Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: quick hot backup question...

Re: quick hot backup question...

From: Howard J. Rogers <dba_at_hjrdba.com>
Date: Tue, 26 Mar 2002 13:53:17 +1100
Message-ID: <a7onsq$kfp$1@lust.ihug.co.nz>


I don't know how you can disagree, since it's not a matter of opinion but basic physics! If the database is up and running, one cannot 100% guarantee that an extra transaction or three will not take place. The database is open, therefore it can accept connections.

Now you might prevent those connections by locking your application down as you describe, but I, as a disgruntled employee, have been taking evening classes in SQL*Plus. How are you going to stop me connecting? Database logons and security? Well, I happened to get your Junior DBA drunk the other night, so I know what SYSTEM's password is. And so on: you get the point, I hope. Yes, you can build all sorts of protections in to say 'I know that no-one can do transactions', but the only way you can be 100% sure, under all circumstances and all situations, is for the database not to permit connections by not actually running. Even killing the Listener won't do it, because whilst the junior DBA was nursing his hangover, I happened to lift his security pass to the Server Room.

So I don't doubt that *you* have things locked down, and get the reassurance you need about the state of play on the database, but that's you and your specifics, which no doubt are perfectly valid for you in your specific environment: but *in principle* if the database is up, one cannot guarantee that no further transactions are permissible or have not occurred or can be prevented.

Not unless you want to make it read only, of course, or use the 9i quiescing feature.

I'm trying to be as generic as possible here, because you weren't the one asking the original question, so the specifics of how specifically you (and your app, of course) gain the assurances you need might not necessarily apply in all cases.

I also get a bit queasy about such important database management matters (who can connect and/or when) being left to the tender mercies of application code (and the developers who have to write it).

Regards
HJR

--
------------------------------------------
Resources for Oracle : www.hjrdba.com
============================


"daniel" <test_at_test.com> wrote in message
news:a7oaca$h2g$1_at_newsg4.svr.pol.co.uk...

