You are viewing a plain text version of this content. The canonical link for it is here.

Posted to users@jackrabbit.apache.org by Skripnikov <ma...@voliacable.com> on 2007/01/31 21:03:37 UTC

Jackrabbit SQL Queries

Hi!
We making research on Jackrabbit to explore possibility to use it for our
Application.
But then we imported some data to jackrabbit (about 50000) nodes it becomes
significantly slow.
To continue our exploration we would like to explore sql queries jackrabbit
makes to DB. 
Is there any way to make jackrabbit log executed queries,their execution
times and result size?
-- 
View this message in context: http://www.nabble.com/Jackrabbit-SQL-Queries-tf3150952.html#a8736653
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.

Re: Jackrabbit SQL Queries

Posted by Skripnikov <ma...@voliacable.com>.

Thanks, checked it out, add try{}finally{} to DatabasePersistenceManager and
DatabaseFileSystem, thanks to jackrabbit team  - the code is much better
than 1.1 - only two places in 1.2 instead of multiple ones in 1.1
 

Nicolas Modrzyk-2 wrote:
> 
> Hi,
> 
>> Is there any way to make jackrabbit log executed queries,their  
>> execution
>> times and result size?
> 
> AFAIK, this is not done in the default jackrabbit code, but it would  
> be really easy to:
> - checkout the code
> - put a simple aspect around the class/methods executing the SQL queries
> - compile and change the jar file
> 
> Result size, from my experience on C-JDBC was a killer, especially as  
> some data and metadata are streamed on demand when you actually  
> access the result.
> So it may not be a valuable indicator.
> 
> Regards,
> 
> Niko,
> 
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Jackrabbit-SQL-Queries-tf3150952.html#a8740321
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.

Re: Jackrabbit SQL Queries

Posted by Nicolas Modrzyk <ni...@macnica.com>.

Hi,

> Is there any way to make jackrabbit log executed queries,their  
> execution
> times and result size?

AFAIK, this is not done in the default jackrabbit code, but it would  
be really easy to:
- checkout the code
- put a simple aspect around the class/methods executing the SQL queries
- compile and change the jar file

Result size, from my experience on C-JDBC was a killer, especially as  
some data and metadata are streamed on demand when you actually  
access the result.
So it may not be a valuable indicator.

Regards,

Niko,