Solaris 11 Express doesn't boot after installing VBox 4.1
Posted: 11. Aug 2011, 23:42
I installed VirtualBox 4.1.0 on my fresh Solaris 11 Express (snv_151a), and it now hangs on boot. It seems like this might be some odd interplay between the VBox kernel driver and the mpt_sas driver that my storage HBA is using.
This is a fresh Solaris 11 Express install on my hardware. Before installing VirtualBox, everything works fine and I can reboot without problems. One time when I installed VirtualBox, the system actually hung right as the kernel driver was being installed. I rebuilt the box (for a different reason) and installed again, and it installed fine, but it always hangs on boot now.
The boot process gets up to the point where it says "Hostname: ....", then the three NOTICE: lines from the VBox kernel driver appear (kthread_t::t_preempt at 42, cpu_t::cpu_runrun at 216, cpu_t::cpu_kprunrun at 217). But then the box just hangs. If I wait long enough (don't remember the exact time, but on the order of 5-10 minutes, probably), the following prints on the console:
The system never seems to recover from that. I boot back into the boot environment that does not have the SUNWvbox package installed, the system works perfectly.
The key hardware in system is:
Anything I can do to further diagnose where the problem is? Is there anyone out there successfully running VirtualBox 4.1.0 on a system with LSI 2008-based storage controllers?
Thanks,
Matthew
This is a fresh Solaris 11 Express install on my hardware. Before installing VirtualBox, everything works fine and I can reboot without problems. One time when I installed VirtualBox, the system actually hung right as the kernel driver was being installed. I rebuilt the box (for a different reason) and installed again, and it installed fine, but it always hangs on boot now.
The boot process gets up to the point where it says "Hostname: ....", then the three NOTICE: lines from the VBox kernel driver appear (kthread_t::t_preempt at 42, cpu_t::cpu_runrun at 216, cpu_t::cpu_kprunrun at 217). But then the box just hangs. If I wait long enough (don't remember the exact time, but on the order of 5-10 minutes, probably), the following prints on the console:
Code: Select all
WARNING: /pci@0,0/pci8086,105@1,1/pci1000,3020@0 (mpt_sas0): Disconnected command timeout for Target 9
WARNING: /pci@0,0/pci8086,1c1@1c,1/pci1000,3020@0 (mpt_sas0): Disconnected command timeout for Target 10
The key hardware in system is:
- Intel S1200BTS motherboard, Xeon E3-1230 processor
2 x LSI SAS9211-8I SAS/SATA HBAs (just doing JBOD with ZFS) - uses the mpt_sas driver
Anything I can do to further diagnose where the problem is? Is there anyone out there successfully running VirtualBox 4.1.0 on a system with LSI 2008-based storage controllers?
Thanks,
Matthew