[Video] Office Hours: Open Q&A About Databases
I'm back in the home office to take your database questions from https://pollgab.com/room/brento.
https://youtu.be/H70Iy6bX970?si=6UZJbG89FC6bVni4&t=15
Restore testing, restore performance, rollback, and recovery procedures.
25 associated posts24 primary posts
I'm back in the home office to take your database questions from https://pollgab.com/room/brento.
https://youtu.be/H70Iy6bX970?si=6UZJbG89FC6bVni4&t=15
Post your Microsoft data platform questions at https://pollgab.com/room/brento and upvote the ones you'd like to see me answer. I sat down by the pool in Cabo with a cup of coffee to go through 'em:
https://www.youtube.com/watch?v=jGVl1NWp2zU
It's the first webcast I've done from my new Vegas pad. You posted questions at https://pollgab.com/room/brento/ and I discussed the highest-upvoted ones:
https://www.youtube.com/watch?v=3wV64rqwxSk
In our last post in this series, I talked about why you should be using Ola Hallengren's maintenance script for backups and sp_DatabaseRestore for faster restores. These two tools have implications for how you build, back up, recover, and fail over your SQL Server.
There's a saying in the cloud: treat your servers like cattle, not like pets.
When I first started out as a SQL Server DBA, I thought things were going well as long as the backup jobs ran successfully. I'd go into SQL Server Agent every now and then, make sure they were still running, and ... that was the end of it. I figured if disaster ever struck, I'd just do a restore. How hard could it be? I was missing the entire point.
Wouldn't it be nice if rollbacks just finished instantly? Wouldn't you love for startup times to be near-zero even when SQL Server crashed just as someone in the middle of a transaction? How much would you pay for all this? (Well, I'm a little afraid to ask that, since we don't know yet whether this is an Enterprise-only feature, but dang, I sure hope not.)
This week, Brent, Erik, Tara, and Richie discuss whether you need to premake a destination database in order to run sp_databaserestore, restoring production statistics on dev servers to simulate productions behavior, dealing with duplicate SPNs, tools for detecting all SQL Servers in a production environment, SQL performance monitoring, dealing with very large indexes, and Brent's recent cruise.
This week, Erik, Tara, and Richie discuss disaster recovery, Python vs C# for SQL Server, whether you should perform log backups on primary or secondary on AGs, why Postgres is so popular, using mount points for tempdb, attention events, database corruption, MERGE statements, restoring a table from a database, features they look forward to in SQL Server 2019, and how Santa is able to visit all those homes in one night...
Anything Brent Can Do
There's some things you can do with Managed Instances, and some ways you can look at data that you can't do elsewhere.
For instance, you have the ability to start an restore ASYNC -- that means if your session drops, the restore will continue headlessly in the background.
You've built your first Managed Instance, and now you wanna play around with real live data. You're going to need a backup in the cloud to do it - SSMS doesn't have a drag-and-drop option.
Big picture, here's what we're going to do:
This week, Brent, Erik, Tara, and Richie discuss database restores, opening a case with Microsoft, buying a pre-built server vs building your own, auto-shrinking all databases with a single command, SQL Server dev environment options, the best way to learn Git, excessive memory grants, and more.
Here’s the video on YouTube:
This week, Tara, Erik, and Richie discuss 3rd-party tools for object-level restore, patching, stringing multiple CTEs, Meltdown and Spectre, instance stacking, organizing DBA teams, DBCC checks, moving local databases to Amazon RDS, CTEs vs temp tables, and backups and restores.
Here’s the video on YouTube:
Carry on
It turns out that the only thing harder than checking for new databases restored to a SQL Server, is checking a folder for a backup of a database that doesn't exist on another SQL Server.
Phantom of the Nopera
Let's say you have a server. Let's go one step further and say SQL Server is running on it. Crazy, I know.
But hey, we specialize in crazy, here.
Brent had an idea that we should set aside a day per month where we could build things in the cloud and blog about our experiences. I was tasked with Amazon RDS for SQL Server: create an instance, configure backups and test point-in-time restores. After grabbing some coffee, I logged into the AWS console. We…
We're trying something new at the company: Builder Day. We define a slightly-out-of-the-norm task, and then the team splits up and tackles the task on different cloud platforms, writing it up as we go. These posts aren't going to make you an expert on the topic - they're just meant to let you skim through technologies you might not otherwise get the chance to play with. The cloud changes fast, so some of this is just about seeing what providers are up to these days.
You use Ola Hallengren's maintenance scripts to do your database backups on SQL Server 2016 or newer, and you want to do things like:
Restore the most recent database backup
Test it for corruption
Initialize restores to get ready for log shipping or replication
It's easy to get caught up in daily incidents, tickets, and special projects. Like a good scout, though, a core task in a DBA's job is to be prepared. You have to set time aside in your schedule to: 1. Practice responding to corruption. Gail Shaw's "Help, my database is corrupt, now what?" post should…
... is to use a SQL Server where a robot does it for you! Let's take a look at how you can do this in AmazonRDS's hosted SQL Server these days. Normally, restoring transaction logs is super tedious. You've got to restore all the files in the right order, for every single database. Even if…
One of the many restore features in SQL Server is the ability to restore one or more pages of data. This can be very convenient in some narrow situations - for example, corruption occurs on one page or an oops update is made to one record. The page restore process is not straightforward, however, and,…