I have one Autosys Job set-up scenario like below, to achieve the aim start/stop multiple servers level deployed application jobs on single click at environment level as well as server level
example:
Box 1 (Environment level like UAT ) >> Box2 (Server level like Host1 ) >> CMD1 (Application 1 deployed on Host1) & CMD2 (Application 1 deployed on Host1) jobs
BOX1 Job I have created at environment level Like UAT environment
This UAT environment has 4 servers which have multiple cmd jobs
So each server I am treating as Box2 and deployed applications as cmd Jobs
BOX Job 1:
insert_job: **150_OneTouch_START_UAT** job_type: box
description: UAT START BOX
owner: RG76WC
alarm_if_fail: n
alarm_if_terminated: n
date_conditions: y
run_window: "12:20-14:30"
days_of_week: su,mo,tu,we,th,fr,sa
start_times: "12:21"
send_notification: n
BOX Job2:
insert_job: **150_OneTouch_START_U1_Server** job_type: box
description: U1 Server App box
owner: RG76WC
box_name: **150_OneTouch_START_UAT **
alarm_if_fail: n
alarm_if_terminated: n
run_window: "12:21-14:29"
date_conditions: y
days_of_week: su,mo,tu,we,th,fr,sa
start_times: "12:22"
send_notification: n
CMD Job1 (This job supposed to run in every hours 4 times as mentioned here in config - start_mins: 3,18,33,48)
insert_job: 150_OneTouch_START_U1_Server_App_CDSEJ job_type: cmd
machine: sd-4567-fghj.nam.nsroot.net
owner: RG76WC
box_name: **150_OneTouch_START_U1_Server **
permission: me,mx,ge,gx,we,wx
alarm_if_fail: y
alarm_if_terminated: y
date_conditions: y
days_of_week: su,mo,tu,we,th,fr,sa
start_mins: 3,18,33,48
send_notification: n
std_out_file: >> /var/tmp/$AUTO_JOB_NAME.out
std_err_file: >> /var/tmp/$AUTO_JOB_NAME.err
command: /home/RG76WC/CDS2IodEtl/cds2config/startETL.sh
CMD Job2
insert_job: 150424_OneTouch_START_U1_Server_App_MQDataFeed job_type: cmd
machine: sd-4567-fghj.nam.nsroot.net
owner: RG76WC
box_name: **150_OneTouch_START_U1_Server**
permission: me,mx,ge,gx,we,wx
alarm_if_fail: y
alarm_if_terminated: y
date_conditions: y
days_of_week: su,mo,tu,we,th,fr
start_times: "12:25"
send_notification: n
std_out_file: >> /var/tmp/$AUTO_JOB_NAME.out
std_err_file: >> /var/tmp/$AUTO_JOB_NAME.err
command: /home/RG76WC/SourceMQDataApp/startSourceMQDataApp.sh
In the mentioned set-up , We want box to keep running until define run_window: "12:20-14:30" duration so that box should remain in running status during this time, and underlying cmd job1 should keep running for define regulars interval [start_mins: 3,18,33,48 ] which is 4 times in an hour
But with this set-up we are observing once cmd job1 and cmd job 2 getting executed at defined start_mins and turning to successful status for first time attempt, box 2 and box 1 job is also getting turn to success due to which we are not able to see cmd 1 job is running for next interval with-in defined run_window: "12:20-14:30" again.
How can I fulfill this business requirement? with this set-up we are turning underlying cmd job start to control at single click from both level environment and server level.