Date: Thu, 11 Oct 2001 12:19:01 -0400 From: Mike Tancsa <mike@sentex.net> To: Luigi Rizzo <rizzo@aciri.org> Cc: freebsd-net@freebsd.org Subject: strange results with increased net.inet.ip.intr_queue_maxlen Message-ID: <5.1.0.14.0.20011011121308.053ddc20@marble.sentex.ca> In-Reply-To: <200110092255.f99Mtd195040@iguana.aciri.org> References: <5.1.0.14.0.20011009165532.0368b110@marble.sentex.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
Queue drops generally corresponded to bandwidth. Charting the bandwidth
going through the box and the rate at which queue drops increased certainly
seemed to correspond. I didnt run any statistical analysis, as the visual
correlation was very evident... But here is a strange result I dont quite
understand. I increased
net.inet.ip.intr_queue_maxlen from 50 to 100. and there didnt seem to be
any positive results in terms of lessening the rate of
net.inet.ip.intr_queue_drops.
On another machine that also was showing drops, I decided to start tracking
it as well. I increased net.inet.ip.intr_queue_maxlen to 256
unintentionally from the standard 50. However, this had the strange
result of reducing the rate at which net.inet.ip.intr_queue_maxlen was
increasing to zero. Huh ?
Why would there not be a more gradual/measured effect ? Does this make sense?
---Mike
At 03:55 PM 10/9/01 -0700, Luigi Rizzo wrote:
> > Also, in terms of queue drops, the fastforwarding did make a small
> > difference, but I still am seeing a series of drops somewhere between 5
> and
> > 10min. If you think it would be useful to track down to see if it is
> > exactly some interval, I can do so.
>
>it is dependent on external drive, anyways.. so the nice thing
>to see perhaps would be run a 'netstat -i' every second and
>see if there are evenly spaced traffic spikes.
>
> cheers
> luigi
> >
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5.1.0.14.0.20011011121308.053ddc20>