Date: Fri, 8 Aug 2008 08:55:53 -0400 From: Jakub Jasinski <jakub.jasinski@utoronto.ca> To: Kris Kennaway <kris@FreeBSD.org> Cc: freebsd-bugs@FreeBSD.org Subject: Re: misc/126230: weird process cpu usage stats and weird load Message-ID: <20080808085553.467482db.jakub.jasinski@utoronto.ca> In-Reply-To: <489C2726.9040003@FreeBSD.org> References: <200808080220.m782K3ra068330@freefall.freebsd.org> <489C2726.9040003@FreeBSD.org>
index | next in thread | previous in thread | raw e-mail
On Fri, 08 Aug 2008 12:59:50 +0200 Kris Kennaway <kris@FreeBSD.org> wrote: > Jakub Jasinski wrote: > > > Now, get cracking and fix select(). ;) > > Except I can't reproduce the bug. As expected (and previously > observed), running rsync does not cause select to use CPU time, > either with your test code or any other application. Something else > bizarre is going on on your system. > When you ran rsync, was your src remote? When src and dest were both local, a.out (and processes other than rsync) did NOT consume cpu for me. To make a.out consume cpu, I made src to be a FreeBSD 6.2 machine on the local network and was getting around 7MB/s. Maybe, something bizarre is going on that is select() _related_. (?) Processes like shells that were just waiting for stdin, did not consume cpu. BUT I'm not going to be surprised if it's specific to the (perhaps odd) hardware I'm using. (?) I'll be switching to FreeBSD 6.3 this weekend and I'll post any results I get. Whatever it is, it's unacceptable for me: my job is to redeploy the mail server on the new hardware by September. FreeBSD has treated me well during the last decade, until now. :) In any event, thanks for trying to reproduce the strange behavior.help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080808085553.467482db.jakub.jasinski>