Re: conflict txns in serialization isolation

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Yi LIN <ylin30(at)cs(dot)mcgill(dot)ca>
Cc: pgsql-jdbc(at)postgresql(dot)org
Subject: Re: conflict txns in serialization isolation
Date: 2004-07-23 21:43:43
Message-ID: 28391.1090619023@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-jdbc

Yi LIN <ylin30(at)cs(dot)mcgill(dot)ca> writes:
> Read-committed is different from Snapshot in that
> Read-Committed txn will read data committed before read operation starts
> but snapshot txn will read data committed before the whole txn starts.

That's a true statement for SELECT, but a read-committed UPDATE will
in fact find and update the latest version of the row; it won't cause
changes to be lost, even if they occurred after the UPDATE statement as
a whole started. The same is true of SELECT FOR UPDATE.

I'm not sure how that relates to your previous observations though.
Can you duplicate the apparent misbehavior in psql? I can't.

regards, tom lane

In response to

Responses

Browse pgsql-jdbc by date

  From Date Subject
Next Message Oliver Jowett 2004-07-24 00:11:34 Re: JDBC memory usage
Previous Message Yi LIN 2004-07-23 21:19:59 Re: conflict txns in serialization isolation