Currently tool_task_adhocqueue check simply gets the oldest unprocessed task and alarms if it over the threshold value.
CRITICAL: Ad hoc task queue: Oldest unprocessed task is 15 hours 50 mins, which is more than 4 hours
In real life this may not actually mean an issue. On big sites where long running autobackup tasks can be stacked in a queue, but eventually get processed which takes time this generates a false alarm.
Should we instead look at the total number of tasks in the queue?
Or maybe we should check if the specific old task has a queue and the queue is getting processed?
Currently tool_task_adhocqueue check simply gets the oldest unprocessed task and alarms if it over the threshold value.
CRITICAL: Ad hoc task queue: Oldest unprocessed task is 15 hours 50 mins, which is more than 4 hoursIn real life this may not actually mean an issue. On big sites where long running autobackup tasks can be stacked in a queue, but eventually get processed which takes time this generates a false alarm.
Should we instead look at the total number of tasks in the queue?
Or maybe we should check if the specific old task has a queue and the queue is getting processed?