You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cloudstack.apache.org by Ewan Mellor <Ew...@eu.citrix.com> on 2012/08/10 02:27:33 UTC
RE: [ASFCS40] CloudStack 4.0 release plan
> -----Original Message-----
> From: Chip Childers [mailto:chip.childers@sungard.com]
>
> [Snip]
>
> Thinking about the milestone along those lines, I agree that cutting a
> release branch over the weekend is a good thing. This is specifically
> because of the pending review board submissions that we should start
> getting integrated into master as a parallel activity to 4.0 work.
OK, so I think from this and David's earlier email and general consensus that we want to get into a time-based schedule, then we're still on to start the feature freeze at the end of Friday, and have a 4.0 release branch ready for Monday morning. Everyone agreed?
> [Snip]
>
> - Regardless of our desire to be time-bound for our releases, our
> first release has a higher-order issue (licensing) that will block us
> if anything significant is outstanding.
> - Given the importance of the licensing aspects, I believe that we
> have to think of the "First release candidate build" milestone as
> being predicated on the community believing / agreeing that we have
> achieved a satisfactory level of compliance to pass an IPMC vote.
Yes, makes sense to me.
> [Snip]
>
> Let's think about "convenience builds" along two lines (because there
> are different issues to deal with):
>
> Binary distribution of the core project - I think this is basically
> sorting through the optional vs. required components. Any binary
> distro we produce would be only include the required build targets,
> not the optional ones. I'm not sure there is much debate open here.
> We're headed down the right path. :-)
Yep.
> System VMs - I've reviewed the discussion notes from OSCON again, just
> to refresh my memory. I might not be understanding the verbal
> conversation correctly, but it doesn't seem to match what I see on the
> apache.org/legal site. Specifically, David mentioned that there was
> verbal consensus that we could possibly "prepare a system VM as we
> distribute it now as a convenience binary". The item that I'm
> concerned about is the "Distribution" section of the Third-Party
> Licensing Policy [1] that says "YOU MUST NOT distribute a prohibited
> work from an apache.org server.". Doesn't the system VM qualify as a
> prohibited work?
I agree that System VM binaries are forbidden by written policy. It is possible to ask for exceptions to this policy -- I have done so for libvirt-java, and Sam Ruby says that he'll approve that next week if no-one disagrees in the meantime. If we felt that we needed to make a similar request regarding System VMs, we can do that.
Getting alternative infrastructure is a possibility. I'm sure Citrix could manage that, though I worry about the perception that would create, if the software is effectively unusable without Citrix servers.
Cheers,
Ewan.
Re: [ASFCS40] CloudStack 4.0 release plan
Posted by David Nalley <da...@gnsa.us>.
On Fri, Aug 10, 2012 at 4:00 PM, Ewan Mellor <Ew...@eu.citrix.com> wrote:
>> -----Original Message-----
>> From: David Nalley [mailto:david@gnsa.us]
>> Sent: 10 August 2012 12:56
>> To: cloudstack-dev@incubator.apache.org
>> Cc: Ewan Mellor
>> Subject: Re: [ASFCS40] CloudStack 4.0 release plan
>>
>> >
>> > The only exception I would make to the "everything in 4.0" rule is for
>> documentation. I don't see any value in branching the docs, and I want to
>> make sure that we can keep updating the docs at a fast pace even as we're
>> deliberately slowing down the rate of change on the code. For this reason, I
>> would have all the docs work continue in master, so there would be no
>> additional gatekeeping there, just the usual review. Is everyone OK with
>> that?
>> >
>>
>> I've read this several times, and not sure that I understand.....Are you saying:
>> that docs will continue to go into both 4.0 and master and will skip the
>> gatekeeper?
>> or that docs (e.g. the docs that will exist for 4.0) will happen only in master
>> going forward?
>>
>> The latter concerns me, as we will release a source tarball whose docs won't
>> match our 4.0 release docs. Presumably master will rapidly move on towards
>> 4.1 docs after release, and we'll have no demonstrable way to hack on 4.0
>> docs or for downstreams to consume 4.0 docs or for us to produce
>> translations of the exact 4.0 docs after they move on. In short my concern is:
>> how do we produce a 'this is the source for docs that match this code
>> release', and ensure that that point in time still exists 6 months down the
>> road?
>
> How about we keep working on docs in master, and then dump the latest snapshot into the release branch before we do the final build?
>
> Ewan.
That worksforme, but lets make sure that's a TODO in the release plan.
--David
Re: FW: [ASFCS40] CloudStack 4.0 release plan
Posted by David Nalley <da...@gnsa.us>.
On Fri, Aug 10, 2012 at 8:49 PM, Jessica Tomechak
<je...@gmail.com> wrote:
>>
>>
>> -----Original Message-----
>> From: Ewan Mellor [mailto:Ewan.Mellor@eu.citrix.com]
>> Sent: Friday, August 10, 2012 1:01 PM
>> To: David Nalley; cloudstack-dev@incubator.apache.org
>> Subject: RE: [ASFCS40] CloudStack 4.0 release plan
>>
>> > -----Original Message-----
>> > From: David Nalley [mailto:david@gnsa.us]
>> > Sent: 10 August 2012 12:56
>> > To: cloudstack-dev@incubator.apache.org
>> > Cc: Ewan Mellor
>> > Subject: Re: [ASFCS40] CloudStack 4.0 release plan
>> >
>> > >
>> > > The only exception I would make to the "everything in 4.0" rule is
>> > > for
>> > documentation. I don't see any value in branching the docs, and I
>> > want to make sure that we can keep updating the docs at a fast pace
>> > even as we're deliberately slowing down the rate of change on the
>> > code. For this reason, I would have all the docs work continue in
>> > master, so there would be no additional gatekeeping there, just the
>> > usual review. Is everyone OK with that?
