Mailing-List: contact dev-help@mesos.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@mesos.apache.org
Received-SPF: pass (athena.apache.org: domain of alex@mesosphere.io designates
 209.85.215.70 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAGBTdW+GCgHUeQwSgeHg46=5zh7gzXhP=CTujxzUe4G8g72-0A@mail.gmail.com>
References: 
 <CAGBTdW+GCgHUeQwSgeHg46=5zh7gzXhP=CTujxzUe4G8g72-0A@mail.gmail.com>
Date: Fri, 10 Apr 2015 13:18:06 +0200
Message-ID: 
 <CACzFoUsVQZW6h8tu9y=oA0E-rB5qA1f77db2gAY97UX1QiAtug@mail.gmail.com>
Subject: Re: Slave exit codes (MESOS-2599)
From: Alex Rukletsov <alex@mesosphere.com>
To: dev <dev@mesos.apache.org>
Content-Type: multipart/alternative; boundary=f46d043749c528b77505135cebf4

--f46d043749c528b77505135cebf4
Content-Type: text/plain; charset=UTF-8

I like the idea. Do you have a list (or want to come up with one) of what
conditions do you want to intercept?

On Fri, Apr 10, 2015 at 1:54 AM, Paul Brett <pbrett@twopensource.com> wrote:

> Currently, whenever an error occurs we have the slave exit with a return
> code of 1.  I propose to allocate separate return codes to each exit
> condition to help external scripts differentiate between failures.  It is
> possible that this will break external tools if they test for a return code
> of 1 rather than the more usual test for not equal to zero.
>
> Thoughts?
>
> -- Paul Brett
>

--f46d043749c528b77505135cebf4--