We have daily scheduled refresh at midnight and sometimes it would fail when the data is not ready. Any way we can set up auto retry when it fails instead of manually kick off the refresh in the morning?
Unfortunately, there is no retry ability.
You could either schedule multiple refreshes daily(one at midnight, one at 1am) or schedule it later (1am).
The only issue I could think of would be that you have data coming in today that you don’t want, but that can be avoided with a where clause.
Like Max shared there is no retry ability through the UI, but if you are doing refreshes from the API CreateIngestion, then you can create a lambda function to execute the API DescribeIngestion, where you can receive a status and if that status equals fail then it can trigger the API CreateIngestion.
I’m marking Bhasi’s response as solution to your question.
Also, since you mentioned that the refresh is failing occasionally due to data not being available, I assume that you have got other dependencies on that data load process which is not always in your control. In that case, it is best to go with triggering the data ingestion with a CreateIngestion API call (made from the end of your data pipline; once you are sure that all data is loaded and available) rather than doing a time based schedule and retrying. This way, you are not baking unnecessary wait time into your processing and can kick off the ingestion as soon as your data is available. If this is not feasible at your end (Maybe due to data pipeline being owned by another department etc), you can fall back on the scheduled refresh with a check and retry option as suggested by Bhasi.
Thanks everyone. Joined the QS office hour, and get the same answer.
In our situation, we are not shared the credential so are unable to leverage any API. The best solution for us is to set up a few more daily refresh in different timestamps.