Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 13 Dec 2001 18:01:03 +0000
From:      Ian Dowse <iedowse@maths.tcd.ie>
To:        Holtor <holtor@yahoo.com>
Cc:        stable@freebsd.org
Subject:   Re: Load Averages 
Message-ID:   <200112131801.aa19310@salmon.maths.tcd.ie>
In-Reply-To: Your message of "Thu, 13 Dec 2001 09:21:08 PST." <20011213172108.10411.qmail@web11606.mail.yahoo.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
In message <20011213172108.10411.qmail@web11606.mail.yahoo.com>, Holtor writes:
>I still fail to see why my systems loads went from
>1.50 - 2.00. There's over 250 processes constantly
>running in "select" state. Loads are now almost always
>0.00 and sometimes touching 0.10

The changes to the load average calculation only added jitter to
the timing of samples; the algorithm used to compute the load average
from the samples is still the same. Could you post a `top' screen
shot from the server in question? (make sure to leave top running
for long enough to get the %idle etc lines filled in).

Is it possible that the processes spend the vast majority of the
time sleeping in select(), but that previously their run periods
were often synchronised with the samples used to calculate the load
average, or maybe it was system processes such as bufdaemon whose
run period was synchronised with the samples? If so, then maybe the
old loads around 2.0 were simply wrong. In general the %idle figure
in top should be close to 100% if the load is close to 0, so that
is worth checking too.

Ian

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi? <200112131801.aa19310>