I’ve noticed when I start a customer’s system that it was taking a long time to settle down to a steady state. I was able to replicate this behaviour by stopping and then restarting FactorySQL. When I did this, I noticed the following:
- The processor usage of MySQL would jump from 0-3% of processor usage to 51-54%.
- Calculated SQLTag points would not have a valid value.
- Any standing alarms were only added to the alarm list very slowly.
- The system responded very sluggishly.
Once all alarms were added to the alarm list (which took 4 or 5 minutes), processor usage would drop back down to normal, whether or not the alarms were acknowledged.
It would appear that the system is having problems coping with standing alarms when the FactorySQL service starts. Any thoughts?