The reason renicing doesn't help is because what you are fighting really is iowait. You have sent an io request out and are selecting on the io system awaiting when you can write again or read again.
What you may need to do is code review what you can of the hog app to see why it is so horribly io intense. Like are they reading large sections char by char instead of large chunks and buffering in the app. Another option might be to segregate the io workload. Such that your hog doesn't starve the main drive system. Part of not being able to log in is not getting a slice to load the bins and configs. tocsa <[email protected]> wrote: >Oh, I reniced the hog processes +5 (which yields 25 priority) from the >beginning. That didn't help. Or just partially: when the load was over >22 the website was still responsive because of that. With 76 load >however the spawned apache threads for handling the HTTP requests were >just starving. There were 150 of them. Also there were 240 chron jobs >related to drupal (49 drupal threads), and we don't even use drupal -- Sent from my Android phone with K-9 Mail. Please excuse my brevity. -- You received this message because you are subscribed to the Google Groups "NLUG" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/nlug-talk?hl=en
