On Monday 16 October 2006 18:01, Mike wrote: > The server is rather new. 2GB of memory with a top grade Opteron (not > sure which). Once again, the server is running with low load for most > of the day, and as much as I want to, I have hard time believing its > the server that is to blame.
But it just might be. it is possible to get into a catastrophic behaviour where the load get sufficient that pages start to take a little longer to service. This means that incoming requests spawn new process because the normal pool of processes is busy. This puts more load on the server and so on. Mostly the bottleneck in these situation is the disk system not the processor. As you are also running at high memory load there will be a fair amount of swap activity on the disk, leading to the database taking even longer to process requests, leading to more processes being spawned. I think you get the picture. Just because it is a shiny new server with good processors does not always mean it is up to the job. The questions about the server spec, disks in particular are therefore valid. Just take a look on the postgres-performance mailing list where you will see dozens of threads about this kind of performance problem. You won't always see it at peak loads if the peak dies down quickly enough for the disks to start serving the database in good time again. Your problem may or may not have anything to do with the above if you are leaking database connections, but it's worth looking into. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-users -~----------~----~----~----~------~----~------~--~---
