On Jan 16, 2014, at 6:46 AM, Joe Gordon <[email protected]> wrote:

> 
> 
> 
> On Wed, Jan 15, 2014 at 1:29 PM, Dugger, Donald D <[email protected]> 
> wrote:
> My thought was to try and get some parallel effort going, do the resync as a 
> continuing task as suffer a little ongoing pain versus a large amount of pain 
> at the end.  Given that the steps for a resync are the same no matter when we 
> do it waiting until the end is acceptable.
> 
>  
> 
> From a `just do it’ perspective I think we’re in violent agreement on the top 
> level tasks, as long as your step 3, integration testing, is the same as what 
> I’ve been calling working functionality, e.g. have the nova scheduler use the 
> gantt source tree.
> 
>  
> 
> PS:  How I resync.  What I’ve done is create a list with md5sums of all the 
> files in nova that we’ve duplicated in gantt.  I then update a nova git tree 
> and compare the current md5sums for those files with my list.  I use 
> format-patch to get the patches from the nova tree and grep for any patch 
> that applies to a gantt file.  I then use `git am’ to apply those patches to 
> the gantt tree, modifying any of the patches that are needed.
> 
> 
> So this sync won't work once we start the nova/gantt rename, so we need a 
> better approach.
> 
> Syncing the gantt tree with nova sounds like a daunting task. Perhaps it 
> would be easier if we use the current gantt tree as a test to see what is 
> involved in getting gantt working, and then redo the fork after the icehouse 
> feature freeze with the aim of getting the gantt tree working by the start of 
> juno, so we can have the freeze nova-scheduler discussion. Syncing nova and 
> gantt during feature freeze should be significantly easier then doing it now.


I would personally just vote for the nuclear approach of freezing nova 
scheduler and doing work in gantt. If close to icehouse 3 we see that gantt is 
not going to be ready in time we can selectively backport stuff to 
nova-scheduler and push gantt to juno.

