Hi Jack,

I checked the ZFS version of both AI image and repository and they 
match. I performed more tests and found that it's only affecting a 
particular platform:

+---------------+---------------------+----------+------------+--------------------+-------+
| NAME          | SYSTEM_DESC         | PROC_NUM | PROC_SPEED | 
PROC_DESC          | RAM   |
+---------------+---------------------+----------+------------+--------------------+-------+
| system1 | SUNW,Sun-Fire-T2000 |       16 |       1000 | 
UltraSPARC-T1      |  8184 |
| system2      | SUNW,Sun-Fire-T2000 |       32 |       1200 | 
SUNW,UltraSPARC-T1 | 16256 |
| system3          | SUNW,Sun-Fire-T2000 |       32 |       1200 | 
SUNW,UltraSPARC-T1 | 16256 |
| system4         | SUNW,Sun-Fire-T2000 |       32 |       1200 | 
SUNW,UltraSPARC-T1 | 16256 |
+---------------+---------------------+----------+------------+--------------------+-------+

 From the systems above, all but system1 are affected with the issue 
reported here. System1, using the same code, AI image and repo is fine.

All 4 systems will install Snv and S10 fine (used in regular DIY/PIT 
testing). Possible ON bug here?

Thanks,
David


Jack Schwartz wrote:
> Hi David.
>
> Does the repository match the AI image used to boot the client?  I'm 
> wondering if the client tries to boot using a different version of ZFS 
> than what is being installed?
>
> There have been similar issues:  For example, see:
> http://www.opensolaris.org/jive/thread.jspa?messageID=337361&#337361
>
>    Thanks,
>    Jack
>
> On 07/07/09 07:28, David Garcia - ONPIT wrote:
>> Hi,
>>
>> We're seeing this in the DIY lab for a while in two of our sparc 
>> boxes (both Sun Fire T2000).
>>
>> - The installation goes fine, the install service reports "Automated 
>> Installation finished successfully", so we reboot and then we'll get:
>> Boot device: /pci at 780/pci at 0/pci at 9/scsi at 0/disk at 0,0:a  File and 
>> args: 
>> kmdb -D kernel/sparcv9/unix
>> ERROR: /packages/zfs-file-system: Last Trap: Division by Zero
>>
>> {0} ok
>>
>>
>> I read some similar bugs that pointed at memory limitations. These 
>> systems have 16GB physical memory so must be something else.
>> No errors in the logs. Tried with lots of different AI images (these 
>> boxes are being used for OpenSolaris testing in DIY) and installadm 
>> tools updated several times (just tried the last nightly).
>>
>> Before I log a bug, does this look familiar to anyone?
>>
>> Thanks,
>> David
>> _______________________________________________
>> caiman-discuss mailing list
>> caiman-discuss at opensolaris.org
>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
>


Reply via email to