Re: Interesting tight loop - Mailing list pgsql-hackers

From Theo Schlossnagle
Subject Re: Interesting tight loop
Date
Msg-id 301C0D96-9C04-4E0A-A716-85091D953A7E@omniti.com
Whole thread Raw
In response to Re: Interesting tight loop  (Gregory Stark <stark@enterprisedb.com>)
Responses Re: Interesting tight loop  (Gregory Stark <stark@enterprisedb.com>)
List pgsql-hackers
On Sep 14, 2006, at 7:03 AM, Gregory Stark wrote:

> Theo Schlossnagle <jesus@omniti.com> writes:
>
>> But the interesting thing is that there were 4.6 million elements  
>> in the
>> s->childXids list.  Which is why it took so damn long.  I can't   
>> quite figure
>> out how I induced this state.  It is an OLAP server with  about 10-20
>> connection that run "long" queries (from 5 seconds to 24  hours).
>
> Wow, I think that means you've had 4.6 million subtransactions  
> within the
> current transaction. Is this a PL/PgSQL function or a script that uses
> SAVEPOINT a lot?

We don't use savepoint's too much.  Maybe one or two across out 1k or  
so pl/pgsql procs.

We use dbi-link which is plperl.  Perhaps that is somehow creating  
subtransactions?

// Theo Schlossnagle
// CTO -- http://www.omniti.com/~jesus/
// OmniTI Computer Consulting, Inc. -- http://www.omniti.com/




pgsql-hackers by date:

Previous
From: Gregory Stark
Date:
Subject: Re: CSStorm occurred again by postgreSQL8.2
Next
From: Simon Riggs
Date:
Subject: Re: Draft release notes