How to handle if http time outs ??

[Manikantahere]

We have a ServiceNow-Jira integration in place.

When a change is created in ServiceNow, a corresponding record is also created in Jira, and the correlation ID is updated back in the ServiceNow change record. However, in negative scenarios, the Jira record is created but updating the correlation ID back in ServiceNow fails due to an HTTP timeout.

If I reprocess in this situation, it will create a duplicate record, which needs to be avoided. What are the best practices in ServiceNow to handle this?

[Bhimashankar H]

Hi @Manikantahere ,

If it fails due to HTTP time out then you try Automatic retries. If the update to ServiceNow fails due to a timeout (or any transient HTTP error such as 408, 500, 502, or 504), the integration should automatically retry the update after a brief wait. Use increasing wait time between retries to avoid overwhelming the downstream system. If timeouts are systematic and not due to application errors, consider increasing timeout values for outbound requests. Check what the time they configure the timeout for outbound integration from JIRA side as well. Log the time to see how much time it is taking to process the data.

Process integration updates asynchronously whenever possible. By queuing update operations, you decouple the real-time system.

You can always see the integration fails by logging the record data so that you can troubleshooting the issue. Make the correlation ID update operation idempotent so that retrying the same update multiple times won’t negatively affect data consistency or system state.

Best Practice:

Retry logic & exponential backoff, 

Asynchronous processing, 

Logging of all integration failures, 

Alerting & notifications, 

Timeout value tuning.

Thanks,
Bhimashankar H

-------------------------------------------------------------------------------------------------
If my response points you in the right directions, please consider marking it as 'Helpful' & 'correct'. Thanks!

[Manikantahere]

What do you mean by retry update? if it is sending one more rest call  it will create another request which should not.

correct me if you are referring  a different functionality.

[Bhimashankar H]

Hey,

It is failing from JIRA side so you need to check from that side the timeout setting and retry mechanism from JIRA side. Like there is functionality in flow designer where you can wait for some specified time and then do the REST again, this is I am referring. It is failing might be because of JIRA has very limited timeout or your instance is taking much time to process it, check the configuration in servicenow for timeout and do not forgot to log the data which is coming from JIRA side. 

Actually, it seems the issue from JIRA side only. Check the data by logging the values.

Thanks,
Bhimashankar H

-------------------------------------------------------------------------------------------------
If my response points you in the right directions, please consider marking it as 'Helpful'/'correct' Thanks!

[Bhimashankar H]

@Manikantahere ,

I hope you saw my reply. 
If my response points you in the right directions, please consider marking it as 'Helpful' & 'Correct'. It will help future readers as well having similar kind of questions and close the thread.

Thanks,
Bhimashankar H