Testing on PR's

[Josh Elser <jo...@gmail.com>]

We're currently running duplicate tests on Travis and ASF jenkins.

Last time I tried to do anything on Travis with a minicluster, it kind 
of blew up.

Given Mike's last PR had a failure in -Psunny, it would be good to be 
able to catch these automatically (let me be lazy and not re-run these 
tests before merging).

As such, I've tweaked 
https://builds.apache.org/job/Accumulo-Pull-Requests/ a little bit to 
try to stabilize it, but figured I would ask, long term, what should our 
goal be?

I'd like to get Yetus working again at some point, but I'm not sure how 
to wire it up with PR's (so that we have one way that contributions are 
tested).

- Josh

Re: Testing on PR's

[Josh Elser <jo...@gmail.com>]

Thanks, Sean. I am fuzzy on how Jenkins presently gets triggered right 
now (how we would get the context of what PR triggered it). I have to 
imagine this is available in some way, probably just a matter of digging :)

Sean Busbey wrote:
> You can give Yetus a link to either a jira or a github PR and it'll
> work out what you mean most of the time.
>
> I think it'll only comment on the github PR if you start with that,
> but I haven't tried making use of the jira-to-github bridge in that
> direction. (I've only started with a jira in Patch Available that has
> a link to a PR in it.)
>
> On Wed, Sep 21, 2016 at 9:53 AM, Josh Elser<jo...@gmail.com>  wrote:
>> We're currently running duplicate tests on Travis and ASF jenkins.
>>
>> Last time I tried to do anything on Travis with a minicluster, it kind of
>> blew up.
>>
>> Given Mike's last PR had a failure in -Psunny, it would be good to be able
>> to catch these automatically (let me be lazy and not re-run these tests
>> before merging).
>>
>> As such, I've tweaked https://builds.apache.org/job/Accumulo-Pull-Requests/
>> a little bit to try to stabilize it, but figured I would ask, long term,
>> what should our goal be?
>>
>> I'd like to get Yetus working again at some point, but I'm not sure how to
>> wire it up with PR's (so that we have one way that contributions are
>> tested).
>>
>> - Josh
>
>
>

Re: Testing on PR's

[Sean Busbey <bu...@cloudera.com>]

You can give Yetus a link to either a jira or a github PR and it'll
work out what you mean most of the time.

I think it'll only comment on the github PR if you start with that,
but I haven't tried making use of the jira-to-github bridge in that
direction. (I've only started with a jira in Patch Available that has
a link to a PR in it.)

On Wed, Sep 21, 2016 at 9:53 AM, Josh Elser <jo...@gmail.com> wrote:
> We're currently running duplicate tests on Travis and ASF jenkins.
>
> Last time I tried to do anything on Travis with a minicluster, it kind of
> blew up.
>
> Given Mike's last PR had a failure in -Psunny, it would be good to be able
> to catch these automatically (let me be lazy and not re-run these tests
> before merging).
>
> As such, I've tweaked https://builds.apache.org/job/Accumulo-Pull-Requests/
> a little bit to try to stabilize it, but figured I would ask, long term,
> what should our goal be?
>
> I'd like to get Yetus working again at some point, but I'm not sure how to
> wire it up with PR's (so that we have one way that contributions are
> tested).
>
> - Josh



-- 
busbey