Hello Cindy,

Looks like everything went fine this tile.
Thanks for the detailed procedure.

Just one last question : how to check the boot on the second disk  
(step 7) ?
Is it as simple as removing my primary disk from the pool and reboot ?
or a more complex procedure ?

Thanks

Seb
---
Sébastien Stormacq
Senior Software Architect
GSS Software Practice,
Sun Microsystems Luxembourg

On 26 Jan 2009, at 19:18, cindy.swearin...@sun.com wrote:

> Hi Sebastien,
>
> I've outlined steps below to help you recovery from the label problem.
> Make sure you can boot from the primary disk, c4t0d0s0, before you
> begin.
>
> I've reproduced these steps correctly on my Nevada, build 104 system
> because my OpenSolaris system doesn't have two disks.
>
> An easier recovery might be to just relabel the disk while it is still
> part of the root pool. I'm not sure how ZFS will react if the disk  
> label
> is changing while the disk is still part of the pool. That's a test  
> for
> another day... :-)
>
> Cindy
>
> 1. Make sure the system is currently bootable from c4t0d0s0.
>
> 2. Detach the second disk, c4t1d0s2, like this:
>
> # zpool detach rpool c4t1d0s2
>
> 3. Convert the existing EFI label on the disk to VTOC (SMI).
>
> # format -e c4t1d0s2
> format> label
> [0] SMI Label
> [1] EFI Label
> Specify Label type[1]: 0
> Auto configuration via format.dat[no]?
> Auto configuration via generic SCSI-2[no]?
> format> q
>
> Make sure all the disk space is in s2 or recreate it in s0. The
> relabeling process might go back to the default sizing so check
> to see that all the disk space is where you want it.
> (Write me offline if you need help with this step.)
>
> 4. Reattach the disk, c4t1d0s2, like this:
>
> # zpool attach rpool c4t0d0s0 c4t1d0s2
>
> 5. Wait for the newly attached disk to resilver completely
> by using zpool status to watch the progress.
>
> 6. Install the bootblock on the newly attached disk.
>
> # installgrub /boot/grub/stage1 /boot/grub/stage2 /dev/rdsk/c4t1d0s2
>
> 7. Confirm that you can boot from c4t1d0s2.
>
>
> Sébastien Stormacq wrote:
>> Hello,
>> I tried that already
>> format shows only a Solaris partition.
>> I did label the disk too
>> but
>> r...@yajug:~# zpool attach rpool mirror c4t1d0
>> cannot label 'c4t1d0': EFI labeled devices are not supported on  
>> root  pools.
>> Seb
>> ---
>> Sébastien Stormacq
>> Senior Software Architect
>> GSS Software Practice,
>> Sun Microsystems Luxembourg
>> On 23 Jan 2009, at 23:42, Evan Layton wrote:
>>> Try running format, choose that disk, run fdisk and remove the  
>>> EFI  fdisk partition and create a Solaris partition. Then in  
>>> format label  the disk.
>>>
>>> -evan
>>>
>>> Sébastien Stormacq wrote:
>>>
>>>> OK will try this
>>>> Any suggestion to repartition ?  I can't remind the exact errors   
>>>> but it seems that fdisk is not happy at all with the EFI   
>>>> partition :-(
>>>> thanks
>>>> ---
>>>> Sébastien Stormacq
>>>> Senior Software Architect
>>>> GSS Software Practice,
>>>> Sun Microsystems Luxembourg
>>>> On 23 Jan 2009, at 23:26, Evan Layton wrote:
>>>>
>>>>> On Jan 23, 2009, at 3:15 PM, Sébastien Stormacq 
>>>>> <sebastien.storm...@sun.com
>>>>>> wrote:
>>>>>
>>>>>> Evan,
>>>>>>
>>>>>> Using BE_PRINT_ERR=true (instead of 1 ;-) gives one more line   
>>>>>> that point to the error
>>>>>>
>>>>>> s...@yajug:~# beadm activate opensolaris-1
>>>>>> be_do_installgrub: installgrub failed for device c4t1d0s2.
>>>>>>
>>>>>>
>>>>>> c4t1d0s2 is a mirror in my zpool
>>>>>>
>>>>>> s...@yajug:~# zpool status
