I wrote a small program to insert the SQL parameters into the SQL so log
file has one SQL statement per line w/ no parameterization.
What are the details of how you set up your JDBC sampler to access the
SQL coming from the CSV dataset?
Did you get parameterized SQL to work when both the SQL statement and
the parameters come from the CSV dataset?
Shane Petroff wrote:
Jeff Baird wrote:
Is this the best way to feed SQL
statements to a database to performance test the database?
It's about the only way you can work with large sets of SQL, yes.
Does anyone have any examples that show how to do this?
The basic layout I use is
ThreadGroup
JDBC Connection Config
CSV Dataset Config
JDBC Sampler
View Results Tree (this one will give you detailed info on
exceptions)
Graph Results
Summary Report
Is your JDBC driver accessible? Either in jmeter's lib directory or
added directly to the test plan. The only issues I remember having is
parameterizing the sql so that caching on the database doesn't skew
the stats, and reliably cleaning up the inserted data at the end of
the script. One can work around both of these issues, but the only
ongoing issue I have with JDBC testing is keeping the throughput high
enough to actually make the database 'sweat'. I was hoping to
overwhelm it, but have never come close.
--
Jeff Baird x40647
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