> > I'd have to question why anybody who can guarantee that no further
> > transactions are taking place because 'user activity stops at close of
> > business hours' is bothering to take hot backups in the first place!
>
> well I want the database up so as to do batch work, i have a system where
> the days statement runs are done over night and the printing process of
the
> statements need to do a select on the database, also i refresh a business
> reporting database over night by selecting from said database... thus i
need
> the database up and need to back it up!
>
> the users can be stopped from changing any data after a certain time via
the
> application untill the next day...
>
> > In principle, if the database is hot, then you cannot "know" that the
log
> > contains the last transaction,
>
> yes you can... well maybe not the last tx but the last business user tx at
> least
>
>
> > because anybody suffering from a bout of
> > insomnia could log on and start doing DML.
>
> as above this is handled and enforced via the app
>
> > You might get away with it 99.9% of the time, but the principle is not
> sound.
>
> disagree...
>
>
> --
> Regards,
>
> Daniel.
>
>
> "Howard J. Rogers" <dba_at_hjrdba.com> wrote in message
> news:a7o81g$389$1_at_lust.ihug.co.nz...
> > Comment below.
> > HJR
> > --
> > ------------------------------------------
> > Resources for Oracle : www.hjrdba.com
> > ============================
> >
> > "daniel" <test_at_test.com> wrote in message
> > news:a7o6m5$nap$1_at_news7.svr.pol.co.uk...
> > > > Lots of people do it, and I've never really understood why.
> > >
> > > maybe they want to differentiate between the two types of redo that
can
> be
> > > written? just a guess...
> > >
> > > however sometimes in a system where user activity stops at close of
> > business
> > > hours then it would be really easy to recover to that log seq num
> knowing
> > it
> > > contains the last business tx's, but no overnight stuff like batch etc
> > > etc...
> >
> > I'd have to question why anybody who can guarantee that no further
> > transactions are taking place because 'user activity stops at close of
> > business hours' is bothering to take hot backups in the first place!
> >
> > In principle, if the database is hot, then you cannot "know" that the
log
> > contains the last transaction, because anybody suffering from a bout of
> > insomnia could log on and start doing DML. You might get away with it
> 99.9%
> > of the time, but the principle is not sound.
> >
> > HJR
> >
> >
> > >
> > > it is also very handy to refresh a test system from this without the
> need
> > > for time based or scn based recovery, again knowing its a capture of
the
> > > data at close of play.
> > >
> > > > The other thing that mystifies me about forcing a log switch in
order
> to
> > > get
> > > > an archive of the current log is that it only makes another log the
> > > current
> > > > log.
> > >
> > > agreed, but it can be used as an easy marker and/or punting it off to
> the
> > > standby server if thats the setup, there's a stack more reasons why
you
> > may
> > > want to do it but i won't bore you with them...
> > >
> > > :O)
> > >
> > >
> > > --
> > > Regards,
> > >
> > > Daniel.
> > >
> > >
> > > "Howard J. Rogers" <dba_at_hjrdba.com> wrote in message
> > > news:a7o4ju$vfn$1_at_lust.ihug.co.nz...
> > > > If you're doing O/S online backups, then there's no need for a
> > preliminary
> > > > checkpoint, because that's precisely what the 'begin backup' command
> > > forces
> > > > (at least for the datafiles of the tablespace involved).
> > > >
> > > > The one about a preliminary log switch (presumably because you are
> about
> > > to
> > > > copy the archives) is always a curious one. Lots of people do it,
and
> > > I've
> > > > never really understood why. Usually the justification is that
> without
> > > it,
> > > > you are short of the current redo log, so you may lose data.
There's
> > > > something to that, I suppose. But it's also usually (so I find)
> because
> > > > people view a backup as an isolated event, whereas the truth of the
> > matter
> > > > is, of course, that what you don't backup tonight you will backup
> > > tomorrow.
> > > > So if there's a bit of current redo left unbacked up, who cares??
> > You'll
> > > > get it when tomorrow's backup is performed, and in the meantime
> there's
> > no
> > > > possible risk of data loss because you've multiplexed your online
redo
> > > logs,
> > > > and then mirrored them with hardware RAID. Haven't you?? ;-)
> > > >
> > > > The other thing that mystifies me about forcing a log switch in
order
> to
> > > get
> > > > an archive of the current log is that it only makes another log the
> > > current
> > > > log. So you can *never* really be completely and utterly up-to-date
> > with
> > > > redo copies, unless you stop all your users doing things: there'll
> > always
> > > be
> > > > a new piece of current redo which you haven't backed up today.
Hence
> > cold
> > > > backups, of course.
> > > >
> > > > That said, a log switch does no real harm -except induce a
> > > > performance-hitting checkpoint.
> > > >
> > > > Regards
> > > > HJR
> > > > --
> > > > ------------------------------------------
> > > > Resources for Oracle : www.hjrdba.com
> > > > ============================
> > > >
> > > > "Glen A Stromquist" <gstromquist_at_nospamyahoo.com> wrote in message
> > > > news:jpMn8.11561$EV.366849_at_news1.telusplanet.net...
> > > > > In my online backup scripts I don't do a logfile switch or force a
> > > > > checkpoint before copying the datafiles.
> > > > >
> > > > > Is this recommended by Oracle?
> > > > >
> > > > > I'm wondering if I overlooked something when writing my scripts, I
> > have
> > > > used
> > > > > my online backups on occasion to create a clone db, so I know they
> > > "work"
> > > > > the way I'm doing it now, but I guess it can't hurt to build in a
> > > logfile
> > > > > switch and/or force a checkpoint as part of the script as well.
> > > > >
> > > > > Curious to hear what others do regarding this....
> > > > >
> > > > >
> > > > > cheers!
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
Received on Mon Mar 25 2002 - 20:53:17 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US