Thread: Problems with autovacuum and vacuum

Hello, <br /> <br />Last week I had a serious problem with my PostgreSQL database. My autovacuum is OFF, but in
Septemberit started to prevent the transaction wraparoud; however last week the following message appeared continuously
inmy log: <br /> <br />WARNING: database "production" must be vacuumed within 4827083 transactions <br />HINT: To avoid
adatabase shutdown, execute a full-database VACUUM in "production". <br /> <br />This message appeared for five to six
hours;after that, the message disappeared from log. Any idea about what could have happened? <br /> <br />Every day the
vacuumis executed on some tables; and on Sundays it's executed on all tables. But as the autovacuum is running since
September,and it runs for a long time, the vacuum was blocked because autovacuum had been running on the same table.
Howshould I procede in this case?<br /><br />The table where the autovacuum is running and where the vacuum was blocked
hasbillion of rows. <br /> <br />I'm using the PostgreSQL 8.3.8 <br /> <br />The configuration of the vacuum parameters
are:<br /> <br />vacuum_cost_limit = 200 <br />vacuum_cost_delay = 0 <br />vacuum_freeze_min_age = 100000000 <br
/>autovacuum= off <br />autovacuum_freeze_max_age = 200000000<br /><br />Regards,<br clear="all" /><br />João Paulo<br
/><br/>-- <br />JotaComm<br /><a href="http://jotacomm.wordpress.com">http://jotacomm.wordpress.com</a><br /> 

On Thu, Dec 30, 2010 at 12:56 PM, JotaComm <jota.comm@gmail.com> wrote:
> Last week I had a serious problem with my PostgreSQL database. My autovacuum
> is OFF, but in September it started to prevent the transaction wraparoud;
> however last week the following message appeared continuously in my log:
>
> WARNING: database "production" must be vacuumed within 4827083 transactions
> HINT: To avoid a database shutdown, execute a full-database VACUUM in
> "production".
>
> This message appeared for five to six hours; after that, the message
> disappeared from log. Any idea about what could have happened?

I'm thinking that autovacuum kicked into gear to prevent transaction
wraparound.  Once it did enough work to stave off disaster, the
warning messages stopped appearing in the log.

> Every day the vacuum is executed on some tables; and on Sundays it's
> executed on all tables. But as the autovacuum is running since September,
> and it runs for a long time, the vacuum was blocked because autovacuum had
> been running on the same table. How should I procede in this case?

I guess the obvious thing to do would be to turn on autovacuum and
forget about manual vacuums.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company