, and do not copy, distribute, or take
action relying on it. Any communication received in error, or subsequent
reply, should be deleted or destroyed.

"
IDMS Public Discussion Forum
IDMS-L@LISTSERV.IUASSN.COM
SMTP
IDMS-L@LISTSERV.IUASSN.COM
IDMS-L@LISTSERV.IUASSN.COM
SMTP








Normal

Normal
XA Storage Type Terminal and Database
"Hi,



We have noticed something funny after going to 16.0 SP1 + apars over the
weekend dealing with XA storage type DB.



We have had our storage pools split since we moved to LE runtime a
number of years ago. We have an XA pool 131 for type DB/TR. We had set
this at 2500 and mostly seemed to use 1000 as a maximum most days. Since
we went to 16.0 on the weekend we have had to increase this to 3500 with
new highs around 3000 for each day. This figure is reached at the end of
day by batch CV jobs. We sit at an HWM of 520 for our online and then
this spikes during the evening.



Has anyone else noticed this behaviour running 16.0 SP1 under z/OS 1.4?



Thanks



Chris Wood

Alberta Department of Energy

CANADA



Note John Siraco has raised QI68061 to describe the output of DCMT D
SUBTASK NUM

This communication is intended for the use of the recipient to which it is addressed, and may contain confidential, personal and or privileged information. Please contact us immediately if you are not the intended recipients of this communication, and do not copy, distribute, or take action relying on it. Any communication received in error, or subsequent reply, should be deleted or destroyed.

"
IDMS Public Discussion Forum
IDMS-L@LISTSERV.IUASSN.COM
SMTP
IDMS-L@LISTSERV.IUASSN.COM
IDMS-L@LISTSERV.IUASSN.COM
SMTP








Normal

Normal
DCMT D SUBTASK NUM
"Hi,



John Siraco has sent me this docup info that will be issued soon. The
wording may change slightly to match the wording for DCMT D SUBTASK.
There is little likelihood of tuning information around the various
counters so just sue it as reference.



Add page 3.53.4.1 Usage to the Advantage CA-IDMS System Tasks and
Operator Commands







DCMT DISPLAY SUBTASK <name or number> Displays the following information
for the specified subtask:





+-----------------------------------------------------------------------
-+

| Field | Value
|

----------------------------------------------------------------------
-|

| Name | The name of each subtask
|

----------------------------------------------------------------------
-|

| Number | The number of each subtask
|

----------------------------------------------------------------------
-|

| Status | The current status of the subtask (IDLE or BUSY)
|

----------------------------------------------------------------------
-|

| Work Type | The type of work the subtask can execute.
|

| |
|

| |
________________________________

I or IDMS -- Database and communication work |

| |
________________________________

R or RRS -- RRS (Resource Recovery System) work. |

----------------------------------------------------------------------
-|

| Count wakeups | The number of times the operating system dispatched
|

| | this subtask
|

----------------------------------------------------------------------
-|

| Count task | The number of times this subtasks dispatched an idms
|

| dispatches | task
|

----------------------------------------------------------------------
-|

| User Mode CPU | The amount of time this subtask spent in user-mode
|

| Time | execution
|

----------------------------------------------------------------------
-|

| System Mode CPU | The amount of time this subtask spent in System-mode
|

| Time | execution
|

----------------------------------------------------------------------
-|

| Cpu | Percentage of CPU cycles this subtask received vs
the|

| effectiveness | the amount of CPU cycles it required to run
|

| (%) |
|

----------------------------------------------------------------------
-|

| Count times fast| The number of times the IDMS postexit was able to
|

| posted | post an ECB without requiring a call to the
operating|

| | system while running on this subtask
|

----------------------------------------------------------------------
-|

| Count times OS | The number of times the operating system's post SVC
|

| posted | was used to post an ECB. This field should always be
|

| | zero unless operating system PC routines are not
|

| | available
|

----------------------------------------------------------------------
-|

| Count found work| The number of times the IDMS dispatcher found work
|

| pass 1 | queued to the global dispatch work queue (CSAWKQUE)
|

| | while using this subtask
|

----------------------------------------------------------------------
-|

| Count found work| The number of times the IDMS dispatcher found work
|

| pass 2 | to dispatch while scanning the DCE active chain
|

| | using this subtask
|

----------------------------------------------------------------------
-|

| Count times | The number of times the operating system woke the
|

| POSTEXIT | IDMS system up via it's PCRESUME routine using
|

| resumed | this subtask
|

+-----------------------------------------------------------------------
-+



Chris Wood

Alberta Department of Energy

Calgary

This communication is intended for the use of the recipient to which it is addressed, and may contain confidential, personal and or privileged information. Please contact us immediately if you are not the intended recipients of this communication, and do not copy, distribute, or take action relying on it. Any communication received in error, or subsequent reply, should be deleted or destroyed.

"
IDMS Public Discussion Forum
IDMS-L@LISTSERV.IUASSN.COM
SMTP
IDMS-L@LISTSERV.IUASSN.COM
IDMS-L@LISTSERV.IUASSN.COM
SMTP








Normal

Normal
Re: XA Storage Type Terminal and Database
"Chris,

A further update after discussing this with colleagues. There is an
Supportconnect issue/solution 3124 which has this description


Title: SOS CONDITION IN DATABASE KEPT STORAGE POOLS

If an application BINDs and FINISHes very large numbers of run-units under
the control of a single task the Central Version may experience short on
storage conditions in pools defined for DATABASE KEPT storage. This is
caused
by the failure to release the storage for the back-end subschema-control
when a run-unit issues a FINISH.

The PTF was TC03902.

After increasing our pool 129 after going to 16.0 we still had the problem
later the following day and it was tracked back to a program doing the
above. So we are uncertain if there was an increase in the pool due to 16.0
or the program or both.

Regards

Sam

Outcomes