Incidents in error - how to rerun last task?

[Antanas]

Hi!

I often encounter errors in incidents due to temporary API integrations failure. I have some automated retries setup for each task which mitigate it somewhat. However if the failure lasts longer - incidents stop in error. I usually open each incident and click "Run automation now" on the failed task. Is there an automated way or command, to force them to retry the last task the incident failed in bulk at this stage? I do not want to rerun the whole playbook, just the last failed task (which could be different for multiple incidents).

Thanks,

Antanas

[jfernandes1]

Hi @Antanas, you can use the retry option for a specific playbook task. Refer - https://docs-cortex.paloaltonetworks.com/r/Cortex-XSIAM/Cortex-XSIAM-Administrator-Guide/Playbook-Ta...

 

[Antanas]

Hi @jfernandes1 ,

Definetely, I'm using this option to overcome temp API issues which happen on short time period, like repeat the task 3 times every 60 sec. But in order to address the longer lasting issues, i need either additional repeat X times every e.g. hour, or manually rerun each task.

Sounds like oximoron - now i'm just manually reruning tasks on an automation tool ¯\_(ツ)_/¯

[jfernandes1 · Accepted solution]

Hi @Antanas, There are some option available for this. Could you install our Integration and Incident Health Check pack? There are 2 automation that will help. 

Get Failed Task - https://xsoar.pan.dev/docs/reference/scripts/get-failed-tasks

Restart Failed Tasks - https://xsoar.pan.dev/docs/reference/scripts/restart-failed-tasks

You could wrap these command into a playbook, then run it as a job. Or maybe even manually triggered. 

[Antanas]

@jfernandes1, seems like exactly what i needed - thanks!