Automic Workload Automation

Dear All,

We got a couple of UC4 workflows started in recurring way. For example they are schduled to run every day for every 30 mins. In Activity Window it is showing as C_PERIOD type. 

I noticed that sometimes job is not starting. It looks like it stuck. I have difficulties to catch this anomalies. I wonder if it possible to setup an alert to send notification if specific job didn't start for example 2 or 3 last hours. In this case UC4 should send email to specific recipients. 

Can You help on this ? 

Regards, Krystian

You can add an SLO and this can monitor the job.   If it does not start, you set an alert.

Hi Richard, Thanks for Your response. What does it mean SLO ? Can You give more light on this ? 

Regards, Krystian

Hi KrystianKochanek629449,

Richard is referring to to the Service Level Objective, see the below link for more information.
Link 1 - Service Level Objective (SLO) 

Link 2 - Service Level Objective (SLO) 

hi LuuLe, Thanks for Your help. I had no idea about SLO. 

My job that I need to monitor is running in recurring way every 30 mins. I see SLO have an option to check 'Latest End Time'. Anyway I don't know how to setup an action when job stuck and it is not running last 1 or 2 hours. It finished recently correctly and should run next time in next 30 minutes but it stuck .. how to catch this setup and SLO to handle this ? 

Can You help on this ? 

Regards, Krystian

Hi

Just a general hint - prefer an EVNT.TIME or a JSCH object to start recurring jobs than start it in recurrent mode.

If you like scripting you can use a watchdog script to check the executions of your recurring script.

I posted one or 2 years ago a watchdog script in automic community (but at the moment I am unable to find it)

cheers, Wolfgang

hi Wolfgang,

We are using JSCH mostly but in jobs running daily or few times per day. When we have jobs running so often (every 30 mins) we are doing it in recurreing way (as it is easier and faster instead of adding each entry to JSCH). 

Are You able to find Your watchdog script from the past ?

Regards, Krystian

Hi

unfortunately the old Watchdog considered MRT only :-(

So I created a new one for you.

Just include JOBI.WATCHDOG in Prescript or Process Tab of the object to check.

Config:

Explanation:

Watchdog checks one minute after activation if JOBP.WATCHDOGTEST runs in state 1550 (ACTIVE)

Watchdog checks one minute after activation if JOBS.UNIX.WATCHDOGTEST ended with State 1900 (ENDED_OK)

Watchdog checks one minute after activation if JOBSUNIX.WATCHDOGTEST.2 runs in state 1550 (ACTIVE)

Watchdog checks one minute after activation if JOBS.WIN.WATCHDOGTEST ended with State 1800 (ENDED_NOT_OK)

Precondition: a TimeZone Object must be defined in Job atrributes!

have fun!

If you have any questions pls let me know.

cheers, Wolfgang

Thank You so much!!

Regards, Krystian

You're welcome :-)

cheers, Wolfgang