Skip site navigation (1) Skip section navigation (2)

Re: [HACKERS] Another nasty cache problem

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
Cc: pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] Another nasty cache problem
Date: 2000-01-30 21:54:14
Message-ID: 26591.949269254@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-hackers
Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Now, if we did proper locking, no SI message could arrive for such an
> entry.

> My assumption is that these are mostly system cache entries, and they
> rarely change, right?  If someone is operating on a table that gets an
> SI entry, odds are that later on the system will fail because the table
> is changed in some way, right?

If the tuple is actually *changed* then that's true (and locking should
have prevented it anyway).  But we also issue cache flushes against
whole system tables in order to handle VACUUM of a system table.  There,
the only thing that's actually been modified is the tuple's physical
location (ctid).  We don't want to blow away transactions that are just
looking at cache entries when a VACUUM happens.

Perhaps the caches shouldn't store ctid?  Not sure.

			regards, tom lane

In response to

Responses

pgsql-hackers by date

Next:From: Bruce MomjianDate: 2000-01-30 21:56:14
Subject: Re: [HACKERS] Another nasty cache problem
Previous:From: joDate: 2000-01-30 20:29:45
Subject: corel9

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group