Too many processes
I'm not a newbie but this is definitely a newbie question...
Quite simply, I have a host that has too many processes on it. Normally I'd just cull unnecessary stuff but the crux of it appears to be kernel related. Here's a list: Code:
[root@cbcvm ~]# ps faux How would I clean up those kernel processes without restarting the machine? It's production. Cheers! |
Why do you think there are too many processes running?
My Debian box runs +/- 375 processes at any given moment and my LFS box +/- 250. I also don't see any defunct ("zombie") processes. |
This is a knee-jerk reaction to my Nagios monitor that just started showing 350+ processes today and hence going critical. I could just up the limit (it is a VM host after all).
Is it a problem having all those procs? |
Quote:
Why is nagios set to alert you when the amount of processes reaches 350+ ? I can only assume this is done for a specific reason and if so you might want to ask the person that set it up why this is so. If a default setting is used then I would suggest to slowly increase this setting to a more suitable/workable setting. But like I stated before: 350+ running processes isn't (normally) problematic. |
I think by default you can have up to 65535 active and zombie processes unless you build the kernel with a different default.
If you're concerned about the system integrity, check free memory using free(1) and run top(1) to see what's going on, whether the CPU and memory are being used up. I really don't know how much metrics one can put into the amount of memory left, but I'd say that if you persistently used more than half of your memory on an everyday basis, I'd get more. If the CPU is loaded a lot however no one process or group of processes are the main culprit and also are the culprit unfairly, then you could find out if there's a problem with a particular program. |
Thanks for the responses. The default of 350 is just a nagios setting out of the box; I'll change it.
There appears to be a lot of disk-orientated processes running. The machine's running a hardware RAID but is also hosting a few VMs, one of which is quite intensive. There's also a backup subroutine which dumps VM snapshots to an attached USB drive. There's no problem with the performance of the machine though so I'll just ignore it. Cheers! |
Quote:
Do make sure you slowly increase the value in Nagios until you have a value that is acceptable. BTW: Can you put up the [SOLVED] tag. - above the first post -> Please Mark this thread as solved if you feel a solution has been provided. - -or- - - first post -> Thread Tools -> Mark this thread as solved |
All times are GMT -5. The time now is 06:41 PM. |