Date: Thu, 23 Feb 2006 09:55:37 -0500 From: Vivek Khera <vivek@khera.org> To: FreeBSD AMD list <freebsd-amd64@freebsd.org> Subject: Re: AMD 64 stability Message-ID: <B6EDA464-D6B5-4E55-9845-FA91959E9C7A@khera.org> In-Reply-To: <43FD3B83.2040109@rtl.fmailbox.com> References: <43FCEF9C.5050308@bluelight.org.uk> <43FD3B83.2040109@rtl.fmailbox.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Feb 22, 2006, at 11:35 PM, Robert Leftwich wrote: > Be wary of FreeBSD's AMD64 support - test, test again and do even > more testing before the box leaves your sight! my experience: On the following it has been rock solid: Dell PE1850, Dell PE800 (though in 6.0 and up ACPI locks up at boot), Dell PE1425 (in testing still), Sun X4100 dual CPU. The Dells are Intel EM64T and the Sun is true Opteron. On the following it has been hit-or-miss: generic systems based on Tyan S2881 with Dual Opteron CPU. I went through 6 motherboards in 4 physically different systems and exacly one has been stable, though I'm afraid to reboot it. And just because your data center is 300 miles away doens't mean you can't manage it. Just hook up a serial console to a terminal server like a Cyclades box (or daisy-chain serial port amongst the other servers there) and ensure your kernel has the debugger enabled so you can control it even if it seems to hang on boot. All of the systems I use also route BIOS POST messages to the serial port. The Sun is best at it, allowing ctrl keys to control the functions. The dell's require appropriate terminal emulation of F keys.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?B6EDA464-D6B5-4E55-9845-FA91959E9C7A>