Issue

• For the past few weeks postgresql on our Satellite 5.6 system has been showing nearly %100 CPU usage. Monitoring postgresql activitity shows that Satellite seems to be repeatedly updating a single table by repeating the commands:

update rhnPushClient set state_id = 2,last_ping_time = NULL,next_action_time = NULL where jabber_id = E'osad-54afab6579@satellite.example.com/osad'

and

update rhnPushClient set state_id = 2,last_ping_time = NULL,next_action_time = NULL where jabber_id = E'osad-20d699a348@satellite.example.com/osad'

• Querying the rhnPushClient table for server_id's with those jabber_id values returns a different set of server_id's every time the query is run, so whatever is happening is happening fast and repeatedly and doesn't seem to be limited to a single client.

• Check-in interval is set to the default 240 minutes on all clients.

• The Satellite server has been updated to the latest version of the application and DB schema (5.6.0.19-3.)

• Is there a way to stop the constant table updates?