Re: hudson matrix build

[Patricia Shanahan <pa...@acm.org>]

On 11/30/2010 1:16 PM, Sim IJskes - QCG wrote:
> I've tried to create a hudson QA testing system:
>
> River-QA-runtime: building the runtime (jars).
> River-QA: matrix job triggered by River-QA-runtime
>
> River-QA runs all combinations from systems and categories as separate
> jobs. Unfortunately every combination requires a workspace (right now)
> and the builds crew have called it a 'less optimal solution'. I think
> they are just beeing polite.
>
> So clearly until a new hudson version is released where the workspace
> can be shared between combinations, we cannot use this job.
>
> Maybe any of the developers on the list has a good idea, so that we can
> still make use of the matrix build system?

What is the advantage to running the categories separately, rather than 
a single all-categories QA job for each system?

Patricia

Re: hudson matrix build

[Sim IJskes - QCG <si...@qcg.nl>]

On 01-12-10 00:03, Patricia Shanahan wrote:
> On 11/30/2010 1:16 PM, Sim IJskes - QCG wrote:
>> I've tried to create a hudson QA testing system:
>>
>> River-QA-runtime: building the runtime (jars).
>> River-QA: matrix job triggered by River-QA-runtime

> What is the advantage to running the categories separately, rather than
> a single all-categories QA job for each system?

- A dashboard giving quick overview which categories failed
- Fair load on hudson (lot of small instead of 1 big)
- Restartable single categories
- Easy enabling, disabling of axis.
- Quicker results of individual categories.

-- 
QCG, Software voor het MKB, 071-5890970, http://www.qcg.nl
Quality Consultancy Group b.v., Leiderdorp, Kvk Den Haag: 28088397