Error retry is a feature which gives a possibility to try to process the failed messages in the integration flows.
Any complex integration flow can fail due to unpredictable problems like network disruption between the platform and the third party source such as your CRM or ERP. If this disruption occurs during the API request there is a high chance the platform would not process the data and report an error like timeout or remote host is unreachable.
In another example with reading the data from an external database every day. Now if the database is under maintenance or offline your flow step responsible for connecting and downloading the data would fail. This would disrupt your integration flow.
To recover from this situation might seem tricky at first, but, the platform features are here to help.
When the platform fails to connect with a third party resource the request or the
intended data would still be in the processing queue. If you have configured the
rebound the system would retry it until it goes through.
Otherwise, if the connectivity is not back in couple of hours the system would
Rebound limit exceeded and would stop trying. The platform would stop
trying after 10 times.
When the automatic retry fails to deliver due to the time constrains you can use the manual retry of the errors. Here is how it works.
Let’s start a generic integration Flow and cause an error. It will be seen here in the logs. Then we run it, see errors, and go to check the runlog to click Retry and see what happens:
When you press Retry the platform will then try to run the erroneous process again. The error will be removed from the execution page.
Note: the integration flow must be active for Retry feature. When the flow is inactive the Retry button would not be visible.
Before you proceed to press Retry button try to assess the situation:
If answer to any of these questions is no then it is highly possible you would get an error again after the Retry.