Openstack in HTTPD URL Scheme

[Adam Young]

There seems to be an agreement that the production configuration of 
Openstack should run in HTTPD using SSL etc.  I'd like to propose the 
following URL scheme for the web Apps so that the various pieces can be 
installed on a single machine without conflict.

All pieces will be served on port 443 using the https protocol.

#If and only if this is installed,  we should put in a forward from / to 
/dashboard
https://hostname/dashboard


#should this be keystone or identity?
https://hostname/keystone/main
https://hostname/keystone/admin

#should this be glance or image?
https://hostname/glance/api
https://hostname/glance/registry

#should this be nova or compute?
https://hostname/nova/api
https://hostname/nova/crt
https://hostname/nova/object
https://hostname/nova/cpu
https://hostname/nova/network
https://hostname/nova/volume
https://hostname/nova/schedule
https://hostname/nova/novnc
https://hostname/nova/vncx
https://hostname/nova/cauth

#should this be swift or storage?
https://hostname/swift/account
https://hostname/swift/object
https://hostname/swift/container


#should this be quantum or network?
https://hostname/quantum/service
https://hostname/quantum/agent


Please let me know if this list is accurate and complete.

[Christian Berendt]

Hello Adam.

On 04/27/2012 11:25 PM, Adam Young wrote:
>  #should this be keystone or identity? 
I would prefer using always the development name of the project, in this 
case keystone.

Regards, Christian.

-- 
Christian Berendt
Solution Architect
Mail: berendt(a)b1-systems.de

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

[Adam Young]

On 04/27/2012 05:33 PM, Christian Berendt wrote:
>  Hello Adam.
>
>  On 04/27/2012 11:25 PM, Adam Young wrote:
> > #should this be keystone or identity?
>
>  I would prefer using always the development name of the project, in 
>  this case keystone.
>
>  Regards, Christian.
>  
Thanks, Christian.  I suspect that will work better with documentation 
as well.

Pete Zaitcev also has pointed out that Swift probably won't play nicely 
with this,  as the URL scheme is pretty much driven by Amazon, and it 
assumes a top level URL along the lines of


http://test-1235163301.kvm-rei.zaitcev.lan/testdata
https://s3.amazonaws.com/test-1235163301/testdata
http://kvm-rei.zaitcev.lan/test-1235163301/testdata

[Daniel P. Berrange]

On Fri, Apr 27, 2012 at 05:37:32PM -0400, Adam Young wrote:
>  On 04/27/2012 05:33 PM, Christian Berendt wrote:
>  >Hello Adam.
>  >
>  >On 04/27/2012 11:25 PM, Adam Young wrote:
>  >>#should this be keystone or identity?
>  >
>  >I would prefer using always the development name of the project,
>  >in this case keystone.
>  >
>  >Regards, Christian.
>  >
>  Thanks, Christian.  I suspect that will work better with
>  documentation as well.
>  
>  Pete Zaitcev also has pointed out that Swift probably won't play
>  nicely with this,  as the URL scheme is pretty much driven by
>  Amazon, and it assumes a top level URL along the lines of
>  
>  
>  http://test-1235163301.kvm-rei.zaitcev.lan/testdata
>  https://s3.amazonaws.com/test-1235163301/testdata
>  http://kvm-rei.zaitcev.lan/test-1235163301/testdata 
If Swift is the only exception, then we would probably be ok, since I
think our other top level name prefixes would be unique enough to not
clash with what Swift used at the top level

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|

[Alan Pevec]

On Mon, Apr 30, 2012 at 10:24 AM, Daniel P. Berrange
<berrange(a)redhat.com&gt; wrote:
>  On Fri, Apr 27, 2012 at 05:37:32PM -0400, Adam Young wrote:
> > Pete Zaitcev also has pointed out that Swift probably won't play
> > nicely with this,  as the URL scheme is pretty much driven by
> > Amazon, and it assumes a top level URL along the lines of
> >
> >
> > http://test-1235163301.kvm-rei.zaitcev.lan/testdata
> > https://s3.amazonaws.com/test-1235163301/testdata
> > http://kvm-rei.zaitcev.lan/test-1235163301/testdata
>
>  If Swift is the only exception, then we would probably be ok, since I
>  think our other top level name prefixes would be unique enough to not
>  clash with what Swift used at the top level 
Even more, Swift should be a separate hostname, so clashing  would
only be a possibility in all-in-one which is a toy/development setup.

Alan

On 04/28/2012 12:25 AM, Adam Young wrote:
>  #should this be quantum or network?
>  https://hostname/quantum/service
>  https://hostname/quantum/agent 
I am not sure that the agent is relevant. In some
cases the agent 
accesses the database directly to retrieve the working configuration. In 
my opinion this should be treated in the same manner as the AMQP 
communication in nova.

Thanks
Gary
>  Please let me know if this list is accurate and complete.
>
>
>
>
>  _______________________________________________
>  cloud mailing list
>  cloud(a)lists.fedoraproject.org
>  https://admin.fedoraproject.org/mailman/listinfo/cloud

[Russell Bryant]

On 04/27/2012 05:25 PM, Adam Young wrote:
>  There seems to be an agreement that the production configuration of
>  Openstack should run in HTTPD using SSL etc.  I'd like to propose the
>  following URL scheme for the web Apps so that the various pieces can be
>  installed on a single machine without conflict.
>  
>  All pieces will be served on port 443 using the https protocol. 
<snip>

Is this something that we should be discussing upstream on the main
OpenStack list?

-- 
Russell Bryant

[Daniel P. Berrange]

On Sat, Apr 28, 2012 at 11:30:40PM -0400, Russell Bryant wrote:
>  On 04/27/2012 05:25 PM, Adam Young wrote:
>  > There seems to be an agreement that the production configuration of
>  > Openstack should run in HTTPD using SSL etc.  I'd like to propose the
>  > following URL scheme for the web Apps so that the various pieces can be
>  > installed on a single machine without conflict.
>  > 
>  > All pieces will be served on port 443 using the https protocol.
>  
>  <snip>
>  
>  Is this something that we should be discussing upstream on the main
>  OpenStack list? 
Indeed, I don't think that Fedora should be the people standardizing
this kind of thing. We should aim to get upstream to declare this the
standard so that all distros follow the same pattern.

Regards,
Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|

[Adam Young]

On 04/30/2012 04:21 AM, Daniel P. Berrange wrote:
>  On Sat, Apr 28, 2012 at 11:30:40PM -0400, Russell Bryant wrote:
> > On 04/27/2012 05:25 PM, Adam Young wrote:
> >> There seems to be an agreement that the production configuration of
> >> Openstack should run in HTTPD using SSL etc.  I'd like to propose the
> >> following URL scheme for the web Apps so that the various pieces can be
> >> installed on a single machine without conflict.
> >>
> >> All pieces will be served on port 443 using the https protocol.
> > <snip>
> >
> > Is this something that we should be discussing upstream on the main
> > OpenStack list?
>  Indeed, I don't think that Fedora should be the people standardizing
>  this kind of thing. We should aim to get upstream to declare this the
>  standard so that all distros follow the same pattern.
>
>  Regards,
>  Daniel 
Yeah.  My origianly thought was that this is a packing issue,  but I 
realize now that it is more fundamental.  I'm going to clean this up,  
integrate in some of hte comments,  and post to the overall openstack 
list this afternoon.