Re: point in time recovery and moving datafiles online - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: point in time recovery and moving datafiles online
Date
Msg-id 200202221830.g1MIU7a02183@candle.pha.pa.us
Whole thread Raw
In response to Re: point in time recovery and moving datafiles online  ("Zeugswetter Andreas SB SD" <ZeugswetterA@spardat.at>)
List pgsql-hackers
Zeugswetter Andreas SB SD wrote:
> > > Yes. But I think you might have to avoid "vacuum full" during data file backup.
> > 
> > Why?  If vacuum is unsafe in this scenario, wouldn't it also be unsafe
> > in event of a system crash?
> 
> I was not sure we are 100% crash safe during "vacuum full".
> But yes, now recalling your last remark on a crash during vacuum full
> I guess we should be.
> 
> > I do believe that vacuum should (but presently does not) emit a WAL
> > record showing its truncation of the file, so that the equivalent
> > truncation can be repeated during replay.
> 
> > However, this is needed in any case.
> 
> I see, but it has only the effect of not freeing the space to the OS, 
> so it is not really a bug ? Next vacuum will do it anyway. 

One idea is that replication can give is point-in-time recovery.  A
recovery is nothing more than a replication slave taking the last full
backup and reading the replication traffic to get itself up to date.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 


pgsql-hackers by date:

Previous
From: Oleg Bartunov
Date:
Subject: Re: problem with restore of functions
Next
From: Bruce Momjian
Date:
Subject: Going through email