Re: Multicolumn index corruption on 8.4 beta 2 - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Multicolumn index corruption on 8.4 beta 2
Date
Msg-id 25575.1244573459@sss.pgh.pa.us
Whole thread Raw
In response to Re: Multicolumn index corruption on 8.4 beta 2  (Josh Berkus <josh@agliodbs.com>)
List pgsql-hackers
Josh Berkus <josh@agliodbs.com> writes:
> It doesn't.  But what I don't trust, and the *first* place I'd look for 
> problems, is whether the OS flushes *all* dirty buffers to disk in the 
> event the application gets killed.

Why wouldn't you trust it?  The sort of thing you seem to be thinking
about would require tracking which process(es) wrote each dirty buffer
and then going back and dropping selected dirty buffers when a process
exits abnormally.  I can hardly imagine any OS wishing to do that.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Not quite a security hole in internal_in
Next
From: "Kolb, Harald (NSN - DE/Munich)"
Date:
Subject: Re: postmaster recovery and automatic restart suppression