Re: [HACKERS] Priorities for 6.6 - Mailing list pgsql-hackers

From ZEUGSWETTER Andreas IZ5
Subject Re: [HACKERS] Priorities for 6.6
Date
Msg-id 219F68D65015D011A8E000006F8590C60267B38C@sdexcsrv1.f000.d0188.sd.spardat.at
Whole thread Raw
List pgsql-hackers
> >By the way, may I ask more question regarding Oracle? You mentioned
> >the magic of no-fsync in Oracle is actually a bug. Ok, I understand. I
> >also heard that Oracle does some kind of redo-log bufferings. Does
> >this mean certain committed data might be lost if the system crashed
> >before the buffered data is written into the disk?
> 
Yes, you might loose a transaction that has been reported committed to the
client.
But, it guarantees that every transaction is eighter committed, or rolled
back as a 
whole. Thus leaving the database in a consistent state. We have a lot of
applications
where this is acceptable, and others where this is not. It is the ability to
choose 
(in Informix buffered versus unbuffered logging) that makes us happy. 

Andreas


pgsql-hackers by date:

Previous
From: ZEUGSWETTER Andreas IZ5
Date:
Subject: Re: [HACKERS] Bizarre coding in _bt_binsrch
Next
From: Vadim Mikheev
Date:
Subject: Re: [HACKERS] Bizarre coding in _bt_binsrch