What to call this type of timer?

[mike_r]

I'm adding a new safety mechanism to an RTOS. It's a count-down timer which is decremented by the kernel, and when it reaches zero it signals that a task should exit.

For example, a task that normally takes 1000 ticks to complete could allocate one of these timers with a value of 2000. Then, when the value reaches zero, the task knows it has been running twice as long as normal (maybe because some hardware isn't responding), so it should exit.

Is there a usual name for this? If not, can someone suggest a name?

 

[sau_sol]

it seems similar to WATCHDOGTIMER used in micro controllers to detect and restart a controller software problem
if the controller hangs somewhere and it does not responds to even watchdog it will be reset by watchdog timer

 

[mike_r]

Actually it is quite a lot like a watchdog. Is there another name for it? This system already has a hardware watchdog, so it feels like that would be using one word for different things.

 

[sau_sol]

up to what I know there is no fixed name for this type of timer

so you can call it by some other name like:--

process monitoring timer
process time limiter
resource manager timer

 

[mike_r]

Thanks for your ideas Sol. I've been calling this "Egg_Timer", but the metaphor's not perfect. I like your "resource manager timer".