[OmniOS-discuss] panic: bd_strategy after new loader installed

[Geoff Nordli]

On 2017-05-28 11:50 PM, Peter Eriksson wrote:
> For what it’s worth I’ve seen the same error message on a Dell server 
> that happened to have 10TB-drives (not the boot disks!) that was 
> formatted using 4104-bytes sectors in order to support the T10-PI/DIF 
> & T10-DIF standards. The errors started to occur when we replaced the 
> RAID controller with an HBA. Non 512/4096-bytes sectors apparently 
> confused the boot loader when scanning the drives for the boot zpool. 
> This was on a FreeBSD system though. The solution was to perform a low 
> level reformatting of the drives in to standard 4096 byte sectors.
>
>
> https://en.wikipedia.org/wiki/Data_Integrity_Field
>
> https://access.redhat.com/solutions/41548
>
> http://www.enterprisestorageforum.com/storage-technology/sas-vs.-sata-4.html
>
>
> —
> [Lı.U] System Administrator ITI-NET IT.LiU.SE +46-13-28 2786
>
>> On 18 May 2017, at 21:55, Geoff Nordli <geoffn at gnaa.net 
>> <mailto:geoffn at gnaa.net>> wrote:
>>
>> Hi.
>>
>> I upgraded to the latest r151022 version.  The upgrade went fine, 
>> rebooted the machine, then upgraded to the new loader.
>>
>> When i rebooted after the loader upgrade, I got an error message:
>>
>> panic: bd_strategy: 512 bytes I/O not multiple of block size.
>>
>> press a key on the console to reboot....
>>
>> Rebooting...
>>
>> panic free: guard1 fail @ 0x2e2e2e67 from ../../common/devopen.c:64
>>
>> The boot disk is an 80GB intel ssd.
>>
>> Any thoughts?
>>
>> thanks,
>>
>> Geoff


follow up on this.

The issue is with 4kn disks.

Toomas provided an updated boot loader and I was able to get the system 
back online again.

https://www.illumos.org/issues/8303

thanks,
Geoff


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://omniosce.org/ml-archive/attachments/20170530/28d8616a/attachment.html>