Mailing-List: contact builds-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: builds@apache.org
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\))
Subject: Re: identifying builder target architecture?
From: Allen Wittenauer <aw@apache.org>
In-Reply-To: <VI1PR0801MB148621705829D5C45D4089C0F2390@VI1PR0801MB1486.eurprd08.prod.outlook.com>
Date: Tue, 5 Jul 2016 07:49:16 -0700
Content-Transfer-Encoding: quoted-printable
Message-Id: <FDF2AF3C-D4A5-45F3-81AA-C223BA0FE089@apache.org>
References: <VI1PR0801MB148621705829D5C45D4089C0F2390@VI1PR0801MB1486.eurprd08.prod.outlook.com>
To: builds@apache.org
archived-at: Tue, 05 Jul 2016 14:49:20 -0000


> On Jul 5, 2016, at 7:42 AM, Richard Henwood <Richard.Henwood@arm.com> =
wrote:
>=20
>=20
> So if you have any guidance here - or general advice how to pursue my =
investigations - I would appreciate it.

	When I start targeting a different set of boxes on the build =
infrastructure, I generally start with a simple job that tells me the =
uname value, a few ls's, etc, to get the lay of the land then go from =
there. =20

	The problem is that the non-Linux-x86 machines always seem to be =
in various states of disrepair and/or just flat out turned off so =
there's a bit of a reliability problem when targeting those machines. :( =
 [Latest casualty appears to be the Mac OS X machine, turned off =
literally a day after I got Hadoop build status published to the mailing =
lists.]=