Hi, We have an alert which is defined for high CPU utilization. If CPU corsses 90 it will alert the windows team. But sometimes, by the time they log in the CPU would have goen back to normal state. Ideally we would like to get the process which would have caused this CPU spike. Any idea on how we can do this? We do not want to create another monitor/monitors which can monitor the process utilization everytime. Rather than that, we just need to monitor and pass the info to ticket tool (or even write to a log file) as in when an alert condition occurs. Any thoughts around this would be appreciated. Dilu
Solved! Go to Solution.
If these are Windows boxes then you should already have a sample alert named something like "Alert on High CPU - top 10 processes". This alert will not only tell you when you have a high CPU, it will grab the top 10 processes at the time of the alert and stuff them into an email message (or logfile, or whatever).
I extended that alert to make the trigger more complex (not just high CPU, but high CPU *and* more processes in the queue than CPU's to handle them) and described the whole thing here:
You may also find the following sections from the Administrator Guide helpful.
If these are Windows boxes then you should already have a sample alert named something like "Alert on High CPU - top 10 processes". This alert will not only tell you when you have a high CPU, it will grab the top 10 processes at the time of the alert and stuff them into an email message (or logfile, or whatever).
I extended that alert to make the trigger more complex (not just high CPU, but high CPU *and* more processes in the queue than CPU's to handle them) and described the whole thing here:
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process.