Oracle 12c Data Guard Architecture
Vložit
- čas přidán 28. 06. 2024
- Data Guard is basically a ship redo and then apply redo, as you know #redo is the information needed to recover a #database transaction. #Dataguard is meant for DISK OUTAGE (or) SITE OUTAGE. Data guard is a disaster Recovery solution.
A production database referred to as a primary database transmits redo to one or more independent replicas referred to as #standby databases.
Standby databases are in a continuous state of recovery, validating and applying redo to maintain #synchronization with the primary database. A standby database will also automatically re-synchronize if it becomes temporary disconnected to the primary due to power outages, network problems, etc.
00:00 Data Guard Architecture
07:17 Data Guard Architecture Scenarios
16:55 Different between Standby Redo log files and Redo log files
19:48 Does that mean only standby has the standby redo log file?
Website: www.dbagenesis.com/
Facebook: / dbagenesis
Instagram: / dbagenesis
Twitter: / dbagenesis - Věda a technologie
Hi Arun, You make hard topics very easy to understand ..Thanks alot for your contribution to DBAs
Short and precise! Excellent
Glad it was helpful!
Very simplistic explanation of Data Guard. Thank You :)
You are welcome!
Very well explained Arun Sir. You make DBA so simple.
Thanks a ton
It's really amazingly easy to understand the concept of DG... It's a WOW explanation...
Glad you like it
I like your paintings. Where do you exhibit them? I would gladly attend your art gallery. Those LNS and RFS are marvelous!
Very helpfull 👍🏻
Looking forward to see more videos like this and learn.
More to come!
Great explanation Arun, Thanks a lot!
Thank you!
Clear explanation, thank you !
You are welcome!
Very well explained
Wonderful Explaination on failover and switchover concept
Glad you enjoyed it!
Your explanation is crystal clear -so awesome, it compelled me to park every aspect of my thought process-so engaging -I would just be so happy to have a quick chat with u. I was curious to know more about you and went to your site-No videos on RAC-Patching-Exadata? I'm eager to discover your videos -especially troubleshooting and optimization-Thank You Arun
Most welcome! Yup, most of the videos (which you are expecting) are part of our one-on-one advance training programs. You can always send a request to our team regarding the topics you want to master - support@dbagenesis.com
We can always custom build courses for the topics you are looking for!
Thank you, good explanation
You are welcome!
great work 👍
Thank you! Cheers!
Good Explanation....done...
Super bro ....clear cut explanation......
Thank you!
The best!
Thank you!
As you said, when it's stand by is being used for reporting purpose or if its a active dataguard the redo logs files on standby will be used. Ok then which process will be writing logs to redo logs? Log writter process? So when data is writtten redo, will mrp read and apply from redo?
By the way very beautifully explained. Loved it. thanks!
Data is written to SRL (standby redo logs) and then read by MRP.
Great Explanation Thanks.
Could you please tell more about how to manually resolve gap when we don't have archive log ?
Also how to manually force rman to take backup only the archive applied on standby ?
Thanks.
you may do it with rman from standby side. RECOVER STANDBY DATABASE FROM SERVICE PRODDBSVC;
Hi Arun, very well explained i wasn't excepting that anyone can explain that topic to easy ,you just made me to think of DBA Genesis every time i need to learn something. Also i had one doubt if you could clear that as the log shipping is going on the MRP is always waiting for the archive to apply as we see in managed_standby view as sequence#.SO as per you we go with first process (redo log buffer) then second process(redo logs) if lgrw is ahead of LNS and then with Third if there is gap but MRP is applying archives only...can you please explain this why we always see archive log(sequence#) in MRP rather than redo buffer or redo logs.
Coz the default configuration of standby is archive apply. You can change it to redo log apply which is also called as real-time log apply. Thats similar to SYNC configuration as compared to ASYNC configuration which uses archives.
Would you be able to explain this DG with one example as you explained oracle scan ips( airport example ) before
Nice video!