Vish

>  
> 
>  
> 
> It may sound a little cumbersome but I’ve got some automated scripts that 
> make it relatively easy and modifying the patches, the hard part, was going 
> to be necessary no matter how we do it.
> 
>  
> 
> --
> 
> Don Dugger
> 
> "Censeo Toto nos in Kansa esse decisse." - D. Gale
> 
> Ph: 303/443-3786
> 
>  
> 
> From: Joe Gordon [mailto:[email protected]] 
> Sent: Wednesday, January 15, 2014 10:21 AM
> 
> 
> To: OpenStack Development Mailing List (not for usage questions)
> Subject: Re: [openstack-dev] [gantt] Sync up patches
> 
>  
> 
>  
> 
>  
> 
> On Wed, Jan 15, 2014 at 8:52 AM, Dugger, Donald D <[email protected]> 
> wrote:
> 
> The current thought is that I will do the work to backport any change that 
> are made to the nova tree that overlap the gantt tree.  I don’t see this as 
> an impossible task.  Yes it will get harder as we make specific changes to 
> gantt but, given that our first goal is to make gantt a drop in replacement 
> for the nova scheduler there shouldn’t be that many gantt specific changes 
> that would make backporting difficult so I think this is a doable path.
> 
>  
> 
> How are you tracking this today? I think its worth having a well documented 
> plan for this, as we will most likely have to keep syncing the two repos for 
> a while.
> 
>  
> 
> If all that is needed to cherry-pick a patch from nova to gantt is a 
> nova=>gantt rename these should be easy and a single +2 makes sense, but for 
> any patch that requires changes beyond that I think a full review should be 
> required.
> 
>  
> 
>  
> 
> For the ordering, the unit tests and working functionality are indeed 
> effectively the same, highest priority, I don’t have an issue with getting 
> the unit tests working first.
> 
>  
> 
> Great, so I would prefer to see gantt gating on unit tests before landing any 
> other patches.
> 
>  
> 
> Whats the full plan for the steps to bootstrap?  It would be nice to have a 
> roadmap for this so we don't get bogged down in the weeds. Off the top of my 
> head I imagine it would be something like (I have a feeling I am missing a 
> few steps here): 
> 
>  
> 
> 1) Get unit tests working
> 
> 2) Trim repo
> 
> 3) Set up integration testing  (In parallel get gantt client working)
> 
> 4) Resync with nova
> 
>  
> 
>  
> 
> As far as trimming is concerned I would still prefer to do that later, after 
> we have working functionality.  Since trimable files won’t have gantt 
> specific changes keeping them in sync with the nova tree is easy.  Until we 
> have working functionality we won’t really know that a file is not needed (I 
> am constantly surprised by code that doesn’t do what I expect) and deleting a 
> file after we are sure it’s not needed is easy.
> 
>  
> 
> Fair enough, I moved trimming after get unit tests working in the list above.
> 
>  
> 
>  
> 
> --
> 
> Don Dugger
> 
> "Censeo Toto nos in Kansa esse decisse." - D. Gale
> 
> Ph: 303/443-3786
> 
>  
> 
> From: Joe Gordon [mailto:[email protected]] 
> Sent: Wednesday, January 15, 2014 9:28 AM
> 
> 
> To: OpenStack Development Mailing List (not for usage questions)
> 
> Subject: Re: [openstack-dev] [gantt] Sync up patches
> 
>  
> 
>  
> 
>  
> 
> On Tue, Jan 14, 2014 at 2:22 PM, Dugger, Donald D <[email protected]> 
> wrote:
> 
> All-
> 
>  
> 
> I want to clear up some confusion I’m seeing in the reviews of these syncup 
> patches.  These patches merely bring recent changes from the nova tree over 
> to the gantt tree.  There is no attempt to actually change the code for 
> gantt, that is a separate task.  Our first goal is to have the scheduler in 
> gantt do exactly what the scheduler in nova does.  We want to be able to 
> reliably change nova to use the gantt source tree as a drop in replacement, 
> get that working before we start making gantt specific changes.
> 
>  
> 
> As far as I can tell this is the first time we have tried to fork a 
> repository without a freeze on the original codebase (when we split out 
> cinder, nova-volume was frozen).  With this form of forking, syncing the 
> repos becomes harder, and I am concerned with the sync method proposed here.  
> Once we do the big rename (%s/nova/gantt/g) in the gantt repo, we can't just 
> cherry-pick patches across without any modifications (assuming we gate on the 
> unit tests). So going forward how are we planning on keeping the two repos in 
> sync?
> 
>  
> 
> Is there an outline of the bootstrap process for Gantt? I would imagine the 
> first goal (before landing any sync patches) would be to get the unit tests 
> working.
> 
>  
> 
>  
> 
>  
> 
> The gantt tree probably has extra code that can be trimmed out later but, as 
> long as that code exists in gantt I want to make it a synced up copy of the 
> code in nova.
> 
>  
> 
> I'm not sure if I agree with the ordering proposed here. I would rather see 
> gantt be trimmed first.
> 
>  
> 
>  
> 
> --
> 
> Don Dugger
> 
> "Censeo Toto nos in Kansa esse decisse." - D. Gale
> 
> Ph: 303/443-3786
> 
>  
> 
> From: Dugger, Donald D 
> Sent: Tuesday, January 14, 2014 2:48 PM
> To: OpenStack Development Mailing List (not for usage questions)
> Subject: [gantt] Sync up patches
> 
>  
> 
> All-
> 
>  
> 
> As threatened, I’ve pushed 24 patches to sync up the gantt tree to recent 
> changes to the nova tree.  They’re all linked in a dependency chain starting 
> at:
> 
>  
> 
>                 https://review.openstack.org/66717
> 
>  
> 
> It’s be good if we can get those reviewed soon.
> 
>  
> 
> --
> 
> Don Dugger
> 
> "Censeo Toto nos in Kansa esse decisse." - D. Gale
> 
> Ph: 303/443-3786
> 
>  
> 
> 
> _______________________________________________
> OpenStack-dev mailing list
> [email protected]
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> 
>  
> 
> 
> _______________________________________________
> OpenStack-dev mailing list
> [email protected]
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> 
>  
> 
> 
> _______________________________________________
> OpenStack-dev mailing list
> [email protected]
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> 
> 
> _______________________________________________
> OpenStack-dev mailing list
> [email protected]
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
OpenStack-dev mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to