canonical-ci-engineering team mailing list archive

[Evan Dandrea <evan.dandrea@xxxxxxxxxxxxx>]

Fixing a CC typo :)

On 21 January 2014 16:26, Celso Providelo <celso.providelo@xxxxxxxxxxxxx> wrote:
> Guys,
>
> Right, that all said about rejected uploads being ephemeral, mainly to
> to avoid *cruft* (let aside malicious attacks) in LP and polling LP
> API for "accepted uploads" strategy working for now ... the whole
> experience is not great.
>
> In times of dput-ng [1] for which we can create nice pre-upload hooks
> using LP API (permissions, version collision, quota, and so on), we
> could certainly do an extra effort on the LP-side to return and store
> an unique upload identifier that would allow us to also create
> post-upload hooks  for inquiring/enforcing results.
>
> It might not be trivial since it will involve a mechanism to send
> information back from the s/ftp server, but we could possibly tweak
> the client transport to fetch the upload identifier from the server
> filesystem (or such).
>
> We will also operate on the assumption that upload email notification
> are good for people (inbox, ML, etc) and this initiative is not an
> attempt to replace them, on the other hand, the automation-front tasks
> could benefit a lot from a more assertive and responsive integration
> path.
>
> [1] http://dput-ng.debian.net/en/latest/
>
>
> On Fri, Jan 17, 2014 at 9:04 PM, Francis Ginther
> <francis.ginther@xxxxxxxxxxxxx> wrote:
>> Adam,
>>
>> Thanks for the response. I suspected this was the case, it's good to
>> get a confirmation.
>>
>> Francis
>>
>> On Fri, Jan 17, 2014 at 4:49 PM, Adam Conrad <adconrad@xxxxxxxxxxxxx> wrote:
>>> On Fri, Jan 17, 2014 at 03:46:07PM -0700, Adam Conrad wrote:
>>>>
>>>> The right way to detect a reject is via the email that's sent back.
>>>
>>> Furthermore, keep in mind that in some cases (like malformed uploads),
>>> there won't even be an email, as we can't safely determine a sensible
>>> target to mail to.
>>>
>>> It might be easier to just check if your upload was *accepted*, and
>>> after a sensible timeout, assume it wasn't, and take action.
>>>
>>> ... Adam
>>
>>
>>
>> --
>> Francis Ginther
>> Canonical - Ubuntu Engineering - Continuous Integration Team
>>
>> --
>> Mailing list: https://launchpad.net/~canonical-ci-engineering
>> Post to     : canonical-ci-engineering@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~canonical-ci-engineering
>> More help   : https://help.launchpad.net/ListHelp
>
>
>
> --
> Celso Providelo
> celso.providelo@xxxxxxxxxxxxx
>
> --
> Mailing list: https://launchpad.net/~canonical-ci-engineering
> Post to     : canonical-ci-engineering@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~canonical-ci-engineering
> More help   : https://help.launchpad.net/ListHelp