Problem with Azure Data Factory IP-Range changing overnight

[marius1106]

Dear Support,

In an Azure Data Factory pipeline, we retrieve data from Snowflake and load it into the Azure Data Lake. So far, we have whitelisted the IP addresses of Azure Data Factory in Snowflake. However, the issue is that these IP addresses sometimes changes overnight and our scheduled Pipeline gets an Error. We have now switched the Integration Runtime to Germany West Central and added the corresponding IP ranges of Germany West Central from Download Azure IP Ranges and Service Tags – Public Cloud from Official Microsoft Download Center to Snowflake's whitelist. Despite this, we are still facing the problem that the IP-address sometimes falls outside the specified IP range, resulting in no access to Snowflake.

It would already help if we knew, for example, that the IPs only change at night between XY o'clock, so we could schedule the job at a different time.

Thank you & Regards,

Marius

[v-yilong-msft · Accepted solution]

Hi @marius1106 ,

Azure Data Factory's integrated runtime IP addresses change without a fixed schedule, so predicting when these changes will occur is not very simple. However, I think you can try the following and check the official documentation below.

1. If possible, consider using a Managed Virtual Network in Azure Data Factory. This can help provide more stable IP addresses for your data flows.

Here is a document I think you can read: 

Azure Integration Runtime IP addresses - Azure Data Factory | Microsoft Learn

2. Utilize Azure Service Tags to simplify network security rule management. Service tags represent a group of IP address prefixes from a specific Azure service, and Microsoft manages the address prefixes encompassed by the service tag and automatically updates the service tag as addresses change.

I think you can also look at this document:

3. Since there's no guaranteed time for IP address changes, you might want to schedule your pipeline to run multiple times throughout the day to increase the chances of successful execution.

Best Regards

Yilong Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.