I modified my tomcat/config/context.xml file as follows:
Code: Select all
<Resource name="redshift" auth="Container" type="javax.sql.DataSource"
initialSize="5" maxActive="5" minIdle="5" maxIdle="5" maxWait="1300000" removeAbandoned="true" removeAbandonedTimeout="1500"
username="myusername" password="mypassword" driverClassName="org.postgresql.Driver"
Restarted tomcat, and create a new db connection via the Designer. Specified my JNDI resource as the one above (java:comp/env/redshift), and fire up my graphs.
The graphs won't fail running, but in my redshift server I never see the sql calls that they are supposed to make. If I do the calls via a non JNDI connection I can successfully run the queries, so there is got to be some miss configuration I have somewhere for JNDI.
Here's the output of my logs:
2015-09-10 17:28:56,756 INFO 1388117 [WatchDog_1388117] DBConnection driver[null]:jndi[java:comp/env/redshift]:url[jdbc:postgresql://myserver.fdasfafgssfgfdh.us-west-2.redshift.amazonaws.com:5439/mydatabase]:user[myusername] ... OK
So as far as cloverETL, it seem to be connecting (OK), but I don't see the sql hitting my redshift server, and the graph just stays there forever, until the timeout is reached, and then it fails.
My questions are:
1) Notice the "driver[null]" part on logs above, should I be concern about that? I am sort of new to the Tomcat env, so not sure if a driver is missing, then again, if I do a postgresql connection that is not JNDI, things work
2) What would be a good way to troubleshoot what is happening while the graph is sitting idle? I am looking at the logs provided by cloverETL itself, but they just also sit idle in there, how can I tell what happens, or where the app failed to send the sql to the redshift server?
Any help will be appreciated, thank you!