Oracle Database 23ai: Backup and Recovery - Part 1
Description
Episode Transcript:
Welcome to the Oracle University Podcast, the first stop on your cloud journey. During this series of informative podcasts, we’ll bring you foundational training on the most popular Oracle technologies. Let’s get started!
Nikita: Welcome to the Oracle University Podcast! I’m Nikita Abraham, Team Lead of Editorial Services with Oracle University, and with me is Lois Houston, Director of Innovation Programs.
Lois: Hi there! For the last two weeks, we’ve been having really exciting discussions on Oracle AI Vector Search. We covered the fundamentals, benefits, the vector workflow, and lots more. Today, we’re going to talk about backup and recovery in Oracle Database 23ai with Bill Millar. If you’ve been listening this season, you’ll know that Bill is a Senior Principal Database & MySQL Instructor with Oracle University.
Nikita: In this two-part special, we’ll dive into some of the things you need to know about backup and recovery, especially if you’re a database and backup admin. So, if you're the person in charge of keeping data safe and handling disaster recovery, this is definitely worth your time.
Lois: That’s right, Niki. Hi Bill, thanks for joining us again. What’s the role of a Database Administrator, or DBA, when it comes to backup and recovery?
Bill: The DBA is typically responsible for ensuring the database is open and available when needed and at times you need to work with system administrators and other people within your organization to achieve that. But we want to try to protect the database from failure wherever possible.
We want to increase the mean time between failures. Hopefully, we don't have failures, and we have to increase that time. But it might mean that we need to ensure we have redundant hardware and that in place, again, maybe out of the realm of the DBA, but people within your organization can help with that. We want to protect those critical components by using the redundancy. And we want to decrease the mean time to recover. Failures happen, but how fast can we get access back to that data after that failure.
The faster we can do it, the happier customers are. Minimize the loss of data. It's never good to lose data, especially in a critical environment, but maybe in test and development, maybe not so bad.
02:39
Nikita: How do we ensure a separation of duties for backup and recovery processes?
Bill: For a separation of duties, we do have a user called SYSBACKUP. It has the privileges that's required to perform backup and recoveries, the privilege to connect and execute the commands in what we refer to as RMAN, our Recovery Manager. As I said, it has permissions for backup and recovery because you do need to shut down the database, start up the database, those type of things.
We're able to connect to that closed database to try to troubleshoot it, to get it to the open state again. It does not include any privileges to access data. The SYSBACKUP user is created when we install the database, when we create the database. We can use it explicitly for privileged user connection. It allows us to connect to the database. So RMAN connects as SYSBACKUP.
Lois: Bill, what should people keep in mind when figuring out what’s considered critical data?
Bill: You want to try to identify your critical data. Some data might be highly required to access and make sure we don't lose don't lose data, but then you might have some environments. OK, I don't need to have them up and running as fast. If we lose a little data, it may not hurt, but we want to identify the difference in the different data that we have on different environments.
So we want to also prioritize that critical data, which data do we need access to first because how much will the company lose per hour of downtime because we can't do business. We want to make sure the access data protection requirements. Not everybody has access to everything. And there are different types of disaster that can happen that are going to be totally out of your control. There's the physical disaster, a hurricane or tornado, outages, power outages, component failures, failures within the building itself, corruption of data because of some of these failures.
And then, the most dreaded one, the one that happens most often, usually those human errors, the logical errors, where the data is just bad, we are able to access and everything. It's just that something has changed that shouldn't have been changed. We want to make sure we access our recovery requirements.
Lois: So, what are they? What are those requirements?
Bill: We want to base that requirement based on how critical is that data, how soon do we need to have access to that? What is our recovery point objective?
Do we have a tolerance for any type of data loss? How frequently should we backups? How often they should be taken? What type of backups will be another thing we'll want to figure out? Is point-in-time recovery required? Are we able to or do we ever need to go back to a previous point in time to do something? It's not always just recovery for a database failure. We might need to do a recovery point in time to a different system so we can investigate something. What is my recovery time objective? Again, what is the tolerance for the downtime? How long can I be down?
The downtime, the biggest part of when a system goes down is trying to identify what is the problem, then next is what is going to be my plan to recover, and then perform in the recovery. We might have a tiered required time objective based off of critical data, and then depending on the failure. Is that failure at the entire database? Is it just a tablespace? Is it just a table? Is it just a row? That also determines how long it takes to recover and what type of recovery we might try to perform.
What is my backup retention policy? Do I have a requirement to where I have to have my backups off site? And it doesn't mean like back in the old days of mainframe computers, you'd back up to tape and you'd take those tapes off site. You might still do that today. Or, am I backing up to a cloud environment? So what do I need to have for that? What about long-term backups? We work with our day-to-day backups, but there's those backups that require for longer, archives like end of year backups. Some places require to keep their end of year backup for like 10 years. How are we going to handle that? So these are some of the things that we have to think about when we start talking about backup and recovery.
Did you know that the Oracle University Learning Community regularly holds live events hosted by Oracle expert instructors. Find out how to prepare for your certification exams. Learn about the latest technology advances and features. Ask questions in real time and learn from an Oracle subject matter expert. From Ask Me Anything about certification to Ask the Instructor coaching sessions, you’ll be able to achieve your learning goals for 2024 in no time. Join a live event today and witness firsthand the transformative power of the Oracle University Learning Community. Visit <
United States
