Re: tracking down hidden SQL???
From: Daniel Fink <danielfink.work_at_gmail.com>
Date: Wed, 16 Oct 2013 13:16:37 -0600
Message-Id: <C994D395-4EA2-4077-996B-C45D054BE675_at_gmail.com>
You might try using the procedure dbms_workload_repository.add_colored_sql to try to capture it for snapshots.
>
> Unfortunately, the SQL_ID in question isn't present. It was an excellent
> suggestion, however.
>
> SQL> select count(*) from dba_hist_sqlstat
> 2 where sql_id = '0p8ucmfzpm28r';
>
> COUNT(*)
> ----------
> 0
>
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
Date: Wed, 16 Oct 2013 13:16:37 -0600
Message-Id: <C994D395-4EA2-4077-996B-C45D054BE675_at_gmail.com>
You might try using the procedure dbms_workload_repository.add_colored_sql to try to capture it for snapshots.
On Oct 16, 2013, at 12:58 PM, Adric Norris <landstander668_at_gmail.com> wrote:
> On Wed, Oct 16, 2013 at 1:51 PM, Chris Taylor <
> christopherdtaylor1994_at_gmail.com> wrote:
>> You could check DBA_HIST_SQLSTAT which is a running collection of SQL >> statement stats (just to see if it ever got captured). If so, from there >> you could grab the force_matching_signature and find any SQLs that share >> the same execution plan. Doubtful its in there, but worth a look. >>
>
> Unfortunately, the SQL_ID in question isn't present. It was an excellent
> suggestion, however.
>
> SQL> select count(*) from dba_hist_sqlstat
> 2 where sql_id = '0p8ucmfzpm28r';
>
> COUNT(*)
> ----------
> 0
>
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Oct 16 2013 - 21:16:37 CEST