Re: Media recovery when using an old controlfile

In article <3ef9d511_1_at_news.estpak.ee>, Tanel Poder wrote:

> 1) If your array supports write caching, are the cached writes written to
> all splits?

As I understand it, yes. The freezing and thawing of I/O happens at the AIX device driver level; from a certain point of time no new I/O actions are accepted, and current I/O actions are completed. The split is done, after which new I/O actions are allowed again. The storage array is an EMC Symmetrix with TimeFinder enabled, and PowerPath on the host.

> 2) Do you split *entire* database atomically, that means all datafiles,
> redos and controlfiles? If half of db is taken at one time and second half
> is taken 1/100 seconds later you could easily have inconsistencies.

The entire database is split atomically.

> 3) I've always put datafiles to backup mode before splitting. There is a
> command alter database suspend / resume which should disallow any writes

With these kind of splits, I have always been able to split off a restartable database. It was only when non-current controlfiles started to get involved that I ran into problems (the main problem being my limited Oracle - or databases in general - knowledge :-) ).

-- 
Jurjen Oskam

PGP Key available at http://www.stupendous.org/