How to never lose data with your Standby Database

Your Oracle Database needs redo logs in order to function. You do not need standby redo logs for your standby database to function, but you would be very foolish to not have them. Just like redo logs are critical to make sure you never lose data on your primary database system, standby redo logs are critical for your standby database to never lose data...


Subscribe for new tech videos every week



Are you serious? A free Oracle database forever ?!?!?!?! Hell yeah!!!




Music: Night Owl (Broke For Free)
, Dyalla

#oracle #dataguard #redologs

[bzielinski]

Thank you Connor. One more gotcha from my side. Do not multiplex standby redo log. This is written in Larry Carpenter book Oracle Data Guard 11g

[FrankyWeber]

It is worth noting that in a RAC system when we have standby redo logs assigned to the primary threads we must have the correct threads for the standby redo logs matching the primary ones. For example If I have a 4 node cluster in the pmy and a 4 node cluster in the sby but my DB is running only with 2 instances, let’s say pmy runs on 1 and 4 while sby runs on 1 and 3. Disregard of the sby instance number, the sby redo log thread must match the pmy threads; in this case threads 1 and 4. Otherwise we will see that pmy thread 4 won’t use a sby log and will wait for an archivelog to be generated. So we must have a sby redo log for thread 4 even if there is no sby instance 4. Long time ago I had this situation and I fried eggs in my head trying to figure out so I just wanted to share it with anyone reading this comment. 🙂

[berndeckenfels]

Is there a guarantee that the archive log is copied before the standby redo is overwritten? Wouldn’t it be more efficient to not copy the archive but instead archive the standby redo as it is already on the remote destination?

[William-B]

Is archive_lag_target “obsolete” now or is there still a purpose for it?