> On Jun 6, 2016, at 7:11 AM, Stian Soiland-Reyes <[email protected]> wrote:
>
> On 6 June 2016 at 14:32, Allen Wittenauer <[email protected]> wrote:
>>> Perhaps it would be better to modify this build to use Docker? That
>>> would also give a more consistent OS to link with.
>> This is something _everyone_ should be doing. With Yetus now
>> cleaning up dead/broken containers, dockerized jobs are less likely to leave
>> processes still running.
>
> Perhaps we need (links to) better documentation or examples.. what are
> good existing Jenkins jobs to use as inspiration?
Over in Apache Yetus land, we definitely have on our plate
documentation on how to set up a Jenkins job using it, including using either
the included Dockerfile or a custom Dockerfile like Apache Hadoop uses.
>
>
>>> Or perhaps restrict it to run on the hadoop slaves (label: Hadoop).
>>> which should have the hadoop libraries installed.
>> hdf5 is a file format and has nothing to do with Hadoop. :)
>
> Bah - silly me thinking hd$anything is Hadoop :)
If it makes you feel any better, the first time I saw HDF5 being
referenced I was seriously confused too. haha.
