Re: hot backup discussion

There's actually zilch problem. What happens is that the database looks suspiciously out of synch to SMON, and hence a 'database recovery required' is signalled -but it's all a total lie. So what you then have to do is to end the hot backup for each affected tablespace -except that, stranded in mount stage as you are, you can't do it at the tablespace level, so instead you are forced to issue the 'alter database datafile X end backup' command.

And if you have the maximum permitted 1023 datafiles per tablespace, you could be stranded in mount stage for a *very* long time whilst you do all that typing.

But apart from, that, no harm is done, and no real recovery is required. So it's a convenience and mean time to get back to work issue rather than anything else.

But your man should have known better, regardless.

Regards
HJR

--
--------------------------------------------------------------------------
Opinions expressed are my own, and not those of Oracle Corporation
Oracle DBA Resources:               http://www.geocities.com/howardjr2000
--------------------------------------------------------------------------



<kal121_at_my-deja.com> wrote in message news:8rb0j8$cu2$1_at_nnrp1.deja.com...


> Yes, very bad idea. But I was wondering what happens to the database if

> you put all the tablespaces in hot backup mode at the same time and

> then the database crashes during the backup. What kind of problems

> would this cause?

>

> The reason I need a reality check is because a dba here wrote a script

> to put all tablespaces in hot backup mode at once and I told him that

> was a very bad idea. Not just because of the tremendous amount of redo

> generated, but because of the danger of the database crashing during

> the backup. But I need to back that statement up.

>

> Thanks

>

> In article <39d54212_at_news.iprimus.com.au>,

>   "Howard J. Rogers" <howardjr_at_www.com> wrote:

> > Lousy idea, unless you want to see your redo log sub-system thrashed

 to


> > within an inch of its life.  When a tablespace is placed in hot

 backup mode,


> > the smallest change to a block within that tablespace causes the

 *entire*


> > block to be written to the log buffer (and hence onwards into the

 online


> > redo logs and the archived logs).  Normally, only the changed data

 itself is


> > written.  Hence the throughput on redo goes up enormously.

> >

> > So, extremely bad idea.

> >

> > Regards

> > HJR

> > --

> > ----------------------------------------------------------------------

 ----


> > Opinions expressed are my own, and not those of Oracle Corporation

> > Oracle DBA Resources:

 http://www.geocities.com/howardjr2000


> > ----------------------------------------------------------------------

 ----


> >

> > <kal121_at_my-deja.com> wrote in message news:8qolvs$ok5

 $1_at_nnrp1.deja.com...


> > > What do you think of the practice of putting ALL OF YOUR

 TABLESPACES in


> > > hot backup mode simultaneously? Is this a good idea or bad idea and

 why?


> > >

> > > (Of course I would never do this myself, but what do you all think

 of


> > > it?)

> > >

> > > Thank you

> > >

> > >

> > > Sent via Deja.com http://www.deja.com/

> > > Before you buy.

> >

> >

>

>

> Sent via Deja.com http://www.deja.com/

> Before you buy.