Re: archive log maintenance on physical standby

> What is best practice for keeping the recovery file dest from filling
> up and stalling the archiver process on the physical standby?

I have a physical standby, but my setup is only similar 'in principle' - I'm on 8i, and in my case it's managed.

The redo is pushed across to the standby dest at the OS level, every 20 minutes.

In my case, the logs are compressed, and then rcp'd to the standby server. My standby is in read only mode (for reporting) except for twice a day, when I apply accrued redo from the master.

At those times, I

-uncompress all .arc.Z files in the recovery dest
-run a 'recover database; auto' to apply those logs
-rm *.arc.

Any redo which might have been shipped to the standby in the interim will still be compressed, and will therefore not be deleted.

As to your scenario, I suspect that you should be able to get the current change number, from 'archive log list', for example, parse that number out to a shell script somehow, and trigger a 'host' command to delete all files whose sequence is lower.

You may need to create some shell scripts - I expect cut and sed should be all that's really required.

And hey - there may be a provision within the Dataguard toolset to do the same thing.
Which I expect is what you were looking for in the first place. But, in the absence of such...

Hope that helps a bit... or at least gives you an option.

BD Received on Wed Aug 02 2006 - 18:16:35 CDT