OPS/MVS - CICS Region restarting immediately after shutdown

Products

• OPS/MVS Event Management & Automation

Issue/Introduction

Created batch jobs to start and stop CICS regions (e.g. CICSA) using following command:

OPSQL UPDATE STCTBL SET DESIRED_STATE='DOWN' WHERE NAME='CICSA' 

Last night the stop jobs were successfully executed and shutdown the CICS regions.  Immediately after regions were down OPS/MVS restarted them.  Example region, CICSA, is defined to be in a schedule up Monday through Friday from 0000-2400.  When I looked at the status for the problem regions they appeared as follows.  I experimented with restarting the regions.  The simple solution turned out to be overriding FAILED with DOWN and the regions started successfully.

 

Cm Sta Resource Name      Current  Desired  Res Pre Ref Action   Message  

-- --- ------------------ -------- -------- --- --- --- -------- ----------

         W CICSA           FAILED   UP        A   A   A  ACTIVE   Transition

OPSLOG shows following in response to the only shutdown job submitted against CICSA.

21:19:19 OPS7914T SSM AUDIT: STCTBL.CICSA UPDATED by *MASTER* *DYNAMIC.ESSMEZ CURRENT_STATE=DOWN

21:20:02 OPS7914T SSM AUDIT: STCTBL.CICSA UPDATED by OPSOSF ASOSMAPI DESIRED_STATE=UP

21:20:02 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: DOWN_UP MVSCMD=S CICSA' 

21:20:02 OPS1181H OPSMAIN       (*Local*) MVS N/A imsID S CICSA'

21:20:02 S CICSA                                                           

21:20:02 OPS7914T SSM AUDIT: STCTBL.CICSA UPDATED by OPSMAIN STATEMAN CURRENT_STATE=STARTING

21:20:02 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STARTING_UP=NO ACTION FOUND

 

Following is the syslog listing of eye-catcher “STCTBL.CICSA”

 

21:18:42.94 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: UP_DOWN TSOCMD=OI SHUTCICS

21:18:42.94 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STOPPING_DOWN=NO ACTION

21:20:02.52 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: DOWN_UP MVSCMD=S CICSA,

21:20:02.53 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STARTING_UP=NO ACTION

21:28:45.19 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: UP_DOWN TSOCMD=OI SHUTCICS

21:28:45.20 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STOPPING_DOWN=NO ACTION

23:00:02.81 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: DOWN_UP MVSCMD=S CICSA,

23:00:02.81 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STARTING_UP=NO ACTION

23:59:53.83 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: UP_DOWN TSOCMD=OI SHUTCICS

23:59:53.84 OPS7902H STATEMAN ACTION FOR STCTBL.CICSA: STOPPING_DOWN=NO ACTION

Environment

Component : OPS/MVS

Cause

Two different entities were defined to control the same resource.   The first entity, the batch job, was attempting to control the resource to be down.   In direct conflict, the Schedule Manager was set to keep the same resource up.   We see this in the log where the Schedule Manager is identified as ASOSMAPI.   

 

Resolution

If the desire is to control the resource through a batch job, as originally indicated, then the resource should be removed from the Schedule Manager control.