Mirroring and Logshipping from the Production DB

Question

Post reply

Mirroring and Logshipping from the Production DB

SQL Bandit

SSCommitted

Points: 1757
More actions
May 2, 2019 at 8:17 am

#3621456

Hi Guru's 🙂
We currently have a Production DB Server and DB mirrored to our DR DB Server and DB.
As a Mirror its is offline and we can't run queries against it.
We are getting more and more requests from our Data Science engineer for data for reporting.
It's is possible to have log shipping and mirroring from the same principle? So we can log ship to a Data Science DB Server in our DR Datacenter so the Data Science engineer can query the log shipped DB to their heart's content?
My recollection is the log shipped DB can be online.
What's is typical practice for accommodating Data Science and reporting queries against a very large production DB that is Mirrored?
Many thanks etc

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply

SQL Bandit SSCommitted Points: 1757 More actions · Answer 1

I'd rather leave the Mirroring in place and have additional log shipping, I'd be interested in your opinions

Beatrix Kiddo SSC-Dedicated Points: 32407 More actions · Answer 2

Bear in mind that if you query the other replica you may be liable for an additional license cost.

SQL Bandit SSCommitted Points: 1757 More actions · Answer 3

Thanks,

But is it actually possible to have Mirroring and log shipping and from the same DB? Plus what is the typical approach? Is this a non-standard approach to interrogating production data against a 24/7 production DB , we have customers in multiple time zones and would rather not let out data scientist loose on a production DB even with just read-only access, giving them a DB backup is too bulky, some form of replication would be preferable.

DinoRS SSCrazy Points: 2683 More actions · Answer 4

To answer your question: Yes you can do both Mirroring and Log Shipping from the same DB. However Mirroring is marked as a deprecated feature so it might disappear in new SQL Server versions at some point.

There is no typical approach as everyone has different requirements in regards to their data, especially when it comes to how up-to-date the information has to be.

Another Approach would potentially be using AlwaysOn Availability Groups however your application would have to support this. With AGs you can have a read-only Instance for your Data Scientist.

This reply was modified 5 years, 6 months ago by DinoRS. Reason: spelling

Andrey SSC Eights! Points: 840 More actions · Answer 5

SQL Bandit wrote:

I'd rather leave the Mirroring in place and have additional log shipping, I'd be interested in your opinions

as an alternative and easy solution : create readable snapshots based on secondary replica

if your Data Science engineer is fine with it and additional license cost is ok for you, it can be kinda low hanging fruit