Order in the Hudson for CogVMs

On 16 April 2011 17:27, Mariano Martinez Peck <[hidden email]> wrote:
>
> Hi...I love what you are doing in Hudson, but the Cog part it is quite a mess right now. I would love to see some organization.
>
> 1) Not all VMs have both: the build and the sources. Example, https://pharo-ic.lille.inria.fr/hudson/view/Cog/job/Cog%20Mac%20Cocoa/   has only the sources. But https://pharo-ic.lille.inria.fr/hudson/view/Cog/job/Cog%20Mac%20Carbon/
>  has both. If someone follows this: http://code.google.com/p/cog/wiki/CMakeVMMaker
> then such info is incorrect.
>
> 2) We are mixing "Release" VMs from experiments. I would create two Hudson tabs
>   2.1) one for Cog, where we have the "release" VMs. This means, things like Cog Mac Carbon, Cog Mac Cocoa, Cog Unix, CogMT Mac Cocoa, MT Cog Unix, Stack VM Unix, StackVM Mac Carbon, StackVM Mac Cocoa, etc.
>
>   2.2) One for experiments, like: Check cogvm.unthinkable.org, Cog Git Tracker (cog-osx), Cog Git Tracker (sig-cog), Cog.FreeBSD.unthinkable,Sista
>

The 'tracker's jobs are for tracking sources. they are triggering jobs
which building vms.

> Only if we do that, we will be able to use Hudson like a place to say to users to download a VM. Otherwise, we cannot. There are several targets and they are confusing.
>
> Cheers
>
> --
> Mariano
> http://marianopeck.wordpress.com
>
>
>

--
Best regards,
Igor Stasenko AKA sig.

On 16 April 2011 22:44, Mariano Martinez Peck <[hidden email]> wrote:
>
>
> On Sat, Apr 16, 2011 at 10:35 PM, Igor Stasenko <[hidden email]> wrote:
>>
>> On 16 April 2011 17:27, Mariano Martinez Peck <[hidden email]>
>> wrote:
>> >
>> > Hi...I love what you are doing in Hudson, but the Cog part it is quite a
>> > mess right now. I would love to see some organization.
>> >
>> > 1) Not all VMs have both: the build and the sources. Example,
>> > https://pharo-ic.lille.inria.fr/hudson/view/Cog/job/Cog%20Mac%20Cocoa/   has
>> > only the sources. But
>> > https://pharo-ic.lille.inria.fr/hudson/view/Cog/job/Cog%20Mac%20Carbon/
>> >  has both. If someone follows this:
>> > http://code.google.com/p/cog/wiki/CMakeVMMaker
>> > then such info is incorrect.
>> >
>> > 2) We are mixing "Release" VMs from experiments. I would create two
>> > Hudson tabs
>> >   2.1) one for Cog, where we have the "release" VMs. This means, things
>> > like Cog Mac Carbon, Cog Mac Cocoa, Cog Unix, CogMT Mac Cocoa, MT Cog Unix,
>> > Stack VM Unix, StackVM Mac Carbon, StackVM Mac Cocoa, etc.
>> >
>> >   2.2) One for experiments, like: Check cogvm.unthinkable.org, Cog Git
>> > Tracker (cog-osx), Cog Git Tracker (sig-cog), Cog.FreeBSD.unthinkable,Sista
>> >
>> The 'tracker's jobs are for tracking sources. they are triggering jobs
>> which building vms.
>>
>
> Thanks for explaining, I didn't know what it was.  But still, it is not a
> "release" VM. What I am saying is to split the VMs for final users. This
> means, a place where a Pharo/Squeak user can enter and pick the VM he wants.
> On the other hand, in another place, the rest of the VM: experiments,
> trackers, blah blah blah. Because right now, it cannot be used for final
> users. It is really confusing.
>

I think putting a correct description for every job will clarify things out. :)

> Cheers
>
> Mariano
>
>>
>> > Only if we do that, we will be able to use Hudson like a place to say to
>> > users to download a VM. Otherwise, we cannot. There are several targets and
>> > they are confusing.
>> >
>> > Cheers
>> >
>> > --
>> > Mariano
>> > http://marianopeck.wordpress.com
>> >
>> >
>> >
>>
>>
>>
>> --
>> Best regards,
>> Igor Stasenko AKA sig.
>
>
>
> --
> Mariano
> http://marianopeck.wordpress.com
>
>

--

Best regards,
Igor Stasenko AKA sig.