Author Name: Bob Gobeille
Original Redmine Issue: 561, http://www.fossology.org/issues/561
Original Date: 2012/01/26
Original Assignee: Alex Norton
When an agent exits with a non-zero status, that exit code should be written to the agent log. It doesn't look like this is happening. Some agents, e.g. nomos, use very specific exit codes to help diagnose an abnormal exit.
Also, it would be extremely helpful if the scheduler could log the signal that killed the agent. For example, to know that an agent segv'd would be very helpful. What I'm thinking of here is the type of information returned by the wait status macros (man waitpid).
Author Name: Bob Gobeille Original Redmine Issue: 561, http://www.fossology.org/issues/561 Original Date: 2012/01/26 Original Assignee: Alex Norton
When an agent exits with a non-zero status, that exit code should be written to the agent log. It doesn't look like this is happening. Some agents, e.g. nomos, use very specific exit codes to help diagnose an abnormal exit.
Also, it would be extremely helpful if the scheduler could log the signal that killed the agent. For example, to know that an agent segv'd would be very helpful. What I'm thinking of here is the type of information returned by the wait status macros (man waitpid).