On Fri, 22 Mar 2002, TORRESANI, Roberto wrote: > The test I took on mon, make me think as alertafter doesn't work with trap. With >normal monitor it works normally. > Is there a bug on alertafter+traps or is my misunderstanding of alertafter+traps >behavior?
unfortunately the alert throttling code is not 100% independent of the trap/monitor handling logic. for example, some of the squelch logic is still in proc_cleanup (which handles the exit condition of a monitor), and some of this logic is duplicated in handle_trap and handle_trap_timeout. this needs to be unified. i made a first-pass at this in 0.99.1, where i pulled a bunch of the logic out of proc_cleanup and put it into do_alert. however, there's more work to be done yet in handle_trap and handle_trap_timeout to make it all consistent.
