Re: Load over 1000

From: Poul-Henning Kamp <phk_at_phk.freebsd.dk>
Date: Mon, 21 Feb 2005 22:12:22 +0100
In message <20050221210834.GB87259_at_opteron.dglawrence.com>, "David G. Lawrence"
 writes:
>> aren't being serviced isn't a bug.  The reason the load on systems with
>> many processes is typically low is that most processes are blocked on I/O
>> -- either waiting for it to complete, waing for a network packet, or
>> waiting for the user, so they're idle the rest of the time.  The CPU sits
>> there waiting for the world to catch up...
>
>   The load average has historically meant the number of processes either
>running/ready to run OR blocked by short term (disk I/O) wait. 

No, disk I/O sleeps is not involved.

The loadavg is the length of the runqueue.  Any process sleeping,
on network, disk or timer, is not counted towards the total.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk_at_FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.
Received on Mon Feb 21 2005 - 20:12:37 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:28 UTC