Priority mapping not respected in Jira Service Management incoming alert rule

[Michael.Gaskin]

I am using the Jira Service Management integration for OpsGenie. In my Jira instance, I have the Jira priority values (Blocker-High-Medium-Low) mapped to OpsGenie priority values (P1-P2-P3-P4).

When using this rule, the priority mapping works fine: ELSE IF eventType equals issue_created THEN create alert

In Jira Service Management, I have Automation rules to set the priority based on Impact and SLA Tier. To give these Automation rules time to run, I use the issue_updated eventType in OpsGenie instead:

ELSE IF eventType equals issue_updated AND labels contains headless_incident THEN create alert

However, when using this rule, although the priority value in "Extra Properties" is Blocker, the priority is set to P3 and not P1.

I could hack around this by creating multiple rules to set the priority manually, but it would be nicer just to use the {{priority}} dynamic value in a single rule.

Any ideas?

[John M]

Hi @Michael.Gaskin ,

I don't think the filter being different would cause an issue.  Normally, if you're using the mapped priorities, when it reaches Opsgenie you need a number to extract. For example:

P{{impact.extract(/[0-9]/) }}

The fact that the alert is showing "Blocker" in the priority field likely means that the mapping could not process anything into a valid Opsgenie priority of P1-P5. 

Can you post what is in your Mapped Priority Value field in Opsgenie?

[Michael.Gaskin]

Hi @John M

I worked on it more last night and found that the rules are indeed a red herring; what's really happening is:

• If the Priority value in Jira is Low, Medium, or High, the mapping works fine.
• It only doesn't work for Blocker.
• However, if I add a condition to my rule that regex-matches "Blocker" and sets the priority to P1, that works. 
  
  In the screenshot below, you can see that the priority mapping is all done in the same place in Jira, so I'm not sure why Blocker is the only one that doesn't work.

[John M]

@Michael.Gaskin can you verify that the issue type for the issue(s) that are not working are 'incident'? The mapping only works for issues with a type of 'incident'