Munin has chart 'interrupts and context switches on the system'.

As far as this is a monitoring tool I assume these values are important for the server performance.

So, the question is: how can I know for each particular server the values are OK or too high?

I assume any Linux at least

  • 1
    "How long is a piece of string?" – lcd047 May 6 '15 at 18:10
  • 2
    What @lcd047 is trying to say is: it all depends on your system and your needs. If it is something like an automatic defibrillator than you want context switches and interrupts so a patient doesn't die. – SailorCire May 6 '15 at 18:30

Monitor the trends over time and look for anomalies. The "normal" values differ depending on the type of application load and in turn, what the application does on a regular basis.

|improve this answer|||||
  • 1
    And just for example i have linux machines with jboss and <1000 CS/s But other have >50k CS/s and for both of them this is normal – Romeo Ninov May 6 '15 at 18:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.