IQSS / glassfish-query-counter

Short query counting script.

Glassfish Query Counter

Count queries printed to the Glassfish logs. This basically:

1. Takes the last 'n' lines from the glassfish log file, marked by a logging statement in Java
2. Writes the lines to a file
3. Runs them though the python collections.Counter.

Tested on python 2.7.

I. Set up Glassfish to log queries

(A) Edit the persistence.xml file

1. Open the persistence.xml file
2. Add this line to the end of the ````` list:

<property name="eclipselink.logging.level.sql" value="FINE"/>

• Example:

<properties>
    <!--property name="toplink.logging.level" value="FINE"/-->
    <property name="eclipselink.weaving" value="false"/>
    <property name="eclipselink.ddl-generation" value="create-tables"/>
    <property name="eclipselink.cache.shared.default" value="false"/>
    <!-- The following property allows primary keys of 0 -->
    <property name="eclipselink.id-validation" value="NULL"/>
    <property name="eclipselink.logging.level.sql" value="FINE"/>
</properties>

(B) For the pages you are interested in, add a logging line

1. Add delimiters to the pages you would like to check.
2. Example:
   1. Edit the DatasetPage.java
   2. Add this line to the top of your init(...) method:

System.out.println("_YE_OLDE_QUERY_COUNTER_");

II. Update the python settings to point to your Glassfish Log

1. Copy the file scripts/settings_template.json to scripts/settings.json
2. Update scripts/settings.json with the path to your Glassfish log file
3. Example:

{ "GLASSFISH_LOG_FILE_PATH" : "/Users/rp/Documents/iqss-git/glassfish4.1/glassfish/domains/domain1/logs/server.log" }

III. Run It

1. Go to the page you would like to check: e.g. http://localhost:8080/dataset.xhtml?id=3&versionId=15
   • The log should contain SQL statements from the page load
2. Run the python script, with the name of an output file:

> cd 'glassfish-query-counter/scripts'
> python count_queries.py dataset-page-check

• This does the following:
  1. Creates a text file in: query_lists/dataset-page-check.txt (the ".txt" is appended automatically)
  2. Creates a .csv file with query counts: query_counts/dataset-page-check.csv
  3. Prints output to the terminal