It looks like the indexes are blown Which may indicate some corruption in
the pg_indexes system catalog table?
Is there a way to check the integrity of pg_indexes for possible page
corruption? Nothing is being reported in the postgresql log file that would
indicate there is a corruption i
In response to Keaton Adams <[EMAIL PROTECTED]>:
> Good point. But
>
> This is a database used to capture logged information, such as
> success/failure of an operation. Daily tables are rolled up to weekly
> tables, weekly tables are rolled up to monthly tables. All logged
> activity are d
Good point. But
This is a database used to capture logged information, such as success/failure
of an operation. Daily tables are rolled up to weekly tables, weekly tables
are rolled up to monthly tables. All logged activity are done by inserts,
never any updates or deletes, to avoid havi
On Tue, Jul 8, 2008 at 12:06 AM, Keaton Adams <[EMAIL PROTECTED]> wrote:
> An analyze is run on the tables every day (even several times a day because
> they are updated very frequently) and a vacuum analyze is run on the
> weekends. I also tried to run an analyze specifically on the customer_id
>
PG: PostgreSQL 8.1.4
OS: RHEL 4.x
I have a set of queries on a production server that have been running fine for
the past few months but as of last Friday started performing poorly. I have
isolated the problem down to a particular part that is common to all queries
involved and have provided
