Ah, Edgar wasn't at the meeting on tuesday it seems, and so didn't hear
about our decision to try and merge your branch first. In hindsight, if we
ever decide on ordering commits, we should make sure to temporarily -2 the
commits we don't want submitted, so that other reviewers don't come in and
approve things.
I think you can just merge it in now on top of Maru's merged branch. You
may want to keep a version of the current code around as well, as it will
probably be easier to backport that version of the change to essex/stable
(assuming there isn't too much additional feedback).
Dan
On Wed, May 2, 2012 at 12:33 AM, Gary Kotton <gkotton@redhat.com> wrote:
> Hi,
> It seems that everything has gone pear shaped. I have conflicts with the
> "HACKING.rst" fix. From the IRC meeting last night I understood that I
> should check in first - so that the fix could be moved to Essex stable.
> Please advise. Sorry for being a pain, I am just new to git and it is a
> bit of a headache at the moment. My humble apologies.
> Thanks
> Gary
>
>
>
> ----- Original Message -----
> From: "Dan Wendlandt" <dan@nicira.com>
> To: gkotton@redhat.com
> Cc: "dan wendlandt (Code Review)" <review@openstack.org>, "Sumit
> Naiksatam" <snaiksat@cisco.com>, "Maru Newby" <mnewby@internap.com>,
> "markmcclain" <mark.mcclain@dreamhost.com>
> Sent: Wednesday, May 2, 2012 7:55:48 AM
> Subject: Re: Change in openstack/quantum[master]: blueprint agent-db-ha
> bug 985470 bug 985646
>
> Ok, we have a couple of changes backed up behind this one, so Gary, please
> rev this change as soon as you can. Thanks!
>
>
> Dan
>
>
> On Tue, May 1, 2012 at 3:05 PM, Gary Kotton < gkotton@redhat.com > wrote:
>
>
> On 05/01/2012 11:41 PM, dan wendlandt (Code Review) wrote:
>
>
> dan wendlandt has posted comments on this change.
>
>
> Change subject: blueprint agent-db-ha bug 985470 bug 985646
> .............................. .............................. ..........
>
>
> Patch Set 7: Looks good to me, but someone else must approve
>
> Overall, looks great. One thing to note is that this introduces new
> mandatory parameters to the config files for the OVS plugin and the Bridge
> plugin, meaning if people update their code but have existing config files,
> the code will puke. Avoiding this would be nice, particularly as it will
> happen for everyone using devstack.
>
> Nova's config system has the property that all config options have a
> default embedding in the code, which avoids this. We don't have quite the
> same capability here, as such defaults would be scattered throughout the
> file.
>
> Thoughts on this?
> Nice point. At the moment if the parameter does not exist then there is an
> exception, it would be best to set this as as default. I'm happy to make
> the change in the code. What do you think? I think that the fix is also
> valid for Essex stable.
>
>
>
> --
> To view, visit https://review.openstack.org/ 6744
> To unsubscribe, visit https://review.openstack.org/ settings
>
>
> Gerrit-MessageType: comment
> Gerrit-Change-Id: I9194f142478b130e8ef198b019539 357a9916d7f
> Gerrit-PatchSet: 7
>
> Gerrit-Project: openstack/quantum
> Gerrit-Branch: master
> Gerrit-Owner: garyk< gkotton@redhat.com >
>
> Gerrit-Reviewer: Isaku Yamahata
> Gerrit-Reviewer: Jenkins
> Gerrit-Reviewer: Maru Newby< mnewby@internap.com >
> Gerrit-Reviewer: Sumit Naiksatam< snaiksat@cisco.com >
> Gerrit-Reviewer: dan wendlandt< dan@nicira.com >
> Gerrit-Reviewer: garyk< gkotton@redhat.com >
> Gerrit-Reviewer: markmcclain< mark.mcclain@ dreamhost.com >
>
>
>
>
>
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Dan Wendlandt
> Nicira, Inc: www.nicira.com
>
> twitter: danwendlandt
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dan Wendlandt
Nicira, Inc: www.nicira.com
twitter: danwendlandt
~~~~~~~~~~~~~~~~~~~~~~~~~~~
about our decision to try and merge your branch first. In hindsight, if we
ever decide on ordering commits, we should make sure to temporarily -2 the
commits we don't want submitted, so that other reviewers don't come in and
approve things.
