zfs on gmirror slice

[Jeff Blank]

On Tue, Sep 01, 2009 at 09:53:28PM +0200, Ivan Voras wrote:
> Maciej Jan Broniarz wrote:
> > Is is a bad ida to create a zfs pool from a gmirrored slice?
> > zpool create tank /dev/mirror/gm0s1g works fine, but after the reboot
> > the filesystem failes consistency check.
> 
> It *should* work. What error(s) do you get?

I'm not the original poster, but I appear to be having similar
problems, though it's a panic rather than a failed consistency check.
I posted about it back in July, see
http://lists.freebsd.org/pipermail/freebsd-stable/2009-July/051167.html

I've since gotten a serial console working (needed to use uart instead
of sio), so I can now provide a backtrace, this from sources around
0600 UTC on 31 Aug.  My gmirror is entire disks (ad6/ad4), and my ZFS
pool/filesystems are still v6, on /dev/mirror/gm0s1d.

Jeff

================================
[...]
WARNING: ZFS is considered to be an experimental feature in FreeBSD.
Timecounters tick every 1.000 msec
ZFS filesystem version 13
ZFS storage pool version 13
ad4: 152587MB <SAMSUNG HE160HJ JF800-24> at ata2-master SATA300
ad6: 152587MB <SAMSUNG HE160HJ JF800-24> at ata3-master SATA300
GEOM_MIRROR: Device mirror/gm0 launched (1/2).
GEOM_MIRROR: Device gm0: rebuilding provider ad4.
acd0: CDROM <PBDS CD-ROM DH-48N1S/BD11> at ata4-master SATA150
acd1: CDRW <TSSTcorp CD-RW/DVD-ROM TS-H493B/D200> at ata5-master SATA150
hdac0: HDA Codec #0: Sigmatel STAC9220
pcm0: <HDA Sigmatel STAC9220 PCM #0 Analog> at cad 0 nid 1 on hdac0
SMP: AP CPU #1 Launched!
Trying to mount root from zfs:zgm0/root


Fatal trap 12: page fault while in kernel mode
cpuid = 1; apic id = 01
fault virtual address   = 0x0
fault code              = supervisor read data, page not present
instruction pointer     = 0x8:0xffffffff80da14fd
stack pointer           = 0x10:0xffffff800001efd0
frame pointer           = 0x10:0xffffff800001f020
code segment            = base 0x0, limit 0xfffff, type 0x1b
                        = DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 1 (swapper)
[thread pid 1 tid 100002 ]
Stopped at      vdev_geom_open+0x10d:   movq    (%r12),%rax
db> bt
Tracing pid 1 tid 100002 td 0xffffff000142eab0
vdev_geom_open() at vdev_geom_open+0x10d
vdev_open() at vdev_open+0xa8
vdev_root_open() at vdev_root_open+0x94
vdev_open() at vdev_open+0xa8
spa_load() at spa_load+0x1d2
spa_load() at spa_load+0x461
spa_open_common() at spa_open_common+0x12d
dsl_dir_open_spa() at dsl_dir_open_spa+0x279
dsl_dataset_hold() at dsl_dataset_hold+0x3b
dsl_prop_get() at dsl_prop_get+0x3c
zfs_mount() at zfs_mount+0x264
vfs_donmount() at vfs_donmount+0xdea
kernel_mount() at kernel_mount+0x88
kernel_vmount() at kernel_vmount+0xcb
vfs_mountroot_try() at vfs_mountroot_try+0x10c
vfs_mountroot() at vfs_mountroot+0x32e
start_init() at start_init+0x4d
fork_exit() at fork_exit+0x11e
fork_trampoline() at fork_trampoline+0xe
--- trap 0, rip = 0, rsp = 0xffffff800001fd30, rbp = 0 ---
db>