ABI stability of internal DBs

[John H Terpstra]

On 07/07/2010 11:21 AM, Jeremy Allison wrote:
> On Wed, Jul 07, 2010 at 06:12:05PM +0200, Volker Lendecke wrote:
>> On Wed, Jul 07, 2010 at 09:10:00AM -0700, Jeremy Allison wrote:
>>> This I think is perfectly reasonable, and is the
>>> level of stability we (unofficially) strive for
>>> today, and doesn't add any extra burdon to our
>>> testing resources.
>>>
>>> Thanks Simo - I think this is a very reasonable
>>> proposal and I'd like to standardize on this
>>> level of "stability" so long as everyone else
>>> agrees. Votes ?
>>
>> Who watches the checkins? What if you detect an smb2 oplock
>> bug in 3.6.0 that requires a format change in one of the
>> internal messages? Make the users wait for 3.7.0 instead of
>> 3.6.1? If you're fine with that, then we can make that
>> promise.
> 
> Yeah, I think so. Keeping stability between minor
> versions is a reasonable goal IMHO. I've only ever
> seen one bug that would have triggered this condition
> (the tdb update where the hash function was changed
> due to a signed -> unsigned change).
> 
> DB format updates inside a minor release version
> is a scummy thing to do to users (IMHO :-) and we
> should really try not to do it.
> 
> Jeremy.

Thank you all for the way this has been handled and resolved. I believe
it is important to remain aligned with the needs of those who depend on
our code.

Cheers,
John T.