I think you can just merge it in now on top of Maru's merged branch. You
may want to keep a version of the current code around as well, as it will
probably be easier to backport that version of the change to essex/stable
(assuming there isn't too much additional feedback).
Dan
On Wed, May 2, 2012 at 12:33 AM, Gary Kotton <gkotton@redhat.com> wrote:
> Hi,
> It seems that everything has gone pear shaped. I have conflicts with the
> "HACKING.rst" fix. From the IRC meeting last night I understood that I
> should check in first - so that the fix could be moved to Essex stable.
> Please advise. Sorry for being a pain, I am just new to git and it is a
> bit of a headache at the moment. My humble apologies.
> Thanks
> Gary
>
>
>
> ----- Original Message -----
> From: "Dan Wendlandt" <dan@nicira.com>
> To: gkotton@redhat.com
> Cc: "dan wendlandt (Code Review)" <review@openstack.org>, "Sumit
> Naiksatam" <snaiksat@cisco.com>, "Maru Newby" <mnewby@internap.com>,
> "markmcclain" <mark.mcclain@dreamhost.com>
> Sent: Wednesday, May 2, 2012 7:55:48 AM
> Subject: Re: Change in openstack/quantum[master]: blueprint agent-db-ha
> bug 985470 bug 985646
>
> Ok, we have a couple of changes backed up behind this one, so Gary, please
> rev this change as soon as you can. Thanks!
>
>
> Dan
>
>
> On Tue, May 1, 2012 at 3:05 PM, Gary Kotton < gkotton@redhat.com > wrote:
>
>
> On 05/01/2012 11:41 PM, dan wendlandt (Code Review) wrote:
>
>
> dan wendlandt has posted comments on this change.
>
>
> Change subject: blueprint agent-db-ha bug 985470 bug 985646
> .............................. .............................. ..........
>
>
> Patch Set 7: Looks good to me, but someone else must approve
>
> Overall, looks great. One thing to note is that this introduces new
> mandatory parameters to the config files for the OVS plugin and the Bridge
> plugin, meaning if people update their code but have existing config files,
> the code will puke. Avoiding this would be nice, particularly as it will
> happen for everyone using devstack.
>
> Nova's config system has the property that all config options have a
> default embedding in the code, which avoids this. We don't have quite the
> same capability here, as such defaults would be scattered throughout the
> file.
>
> Thoughts on this?
> Nice point. At the moment if the parameter does not exist then there is an
> exception, it would be best to set this as as default. I'm happy to make
> the change in the code. What do you think? I think that the fix is also
> valid for Essex stable.
>
>
>
> --
> To view, visit https://review.openstack.org/ 6744
> To unsubscribe, visit https://review.openstack.org/ settings
>
>
> Gerrit-MessageType: comment
> Gerrit-Change-Id: I9194f142478b130e8ef198b019539 357a9916d7f
> Gerrit-PatchSet: 7
>
> Gerrit-Project: openstack/quantum
> Gerrit-Branch: master
> Gerrit-Owner: garyk< gkotton@redhat.com >
>
> Gerrit-Reviewer: Isaku Yamahata
> Gerrit-Reviewer: Jenkins
> Gerrit-Reviewer: Maru Newby< mnewby@internap.com >
> Gerrit-Reviewer: Sumit Naiksatam< snaiksat@cisco.com >
> Gerrit-Reviewer: dan wendlandt< dan@nicira.com >
> Gerrit-Reviewer: garyk< gkotton@redhat.com >
> Gerrit-Reviewer: markmcclain< mark.mcclain@ dreamhost.com >
>
>
>
>
>
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Dan Wendlandt
> Nicira, Inc: www.nicira.com
>
> twitter: danwendlandt
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dan Wendlandt
Nicira, Inc: www.nicira.com
twitter: danwendlandt
~~~~~~~~~~~~~~~~~~~~~~~~~~~