Date: Tue, 07 Mar 2017 21:52:33 +0200 From: Toomas Soome <tsoome@me.com> To: "Michael W. Lucas" <mwlucas@michaelwlucas.com> Cc: current@freebsd.org Subject: Re: two recent snapshot installer problems Message-ID: <40408F86-006D-40FF-B6B1-777E24FB53EE@me.com> In-Reply-To: <20170307190549.GA97879@mail.michaelwlucas.com> References: <20170307190549.GA97879@mail.michaelwlucas.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> On 7. m=C3=A4rts 2017, at 21:05, Michael W. Lucas =
<mwlucas@michaelwlucas.com> wrote:
>=20
> Hi,
>=20
> I want to open a bug report on this, but have no idea how to gather
> useful info.
>=20
> Attempting to install
> FreeBSD-12.0-CURRENT-amd64-20170301-r314495-memstick.img onto an
> eight-drive iX system. This machine previously ran -current, but the
> OS has been sadly used by book testing and I decided to do a clean
> install.
>=20
> Used the Guided ZFS install, no networking, no extra users, default
> security.=20
>=20
> I want a two-way mirror for the base install, and the BIOS didn't
> provide serial numbers for each disk, so I wound up doing multiple
> installs, thinking that once I installed to the disk that the BIOS
> expected to be a boot disk, it'd find the pool and boot.
>=20
> Each of the four installs failed to boot, giving me:
>=20
> gptzfsboot: error 1 lba 32
> error 1
> gptzfsboot: error 1 lba 4294967288
> gptzfsboot: error 1 lba 1
> gptzfsboot: No ZFS pools located, can't boot
The messages are from low level disk IO, INT13.
if (V86_CY(v86.efl)) {
printf("%s: error %u lba %u\n",
BOOTPROG, v86.eax >> 8 & 0xff, lba);
return (-1);
}
So the error 1 is:
01h Invalid Command
Which means the INT13 AX=3D0x4200 is not supported by this BIOS. So, =
the question is, does your HBA is set in bios to IDE or AHCI mode? =
Changing that may make an difference, also usual suggestion - check for =
bios update, and if the host supports it, check the uefi boot=E2=80=A6 =
Of course, the fact that it did run fbsd before, points towards bios =
settings change...
the invalid command from INT13 extended read is by itself quite odd, but =
there is also possible another reason - if by any reason somehow the =
memory buffer for this call is from memory above 640k and so the invalid =
memory area was provided=E2=80=A6
rgds,
toomas
>=20
> Old disklabels? gpart -F destroy da0 through ada3. and
> reinstall. Nope.
>=20
> In frustration I did an eight-disk mirror install. Got the same
> message.
>=20
> Fine, I'll install to the Intel RAID satadom. It's a raid config, but
> it'll get me a working system. Select a ZFS stripe on raid/r0.
>=20
> gpart: arg0 'raid/r0': Invalid argument
>=20
> Mirror on the individual satadom drives?
>=20
> Same boot message.
>=20
> Any suggestions on how to gather debugging info for this?
>=20
> Thanks,
> =3D=3Dml
>=20
>=20
> --=20
> Michael W. Lucas Twitter @mwlauthor=20
> nonfiction: https://www.michaelwlucas.com/
> fiction: https://www.michaelwarrenlucas.com/
> blog: http://blather.michaelwlucas.com/
> _______________________________________________
> freebsd-current@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to =
"freebsd-current-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?40408F86-006D-40FF-B6B1-777E24FB53EE>