I have created and am using an SQL Agent for the first time. I have searched google but either the info isnot there or I am not getting the keywords correct (SQL agent + back log/callstack overflow etc !).
My agent reads in and does some processing on a (variable) volume of data. I observed an error recently where i deadlocked a database i think due to an issue with the SQL agent and it started me thnking about the following: If I set the schedule frequency to be shorter than the duration that it took to compete the action that is performed by the agent, such that the 2nd action is scheduled while the first is still processing ...and so on. Will this generate an unrecoverable situation by accumulating jobs that it will never have time to process or if the agent starts while a previous job is still running does it just abandon the attempt and check again at the next scheduled interval.
Obviously one wouldnt want to do the above on purpose but it just got me thinking about whether i need to handle for the above possibility.