boot errors since upgrading to 12-current

[Toomas Soome]

> On 15 Aug 2018, at 16:52, tech-lists <tech-lists at zyxst.net> wrote:
> 
> On 15/08/2018 13:20, tech-lists wrote:
>> On 15/08/2018 12:49, Kyle Evans wrote:
>>> On Wed, Aug 15, 2018 at 6:22 AM, tech-lists <tech-lists at zyxst.net> wrote:
>>>> Hello Toomas,
>>>> 
>>>> On 15/08/2018 07:31, Toomas Soome wrote:
>>>>> 
>>>>> Well that does explain the problem, if you look on the sizes reported… so
>>>>> your BIOS is reporting wrong sizes, is unable to access whole 4TB space and
>>>>> the zfs reader is not getting the correct data from the disks
>>>> 
>>>> 
>>>> Do you know why this was not happening on 11-stable but is happening on
>>>> 12-current?
>>>> 
>>>> The BIOS has not been modified since the machine was bought back in 2014.
>>>> 
>>> 
>>> Perhaps r335245, which fixed a logical error and completely changed
>>> the calculations for this stuff and possibly sent you on dramatically
>>> different path. There's not too many commits to this area that haven't
>>> yet landed back in stable/11, fortunately.
>>> 
>>> Thanks,
>>> 
>>> Kyle Evans
>>> 
>> Thank you Toomas and Kyle for the clarifications.
> 
> addendum to this - installed new world and kernel r337835 and the zfs warning at boot seems to have disappeared. Something else odd has started happening though, will make a new email for it - usb keyboard no longer functioning at the beastie menu, but functions when boot completes.
> 
> 


The size related problem is entirely about luck - to read the pool, we get pool config from label, and then read MOS (meta object set), and if we can not read MOS, we are busted. if we can read MOS, the zfs probe in boot loader is happy (actual file reads is still different story). However, whenever the pool is changed, the MOS will get updates and its location can (and will) change.

rgds,
toomas