Thank You!
If SRL space gets full in standby database, then we consider switchover from standby to primary DB in the data guard environment?
Thanks a lot sir. Could you please elaborate more about how to manually resolve gap when we don't have archive log.
you should not delete archive unless they are applied to standby ) but still they are in your backupsets so you may restore them at any time
Hi,
Can we create SRL on primary after standby configuration setup ?
Can I get the 1st video related to data guard please. I'm missing some concepts it seems.
you mentioned that standby database need not be in archive mode-however if the standby is now created as active standby database then would it be necessary to put in archive mode, though if u use it for only reporting it will not be required but pls could you throw some light
Archivelog mode on standby is only required when its in Logical Standby mode. Except that, you do not need to put it in.
One more situation would be when you configure cascading DR.
Very clear explanation. Keep it up :)
There is a formula for calculating the number of standby redo logs
(maximum # of logfile groups on the source db +1) * maximum # of threads
--ASSUMPTION
SOURCE IS RAC - 2 NODES
TARGET IS RAC - 2 NODES
Suppose on source there are 10 thread 1 redo log groups and 10 thread 2 redo log groups, this means there will be 21 thread 1 redo log groups and 21 thread 2 redo log groups on target? What if by mistake I created 10 thread 1 and 10 thread 2 redo logs?
Thanks!
So let's drill down to basics.
Redolog Thread = Redolog groups that belong to an instance
In RAC, # redo log threads = # of RAC instances
What you mentioned (10 threads) is not possible for a two node RAC (off course you can create as many threads as you want but you can assign only one redo log thread to one RAC instance at a time).
But, yes, each RAC instance can have min 2 redo log groups or more. Each redo log group can have one or more redo log members.
Example:
If I assume that 2 NODE RAC will have 2 redo log threads and each thread is having 10 redolog groups with 2 redolog members, then going by below formula:
SRL = (maximum number of logfiles for each thread + 1) * maximum number of threads
SRLs = ((10 * 2) + 1) * 2 = (20+1)*2 = 42!
Don't be surprised by the 42 number. Understand, on RAC primary you have:
ORLs = (redolog group * redolog member) * log threads
ORLs = (10 * 2) * 2 = (20)*2 = 40!
ORL = Online Redolog Files
Hope this helps !!
@@dbagenesis Thanks for clarifying on the redo thread. If by mistake I created 22 standby redo logs (each thread having 11 redo log groups), will that cause a problem? I did it on my target but hopefully replication is working fine.
Hi DBA Genesis , Do you have any courses for Cloud DBA - on AWS or Oracle Cloud ?
Not yet, but we shall have it very soon.
Hi Sir. Thanks for sharing the good content with us.
Once we set up the data guard environment. Incase if I forget to create a Standby Redo Logs on primary database. Can i create it later.
Yes, you can
@@dbagenesis Thank you sir
Excellent video Arun ji, Now One more que apart from this.
Can You let me know how much minimum number of archives should be applied to open a database when we are performing a restoration of db on another server? It's help me alot as I have searched it on many sites but not got any solution
I think Lns process read from online redolog, not from redolog buffer
sir, if rfs is down and lns is down how to recover?
And if mrp is not apply log in standby in standby database ? How to recover?😊
how to create secondary database sir can u make any video for us
How can we enroll for your courses? I am looking for a RAC course..
To enroll into RAC course: dbagenesis.com/p/oracle-rac-beginners
To check all our DBA courses, go to: dbagenesis.com/courses
u never answered ur question-what happens if rman deletes archivelogs required by standby db?
Why don't you figure out answer and reply to the comment? If you go by logics, your question has the answer!
@@dbagenesis yes I think I have the answers
@@zulkifal111 Hi Fernandez .What's the answer?
DataGuard is free or not?
It's free with enterprise edition. Active data guard is extra license.
@@dbagenesis Is Data Guard different from Active Data Guard?
Good Explanation..done...