>>>>>> pool: rpool
>>>>>> state: ONLINE
>>>>>> scrub: resilver completed after 0h4m with 0 errors on Wed Jan  
>>>>>> 21  13:57:10 2009
>>>>>> config:
>>>>>>
>>>>>> NAME          STATE     READ WRITE CKSUM
>>>>>> rpool         ONLINE       0     0     0
>>>>>>  mirror      ONLINE       0     0     0
>>>>>>    c4t0d0s0  ONLINE       0     0     0  23.9M resilvered
>>>>>>    c4t1d0s2  ONLINE       0     0     0  4.16G resilvered
>>>>>>
>>>>>> errors: No known data errors
>>>>>>
>>>>>>
>>>>>> I did not install grub (yet) on the second disk
>>>>>> Actually, I had a whole bunch of issues to include that disk  
>>>>>> in  the mirror because it is EFI partitioned
>>>>>>
>>>>>> Any suggestion ?
>>>>>
>>>>> EFI labeled disks are not supported for root pools. The only  
>>>>> thing  I can suggest is to remove that device from the pool and  
>>>>> remove  the EFI labeled partition, create a solaris fdisk  
>>>>> partition (vtoc)  and re-attach the device.
>>>>>
>>>>> -evan
>>>>>
>>>>>
>>>>>>
>>>>>> Thanks for your help
>>>>>>
>>>>>>
>>>>>> ---
>>>>>> Sébastien Stormacq
>>>>>> Senior Software Architect
>>>>>> GSS Software Practice,
>>>>>> Sun Microsystems Luxembourg
>>>>>>
>>>>>> On 23 Jan 2009, at 23:11, Evan Layton wrote:
>>>>>>
>>>>>>
>>>>>>> Sébastien Stormacq wrote:
>>>>>>>
>>>>>>>> Hello,
>>>>>>>> I installed OpenSolaris 2008.11 on a Sun v40z machine.
>>>>>>>> Unfortunately the build I use, although being 101b was not  
>>>>>>>> the  latest  one, i.e. it was a couple of hours before GA :
>>>>>>>> Sun Microsystems Inc.   SunOS 5.11      snv_101b          
>>>>>>>> November 2008
>>>>>>>> I tried to update to GA with image-update
>>>>>>>> r...@yajug:~# pkg list -u
>>>>>>>> NAME (AUTHORITY)                               
>>>>>>>> VERSION           STATE      UFIX
>>>>>>>> entire                                         
>>>>>>>> 0.5.11-0.101      installed  u---
>>>>>>>> r...@yajug:~# pkg image-update
>>>>>>>> PHASE                                        ACTIONS
>>>>>>>> Update Phase                                     1/1
>>>>>>>> PHASE                                          ITEMS
>>>>>>>> Reading Existing Index                           9/9
>>>>>>>> Indexing Packages                                1/1
>>>>>>>> pkg: unable to activate opensolaris-1
>>>>>>>> r...@yajug:~# beadm activate opensolaris-1
>>>>>>>> Unable to activate opensolaris-1.
>>>>>>>> Unknown external error.
>>>>>>>> How can I further diagnose / solve this ??
>>>>>>>
>>>>>>> Can you set "export BE_PRINT_ERR=true" before running beadm   
>>>>>>> activate and send the debug output? That should help narrow  
>>>>>>> down  where the problem is.
>>>>>>>
>>>>>>> Thanks,
>>>>>>> -evan
>>>>>>>
>>>>>>>
>>>>>>>> I followed the same procedure with the same image on a  
>>>>>>>> couple  of VMs  and it was always OK
>>>>>>>> Thanks for your help
>>>>>>>> ---
>>>>>>>> Sébastien Stormacq
>>>>>>>> Senior Software Architect
>>>>>>>> GSS Software Practice,
>>>>>>>> Sun Microsystems Luxembourg
>>>>>>>> _______________________________________________
>>>>>>>> indiana-discuss mailing list
>>>>>>>> indiana-discuss@opensolaris.org
>>>>>>>> http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
>>>>>>>
>> _______________________________________________
>> indiana-discuss mailing list
>> indiana-discuss@opensolaris.org
>> http://mail.opensolaris.org/mailman/listinfo/indiana-discuss

_______________________________________________
indiana-discuss mailing list
indiana-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss

Reply via email to