Philip Gorwell wrote:> Hello everyone!
>
> got problem to boot dom0 on 3.3.2-xvm in snv b123/4
>
> the important part:
> arcmsr0:Areca SATA Host Adapter RAID Controller arcmsr0:1.20.00.17Sun
WARNING: arcmsr0: ddi_dma_mem_alloc failed
> WARNING: arcmsr: arcmsr0: arcmsr_initialize failed
> arcmsr0:Areca SATA Host Adapter RAID Controller arcmsr0:1.20.00.17Sun
WARNING: arcmsr0: ddi_dma_mem_alloc failed
> WARNING: arcmsr: arcmsr0: arcmsr_initialize failed
> arcmsr0:Areca SATA Host Adapter RAID Controller arcmsr0:1.20.00.17Sun
WARNING: arcmsr0: ddi_dma_mem_alloc failed
> WARNING: arcmsr: arcmsr0: arcmsr_initialize failed
> arcmsr0:Areca SATA Host Adapter RAID Controller arcmsr0:1.20.00.17Sun
WARNING: arcmsr0: ddi_dma_mem_alloc failed
> WARNING: arcmsr: arcmsr0: arcmsr_initialize failed
> NOTICE: Can not read the pool label from
''/pci@0,0/pci10de,378@d/pci8086,370@0/pci17d3,1210@e/disk@0,0:a''
> NOTICE: spa_import_rootpool: error 5
> Cannot mount root on
/pci@0,0/pci10de,378@d/pci8086,370@0/pci17d3,1210@e/disk@0,0:a fstype zfs
>
> panic[cpu0]/thread=fffffffffbc60c20: vfs_mountroot: cannot mount root
>
> fffffffffbc94960 genunix:vfs_mountroot+350 ()
> fffffffffbc94990 genunix:main+e7 ()
> fffffffffbc949a0 unix:_locore_start+80 ()
>
> skipping system dump - no dump device configured
> rebooting...
I''m not sure that this is an xVM problem, because I came
across pretty darned close to the same problem with non-xVM
booting.
http://bugs.opensolaris.org/view_bug.do?bug_id=6890524
image-update to snv_124 from snv_122 results in vfs_mountroot panic
(spa_import_rootpool: error 5)
Here''s what I did to get past the problem:
=======
After running through the following sequence of operations, I was
unable to reproduce the problem:
1. boot the previous BE
2. mount the snv_124 BE under /mnt and rm /mnt/etc/zfs/zpool.cache
3. umount the snv_124 BE
4. reboot
5. at the grub menu, choose the snv_124 BE and add "-kd" to the
kernel$ line
6. when you get to kmdb, set a breakpoint
::bp zfs`vdev_disk_read_rootlabel
(and continue) :c
7. step through the vdev_disk_read_rootlabel function, remembering to step
_over_
any function calls.
8. after the root pool''s first device has been probed and attached in
the
ldi_open_by_devid() call, you should be ok to hit :c and continue the boot.
=======
hth,
James C. McPherson
--
Senior Kernel Software Engineer, Solaris
Sun Microsystems
http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog