On Friday 05 December 2003 16:51, Steve Wampler wrote:
> I need some help tracking down a sudden, massive slowdown
> in inserts in one of our databases.
>
> PG: 7.2.3 (RedHat 8.0)
>
> Background. We currently run nearly identical systems
> at two sites: Site A is a 'lab' site used for development,
> Site B is a production site.
>
> The databases in question have identical structure:
>
> A simple table with 4 columns with a trigger function
> on inserts (which checks to see if the entry already
> exists, and if so, changes the insert into an update...)
> A simple view with 4 columns into the above table.
>
> All access is through jdbc (JDK 1.3.1, jdbc 7.1-1.3),
> postgresql.conf's are identical.
>
> The two sites were performing at comparable speeds until
> a few days ago, when we deleted several million records
> from each database and then did a vacuum full; analyze
> on both. Now inserts at Site B are several orders of
> magnitude slower than at Site A. The odd thing is that
> Site B's DB now has only 60,000 records while Site A's is
> up around 3 million. Inserts at A average 63ms, inserts
> at B are now up at 4.5 seconds!
>
> EXPLAIN doesn't show any difference between the two.
>
> Can someone suggest ways to track this down? I don't know
> much about postgresql internals/configuration.
>
What does explain analyze show for the insert query?
Are there FK and/or Indexes involved here? Did you you reindex?
A vacuum verbose could give you a good indication if you need to reindex,
compare the # of pages in the index with the # in the table.
Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
