Date: Tue, 30 Jan 2007 09:59:33 -0600 From: Stephen Montgomery-Smith <stephen@math.missouri.edu> To: freebsd-hackers@freebsd.org Subject: Re: top delay value Message-ID: <45BF6B65.1000105@math.missouri.edu> In-Reply-To: <20070130152633.GF19656@dan.emsphone.com> References: <20070130140227.26613101832@hk2.uwaterloo.ca> <20070130152633.GF19656@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Dan Nelson wrote: > In the last episode (Jan 30), waldeck@gmx.de said: > >>An unprivileged user could waste all CPU time by setting a low delay >>value in top (interactive or via -s). > > > Are you sure? In 6.2 at least, "s0" in interactive mode results in a > 1-second delay, and "top -s0" prints > > top: warning: seconds delay should be positive -- using default > .... You can run "top -s0" as root (not that this negates your point in any way). Incidently, this is an excellent way to generate those "calcru going backwards" errors, at least in some recent versions of FreeBSD. Just run a highly threaded, CPU intensive job at the same time. I did this on SMP systems, so I don't know if it also does this on non-SMP systems. On versions of FreeBSD 5.x I was able to induce kernel panics, the likes of which produced useless core dumps, but this seems to have been fixed with more recent versions. Stephen
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?45BF6B65.1000105>