[Container-tools] Ways to enable Vagrant Service-Manager plugin

[Lalatendu Mohanty]

On 02/29/2016 03:16 PM, Navid Shaikh wrote:
>
>
> On Mon, Feb 29, 2016 at 2:45 PM, Budh Ram <budhram.gurung01 at gmail.com 
> <mailto:budhram.gurung01 at gmail.com>> wrote:
>
>     On Mon, Feb 29, 2016 at 2:41 PM, Lalatendu Mohanty
>     <lmohanty at redhat.com <mailto:lmohanty at redhat.com>> wrote:
>
>         On 02/29/2016 02:34 PM, Budh Ram wrote:
>>         On Mon, Feb 29, 2016 at 2:09 PM, Brian (bex) Exelbierd
>>         <bex at pobox.com <mailto:bex at pobox.com>> wrote:
>>
>>             On Feb 26, 2016, at 6:15 PM, Budh Ram
>>             <budhram.gurung01 at gmail.com
>>             <mailto:budhram.gurung01 at gmail.com>> wrote:
>>             >
>>             > Hi All,
>>             >
>>             > We would like to get input on different ways of
>>             enabling "vagrant service-manager" plugin [1].
>>             >
>>             > We could come up with following possible ways:
>>             >
>>             > 1. Enable for all Vagrant boxes regardless.
>>
>>             I think this is too invasive.
>>
>>         I believe this one is default behavior of every vagrant
>>         plugin. If one don't want feature, then it should not be
>>         installed.
>>
>>
>>             > 2. Set "enabled" flag "true" in Vagrantfile unless we
>>             specify any service(s).
>>             >
>>             > config.servicemanager.enabled = true
>>             >
>>             >    or
>>             >
>>             > config.servicemanager.services = 'openshift'
>>
>>             This would work.  I believe that
>>             "config.service-manager.services = 'docker'" should also
>>             be accepted, even though it is a noop as pointed out below.
>>
>>             > 3. Detect when to enable based on a set of
>>             names(particulary box name like 'projectatomic/adb')
>>
>>             I like box detection.  I do not like detecting it via
>>             name. Could we scan for /etc/release information? How
>>             heavy would this be on unrelated boxes, especially
>>             non-Linux ones?
>>
>>
>>         Yes we can scan "/etc/os-release" and get the "VARIANT_ID" as
>>         "adb" or "cdk" and detact against the box mentioned in
>>         Vagrantfile. One concern is that the box name in Vagrantfile
>>         is combination of "projectatomic" and "adb"/"cdk". So, we can
>>         extract out "adb" / "cdk" and compare it. I hope this
>>         approach is fine.
>
>         We should just use /etc/os-release to detect the ADB/CDK box.
>         Vagrant-registration also only detects RHEL and prompts for
>         user name and password AFAIR.
>
>
>     Ok. That's great. I think I got the answer. Here is what I think
>     to trigger the pluggin:
>     1. Scan "/etc/os-release" file.
>
>     2. Trigger the plugin if it has "VARIANT_ID" as "adb"  or  "cdk".
>
>
> Finding out the box and triggering if applicable is the way to go. +1
>
> IMO, we should include "Guests" type of plugin feature [1] to figure 
> out underlying box.
>
> [1] https://www.vagrantup.com/docs/plugins/guests.html

Agree . Vagrant-registration uses the same framework.

>
>     Please acknowledge.
>
>
>         The name in Vagrantfile can be changed by the user easily, so
>         it is not a reliable thing.
>
>         -Lala
>
>
>         _______________________________________________
>         Container-tools mailing list
>         Container-tools at redhat.com <mailto:Container-tools at redhat.com>
>         https://www.redhat.com/mailman/listinfo/container-tools
>
>
>     -- 
>     Cheers!!!
>     Budh Ram Gurung
>
>     _______________________________________________
>     Container-tools mailing list
>     Container-tools at redhat.com <mailto:Container-tools at redhat.com>
>     https://www.redhat.com/mailman/listinfo/container-tools
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/container-tools/attachments/20160229/a0154edb/attachment.htm>