autobuild failure due to python replication tests - why ?

[Uri Simchoni]

On 08/03/2016 07:14 PM, Jeremy Allison wrote:
> On Wed, Aug 03, 2016 at 05:33:55PM +1200, Andrew Bartlett wrote:
>> On Tue, 2016-08-02 at 16:37 -0700, Jeremy Allison wrote:
>>> On Tue, Aug 02, 2016 at 05:13:34PM +1200, Andrew Bartlett wrote:
>>>>
>>>> On Mon, 2016-08-01 at 19:47 -0700, Jeremy Allison wrote:
>>>>>
>>>>> On Mon, Aug 01, 2016 at 07:32:03PM -0700, Jeremy Allison wrote:
>>>>>>
>>>>>>
>>>>>> Then the replication tests and ctdb tests NEED TO BE REMOVED
>>>>>> NOW !
>>>>>
>>>>> Apologies. Error EIAMTOOMUCHOFANASSHOLE :-).
>>>>>
>>>>> That was a little too strong. However I hope you understand
>>>>> my frustrations with this, we really do need to get a fix
>>>>> for this sooner (this week ?) rather than later. This week
>>>>> please ?
>>>>>
>>>>> In the meantime I'll resubmit, but that can't be the long
>>>>> term solution.
>>>>
>>>> I'm running tests to see if
>>>> reverting 055544321389de5d5cea8e129e0923369e9f558b or
>>>> e0b6d6bb10479759905932ea5438fc57713f0a8f helps, as this got much
>>>> worse
>>>> just as we forked for 4.5, which is where this commit was.
>>>>
>>>> It may simply be that we don't correctly notice the new schema, or
>>>> the
>>>> need for new schema to replicate schema.
>>>
>>> Unfortunately the autobuild failed again. That's 3 times in a row.
>>> I'm sorry but these replication tests have completely broken
>>> development on Samba, and we can't continue with this.
>>
>> Did you try with the tests referenced by those commits skipped, or with
>> those patches reverted?
> 
> Just to be clear, here is the patchset I'm trying to push at the moment
> (attached).
> 
> Do you want me to try with 055544321389de5d5cea8e129e0923369e9f558b
> and e0b6d6bb10479759905932ea5438fc57713f0a8f reverted also ?
> 

I tried a non-committing background buildnice autobuild on sn-devel,
skipping the tests mentioned in those commits (the ones in knownfail in
the commits put in skip) and it failed twice, but in places that don't
seem related.

One peculiar thing - is the following normal behavior?

tee: ./bin/ab/fl2008r2dc/samba_test.log: No space left on device

(and there was space when I checked)

Uri.