Date: Wed, 18 Oct 2000 23:44:55 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: scrappy@hub.org (The Hermit Hacker) Cc: kris@catonic.net (Kris Kirby), tlambert@primenet.com (Terry Lambert), freebsd-stable@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Dual CPU Xeon server hangs on SMP kernel ... Message-ID: <200010182344.QAA18472@usr01.primenet.com> In-Reply-To: <Pine.BSF.4.21.0010182006580.974-100000@thelab.hub.org> from "The Hermit Hacker" at Oct 18, 2000 08:07:34 PM
next in thread | previous in thread | raw e-mail | index | archive | help
> > > Okay, since I can't actually boot into SMP, how would you suggest I get > > > that information? MPtable I know how to get, and can do that without > > > being in SMP, so will get that one later tonight and send it on, but any > > > sugggestions on how to get the dmesg from an actual SMP boot is most > > > welcome? > > > > Pull the keyboard and connect to the box with a null-modem connected to > > COM1 of the Xeon box. Capture to logfile. cat ~/logfile | mail > > freebsd@lists. > > damn, I was afraid that was going to be the only solution :( ah well, > will try and get into office tonight and do this ... It's important to see the stuff that comes up differently when attempting an SMP boot. However, assumming you had a UP boot in hand, you could get most of the information by comparing it to a hung SMP boot, and manually filing in the differences. The first thing to check is that chipset features, Id, and stepping are identical; if they aren't, you could have problems. The other issues are MPtable version, interrupt routing, and BIOS programming of your chipset (in general, I think that selecting "UnixWare" has shown the most overall success, but ou may need to go into advanced BIOs options for specific settings). BIOS versions and other stuff are also useful. The reason you didn't get this shotgun to start with (the way you asked the questions is basically a request for a shotgun) is that it's more useful to the project as a whole to know exatly why you are failing, so that it can be surgically corrected. Giving you a shotgun, and getting back a "Thanks; it works now" message doesn't provide a way for changing the code to avoid your situation in the future, and handing out shotguns for all of eternity isn't really a tenable support situation. Now that I've given you a shotgun, at least fire off only one pellet at a time, and let us know which one, if any, hits the mark. Preferrably, you'll go to the trouble of capturing the data, so that we can give you a better diagnosis, and help avoid this pain for others (or your next system) in the future. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200010182344.QAA18472>