Date: Thu, 28 Aug 2008 11:12:59 +0100 From: "Joao Barros" <joao.barros@gmail.com> To: "Pascal Hofstee" <caelian@gmail.com> Cc: current@freebsd.org Subject: Re: SATA disks no longer showing up after updating CURRENT last night Message-ID: <70e8236f0808280312l55476d5bt55924f89e2c96bff@mail.gmail.com> In-Reply-To: <d8a0b7620808272328q301d6e1buc999031954f33852@mail.gmail.com> References: <d8a0b7620808272328q301d6e1buc999031954f33852@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Aug 28, 2008 at 7:28 AM, Pascal Hofstee <caelian@gmail.com> wrote: > Moin, > > Last night i updated my 8.0-CURRENT/amd64 system to svn revision r182296. > I then re-applied pjd's zfs-patchset and rebuild world and kernel overnig= ht. > > This morning i actually went ahead and installed the resulting world and = kernel > only to discover that during boot my two SATA disks no longer seem to sho= w up > at all. > > The onboard nVidia MCP55 SATA controller is still probed and attached tho= ugh if > memory serves me right the failing boot only showed a single nVidia MCP55= SATA > controller on atapci1, instead of two as the working boot does ... one > on atapci1 > the other on atapci2. > > My SATA disks are attached to ata2-master and ata3-master respectively > (ad4 and ad6) > > Somebody out there has any idea why the SATA Controller on atapci2 no > longer seems > to be probed and therefore my disks no longer show up ? > I'm having a "similar"ish problem on my system. I was running 7.0R and upgraded to CURRENT as of 26/08/2008 with pjd's zfs-patchset. After rebooting with the new kernel my system hangs here: http://img409.imageshack.us/img409/9995/picture1dk3.png My system consists of 4 disks, 2 on the onboard SATA and 2 on a onboard Promise, kernel on UFS, the rest on zfs using all 4 disks with raidz I was talking with Rui Paulo last night and he gave me the idea to disable the Promise, and voil=E1, the kernel boots but since I need the other 2 disks to complete the raidz I can't boot the system completely. This onboard Promise is a bit picky about sharing interrupts and I checked that, kernel.old boots ok (7.0R) On a side note, boot loader prompt for me can be a 2 seconds experience resulting either on a hang, a reboot or this ZX Spectrum era screen: http://img253.imageshack.us/img253/5765/picture2ua5.png --=20 Joao Barros
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?70e8236f0808280312l55476d5bt55924f89e2c96bff>