Enable Query Store for collection on a read-only replica in an Availability Group
Currently, Query Store can only be enabled for the read-write database in Availability Group. As many customers issue queries against the read-only replicas, capturing query and performance metrics for those queries would be beneficial for understanding the workload, troubleshooting performance issues, etc.
Mike Molnar commented
We offload reporting traffic to the read-only secondary node. The lack of Query Store makes it much more difficult to analyze performance of those report queries.
Jeremy Weir commented
This is not a nice to have, this is a requirement. Anyone who enables read scale out has no idea if they are under or over utilizing it.
it is a great feature
Jesús Eduardo León Fonseca commented
This feature should be included because the read-only replicas have many different use cases and, it will really be useful to have this tool helping us to understand the behaviour and how the data is queried in a ReadOnly replica.
This would be very beneficial
Martin Smith commented
Hopefully this will be implemented for Read Scale-Out replicas too.
this is an important missing feature. our readonly replicas have very different use cases and its important to understand granular RO query data
very important for us to get some important figures on runing Store procedures
Jan Wróblewski commented
Much needed feature, as this is very common scenario.
Peter Åkerlund commented
This feature should be essential for us when using heavy queries against the read only replica.
This is a very important feature which is much needed. We have been using Geo Replication and secondary database for read-only workload but there is no way to identify the query performance or even the performance of the readonly database itself.
Adam Wiedenhaefer commented
Include this for Geo-replicated databases in Azure.
Aakash Patel commented
This feature will make our life much easier for optimizing secondary work load, will be a great addition!!
Vrambout Peter commented
Maybe add it as an option if you want to include replica data, that way you can enable it when needed and are aware of the performance impact it may have.
this is will be very much required feature, for Read-only replicas to leverage the real benefit of Secondary replicas.
Probably , we can keep this data at local file level... just like Trace or Extended Event data or writing to primary replica thru a listener with a simple flag of distinguish b/w primary & secondary queries
Sivakumar Vengala commented
We would like to enable query store on read only replicas since we have n number of downstream application accessing the databases. We would be able to provide some insight during performance issues if we can enable the query store on AG Read-Only Replicas.
This would be tremendously useful - even if the read-only replica was an in-memory version of query store that did not persist after reboot (may be easier to implement)
David Solomon commented
We depend on query store to do a tremendous amount of optimization. Please add this feature.
Christian Rodríguez Giménez commented
When your'e at it, do it for Azure Sql Database too