[Question] High SNI_CONN_DUP Wait

[ppozeti]

Hi everyone,

This isn't directly related to DBA Dash, but the tool provided visibility into a metric that caught our attention.

We connected DBA Dash to our Azure SQL Server databases, which run on Elastic Pools.

While investigating an unrelated issue and reviewing some wait metrics, we noticed that about a week ago, there was a significant and sustained increase in SNI_CONN_DUP waits. We resolved this by scaling up the Azure SQL elastic pool. You can see the metrics over time and right after the scale up operation.

After looking into it further in Azure, we discovered that maintenance had been initiated just before the increase in SNI_CONN_DUP waits. We opened a ticket with Azure support, but they told us not to worry.

Even though they said it’s not a concern, we have some questions and would appreciate your input:

1. Have you experienced issues with sustained high SNI_CONN_DUP waits?
2. Does anyone have more information about this specific wait type?
3. If it’s not important, should we just exclude the wait in SQLWaits.sql?

[DavidWiseman]

Hi,

I can identify this wait type occurring on some Azure SQL DBs in my repository. I don't see this wait type at all on regular SQL instances. There is very little information on this wait type.

SQLSkills doesn't provide any additional information. I guess SNI refers to "Server Network Interface".

I suspect this can be added as an ignorable wait type, though it would be useful to know more about it.

Thanks