[Lldb-commits] [lldb] r327625 - Next batch of test-tree-cleaning changes

[Vedant Kumar via lldb-commits]

All right, I've configured greendragon's cmake/xcode builds to become unstable when there are test failures. I've also enabled email notifications for the 'default set of recipients'. From what I've gathered, Jenkins determines this set by looking at the changelist.

Let me know if there continue to be issues with failure notifications.

thanks,
vedant

> On Mar 15, 2018, at 11:02 AM, Vedant Kumar <vsk at apple.com> wrote:
> 
> I haven't received any greendragon bot failure notifications. I'll look into that.
> 
> vedant
> 
>> On Mar 15, 2018, at 8:32 AM, Davide Italiano <dccitaliano at gmail.com> wrote:
>> 
>> On Thu, Mar 15, 2018 at 8:28 AM, Pavel Labath <labath at google.com> wrote:
>>> Yea, this is strange. The bot does not even appear to be red. It's just
>>> yellow.
>>> 
>> 
>> This is historical. We ought to change this. We just delayed this
>> because we wanted to be *really sure* things were green for a while.
>> Also, UNEXPECTED SUCCESSES should be failures (we ought to change that as well).
>> 
>>> Anyway, r327633 ought to fix the breakage. As I said in the commit message,
>>> the cause is different treatment of relative paths to "process launch
>>> --stdin". Do you guys have any opinion on which behavior makes more sense?
>>> 
>> 
>> I don't have strong opinion on this, your fix seems reasonable.
>> 
>> Thanks,
>> 
>> --
>> Davide
>