Date: Wed, 21 Jan 2009 20:42:27 +0100 From: Rainer Hurling <rhurlin@gwdg.de> To: Gary Palmer <gpalmer@freebsd.org> Cc: Reuben <rapopp@marvin.eastcentral.edu>, John Rushford <jjr@alisa.org>, FreeBSD-stable@freebsd.org, Mark Kirkwood <markir@paradise.net.nz> Subject: Re: Broken loader on 7.1-STABLE? Message-ID: <49777AA3.5070202@gwdg.de> In-Reply-To: <20090121140307.GE81380@in-addr.com> References: <4975ADF4.1070103@paradise.net.nz> <C59BC6C8.259E%jjr@alisa.org> <20090121140307.GE81380@in-addr.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I have this issue only with amd64, not with i386. Both on the same hardware with up to date 8.0-CURRENT. In the not booting (amd64) version the following both lines, which indicate the detection of disks, are absent: ad4: 381554MB <SAMSUNG HD401LJ ZZ100-15> at ata2-master SATA300 ad6: 476940MB <SAMSUNG HD501LJ CR100-11> at ata3-master SATA300. I think the boot loader is ok, because I can load old amd64 kernel structures (before Oct. 9th 2008), but no newer one on this hardware. I have no clue what could be reason. Perhaps you are right with your suspicion about a kernel config problem? Please let me know if I can help. Thank you, Rainer On 21.01.2009 15:03 (UTC+1), Gary Palmer wrote: > On Tue, Jan 20, 2009 at 06:22:00PM -0700, John Rushford wrote: >> Hello, >> >> Perhaps mine is a different issue but I ran into this today. >> >> I loaded FreeBSD 7.1-RELEASE from a DVD onto an Apple Mac Pro after >> partitioning a drive with bootcamp. The install went quite well and the >> machine came up with no issues. I configured networking, X, and got gnome >> working. I setup cvsup and loaded the STABLE sources, ran make buildworld, >> make kernel, etc... >> >> I attempted to boot into single user to run mergemaster, installworld, etc.. >> But the machine will not boot, I now get the following errror and prompt: >> >> Trying to mount root from ufs:/dev/ad9s1a >> >> Manual root filesystem specification: >> <fstype>:<device> Mount <device> using filesystem <fstype> >> e.g. ufs:da0s1a >> ? List valid disk boot devices >> <empty line> Abort manual input >> >> I input ufs:/dev/ad9s1a to no avail and ? Does not list any other devices. >> >> I've booted from the DVD and gone into Fixit mode and can mount /dev/ad9s1a >> just fine. I thought that I'd try putting back loader.old on the off chance >> that might solve the problem but, I do not see a loader.old, just loader. >> The kernel.old directory is there. >> >> Any ideas on what the issue might be and how to fix? >> >> Appreciate the help as I'd like to get 7.1-STABLE running on this machine. > > Did the kernel probe messages prior to the mountroot prompt indicate that > the appropriate disk controllers were found/probed and that the > disk devices were found on the disk controllers? The lack of > any devices listed when you asked might indicate a kernel configuration > problem. > > Regards, > > Gary > > P.S. Please do not top post. > >> Regards >> John >> >> >> >> >> On 1/20/09 3:56 AM, "Mark Kirkwood" <markir@paradise.net.nz> wrote: >> >>> Reuben wrote: >>>> I was wondering if anyone else was seeing loader (v1.02) break after updating >>>> from 7.1-RELEASE to 7.1-STABLE. After performing the prescribed updating >>>> procedure (via the handbook), the system will go through the normal steps and >>>> after the boot menu will present the following error: >>>> >>>> Can't work out which disk we are booting from. >>>> Guessed BIOS device 0xffffffff not found by probes, defaulting to disk0 >>>> >>>> According to the bugbusting page on the FreeBSD wiki there's two issues at >>>> work that cause this behavior; patches were committed to HEAD/RELENG earlier >>>> last year in Mar and Aug. Up until now I've never come across this problem >>>> in 6.x or 7.0. In doing a little research I've come across a few older >>>> threads via google where it was believed that the problem was caused by >>>> improper CFLAGS in make.conf. I've commented mine out and rebuilt things.. >>>> with the same end result. In fact, if it's any help, my CFLAGS declaration >>>> in make.conf is taken verbatim from the /usr/share/examples/etc/make.conf. >>>> Furthermore, on selecting option 6 from the boot menu (escape to loader >>>> prompt), the system [I'm assuming] crashes displaying a blinking ASCII >>>> pattern from which only a hard reboot will work. >>>> >>>> FWIW, this is a fairly plain system.. nothing special in sysctl.conf or >>>> loader.conf, and the kernel is pretty stock as well (more or less GENERIC >>>> with my sound device and pf). >>>> >>>> A temporary fix for me was to copy over loader.old to loader in /boot. >>>> >>>> >>>> >>> I am getting this too - update from RELENG_7 @12 Jan src to 20 Jan and I >>> have: >>> >>> panic: free: guard1 fail @ 0x511d >>> from /usr/src/sys/boot/i386/loader/../../common/module.c:959 >>> >>> Can't work out which disk we are booting from. >>> Guessed BIOS device 0xffffffff not found by probes, defaulting to disk0 >>> >>> >>> >>> Copying loader.old to loader did not work around this for me - I had to >>> empty loader.conf too. >>> >>> Here is what *was* in there: >>> >>> ---> begin loader.conf >>> # Sound >>> sound_load="YES" >>> snd_hda_load="YES" >>> >>> # IPC >>> kern.ipc.semmni=500 >>> kern.ipc.semmns=500 >>> >>> >>> # CD/DVD DMA >>> atapicam_load="YES" >>> hw.ata.atapi_dma=0 >>> --> end loader.conf >>> >>> >>> At this point I am not sure which if are the bad guys... >>> >>> >>> The machine is an Asus a8vx. >>> >>> regards >>> >>> Mark
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49777AA3.5070202>