Comment 2 for bug 1901191

Here is a SQL query that you can use to find the queries run by the storage drone that is looping. You will want to scan the output for look ups on the actor.hours_of_operation and actor.org_unit_closed tables. Note that you will have to run the SQL several times to find any/all processes. These queries are short-lived and there is a decent chance that you will miss some if you only run the query once or twice. If you run this on a database with a lot going on, i.e. a production database, then I would suggest running it several times before executing any of the following steps to make sure that you are getting processes stuck in a loop.

You can find the process ID of the storage drone(s) that is/are looping over the queries by copying the client port column value, logging into the machine with the client IP address and running the following command:

lsof -iTCP | grep {client port}

Where {client port} is the port number reported by the query.

The second column of the output from lsof is the drone's process ID (PID). You can terminate the drone with the following command:

kill {PID}

Where {PID} is, of course, the numeric process ID reported by lsof.