[Testing] Closing Test in Release and QA Sign off bugs

[Mel Chua]

Oh. Maybe I shouldn't flash all the XOs with 765 tonight, then... I'm en 
route to the office now, and will check in with Ed and Michael first, 
but unless I hear otherwise from someone I'll install the latest 
candidate that's out when my train gets to Kendall.

PS: Any objections to me lining the edge of the Patmos with testing XOs 
for Saturday and Sunday? There's not much choice for space to do this in.

Kimberley Quirk wrote:
> The only issue I see is that 765 is probably not the final release and 
> we are going to need to 'cram' test the final release more than 
> spending a lot of time with 765.
>
> Copying Ed and Michael on this as well as they will help figure out 
> what release is our final one and they need to know that you (and 
> others) are geared up to test... we just need to know what to test!
>
> Thanks, Mel,
> Kim
>
>
> On Sep 26, 2008, at 3:28 PM, Mel Chua wrote:
>
>> (Joe, please feel free to trump me on this - I'll keep plugging on 
>> this unless somebody tells me otherwise.)
>>
>> Greg, by Monday at 11 I can get you and the testing list a breakdown 
>> of the bugs in report 28 sorted by the following:
>>
>> 1) Verified as closed - celebrate!
>> 2) Verified as not-closed (reopened), add to release notes
>> 3) We have no idea (probably treat the same as #2) - I hope this list 
>> will be empty.
>>
>> Plan for producing that email: (please revise/suggest as needed, 
>> partial strawman)
>>
>> * I've been writing test cases and procedures in the last few days 
>> for everything in report 33; I'll expand that to include the tickets 
>> in 28, which is a superset of 33.
>>
>> * You'll see me back at 1cc this afternoon installing 765 on the 20 
>> XOs we got from Richard. I'm going to use those as the in-1cc 8.2.0 
>> blockers test pool since our other testbeds are being used for 
>> connectivity tests (I think- Joe?) that require all 30 XOs running 
>> full-tilt on that.
>>
>> * This weekend: running tests. And running tests. And running tests. 
>> And - this is why I've spent so long on how-to-test tutorials and 
>> test cases - getting the visiting volunteers (+whoever I can pull in 
>> over IRC) on Saturday (MassXO comes in Sat. morning) and Sunday 
>> (support-gang) to run tests. I've been trying to build our capacity 
>> to run a lot of tests in parallel, and the bottleneck (imo) was 
>> people with XOs who were able to, with the given instructions, run 
>> all these tests. I hope I've fixed the "with the given instructions" 
>> part. You'll see this when I revise the [[Testing]] page later tonight.
>>
>> * Monday morning: Sit down and write up those 3 lists for Greg.
>>
>> How does this sound?
>>
>> -Mel
>>
>> Greg Smith wrote:
>>> Hi Mel, Joe and team,
>>>
>>> I am writing the 8.2 release notes and I see a number of bugs which 
>>> are marked Test In Release, Test in Build or QA Signoff.
>>>
>>> If they really are fixed then I don't need to document them. 
>>> However, if they fail final test then I should document them.
>>>
>>> I believe the relevant list is at:
>>> http://dev.laptop.org/report/28
>>>
>>> Can you give me an ETA when you think you can have all of those 
>>> verified or sent back for more work?
>>>
>>> For now I will assume that they are fixed and I will not document 
>>> them. If you find one that is not fixed, please cc me on the bug or 
>>> let me know and I will add it to the release notes later.
>>>
>>> Thanks,
>>>
>>> Greg S
>>>
>>
>