Date: Tue, 11 Mar 1997 15:05:34 -0500 (EST) From: spork <spork@super-g.com> To: "Jordan K. Hubbard" <jkh@time.cdrom.com> Cc: Samara McCord <mccord@zytek.com>, freebsd-questions@FreeBSD.org Subject: Re: make world exposes memory errors!? Message-ID: <Pine.BSF.3.95.970311150133.16335D-100000@super-g.inch.com> In-Reply-To: <8152.858069596@time.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I'm a believer... I had a machine running er- Linux for months with no problems. Never did any heavy duty compile on it. Installed Solaris x86, and for the life of me, I couldn't build a new GCC. This was before I knew what signal 11 might have been trying to tell me. About a month ago I started tinkering with it and installed FBSD. No problems installing, but during the *second* make world, sig 11. Sent memory back for a refund (my hardware dealer is a nice guy) and did 4 more make worlds with new memory, and I never saw the problem again. I've found *lots* of bad memory this way. Putting this memory in a Windoze machine is kind of fun if you don't like the person you're giving it to. The incidences of unexplained crashes really increase under 95... Charles On Tue, 11 Mar 1997, Jordan K. Hubbard wrote: > > Assuming this is a memory problem, my question is: how is it that this machin > e, > > which has been running continuously for six months as a server, several domai > ns > > and websites and some heavy computational processing could avoid finding this > > problem while "make world" would expose it? Is there something peculiar > > going on here that would stress the system differently than normal usage? > > Not peculiar, but certainly more stress over a shorter period than > most machines seem to experience any other way. The `make world' is a > well accepted quality-assurance test for most FreeBSD hardware vendors > now, and I know that I certainly won't check off a box as "seaworthy" > until it's passed two consecutive make worlds. Nowadays, with machines > as fast as they are, I might even consider running 4 or 5 of them over > a 24 hour period and test the thermal sensitivity of the system at the > same time. > > Jordan >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.970311150133.16335D-100000>