>> > >
>> >
>> > I've read this several times, and not sure that I understand.....Are you
>> saying:
>> > that docs will continue to go into both 4.0 and master and will skip
>> > the gatekeeper?
>> > or that docs (e.g. the docs that will exist for 4.0) will happen only
>> > in master going forward?
>> >
>> > The latter concerns me, as we will release a source tarball whose docs
>> > won't match our 4.0 release docs. Presumably master will rapidly move
>> > on towards
>> > 4.1 docs after release, and we'll have no demonstrable way to hack on
>> > 4.0 docs or for downstreams to consume 4.0 docs or for us to produce
>> > translations of the exact 4.0 docs after they move on. In short my
>> concern is:
>> > how do we produce a 'this is the source for docs that match this code
>> > release', and ensure that that point in time still exists 6 months
>> > down the road?
>>
>> How about we keep working on docs in master, and then dump the latest
>> snapshot into the release branch before we do the final build?
>>
>> Ewan.
>>
>
> I like this idea.
>
> The rate of change in documentation speeds up the closer we get to a
> release. This makes docs the opposite of code.
>
> I'd like to be able to fix doc bugs for any past release, too, then
> auto-post the update to docs.cloudstack.org under the correct release
> number. I will add that to the requirements for our automated doc
> publishing tool (which, by the way, we could use help implementing!)
>
> https://cwiki.apache.org/confluence/display/CLOUDSTACK/Functional+Requirements+for+Documentation+Publishing+Tool
>
> Jessica T.
> CloudStack Tech Pubs
Not to derail to a completely different topic, but publican doesn't do
that already?
--David
Re: FW: [ASFCS40] CloudStack 4.0 release plan
Posted by Jessica Tomechak <je...@gmail.com>.
>
>
> -----Original Message-----
> From: Ewan Mellor [mailto:Ewan.Mellor@eu.citrix.com]
> Sent: Friday, August 10, 2012 1:01 PM
> To: David Nalley; cloudstack-dev@incubator.apache.org
> Subject: RE: [ASFCS40] CloudStack 4.0 release plan
>
> > -----Original Message-----
> > From: David Nalley [mailto:david@gnsa.us]
> > Sent: 10 August 2012 12:56
> > To: cloudstack-dev@incubator.apache.org
> > Cc: Ewan Mellor
> > Subject: Re: [ASFCS40] CloudStack 4.0 release plan
> >
> > >
> > > The only exception I would make to the "everything in 4.0" rule is
> > > for
> > documentation. I don't see any value in branching the docs, and I
> > want to make sure that we can keep updating the docs at a fast pace
> > even as we're deliberately slowing down the rate of change on the
> > code. For this reason, I would have all the docs work continue in
> > master, so there would be no additional gatekeeping there, just the
> > usual review. Is everyone OK with that?
> > >
> >
> > I've read this several times, and not sure that I understand.....Are you
> saying:
> > that docs will continue to go into both 4.0 and master and will skip
> > the gatekeeper?
> > or that docs (e.g. the docs that will exist for 4.0) will happen only
> > in master going forward?
> >
> > The latter concerns me, as we will release a source tarball whose docs
> > won't match our 4.0 release docs. Presumably master will rapidly move
> > on towards
> > 4.1 docs after release, and we'll have no demonstrable way to hack on
> > 4.0 docs or for downstreams to consume 4.0 docs or for us to produce
> > translations of the exact 4.0 docs after they move on. In short my
> concern is:
> > how do we produce a 'this is the source for docs that match this code
> > release', and ensure that that point in time still exists 6 months
> > down the road?
>
> How about we keep working on docs in master, and then dump the latest
> snapshot into the release branch before we do the final build?
>
> Ewan.
>
I like this idea.
The rate of change in documentation speeds up the closer we get to a
release. This makes docs the opposite of code.
I'd like to be able to fix doc bugs for any past release, too, then
auto-post the update to docs.cloudstack.org under the correct release
number. I will add that to the requirements for our automated doc
publishing tool (which, by the way, we could use help implementing!)
https://cwiki.apache.org/confluence/display/CLOUDSTACK/Functional+Requirements+for+Documentation+Publishing+Tool
Jessica T.
CloudStack Tech Pubs
RE: [ASFCS40] CloudStack 4.0 release plan
Posted by Ewan Mellor <Ew...@eu.citrix.com>.
> -----Original Message-----
> From: David Nalley [mailto:david@gnsa.us]
> Sent: 10 August 2012 12:56
> To: cloudstack-dev@incubator.apache.org
> Cc: Ewan Mellor
> Subject: Re: [ASFCS40] CloudStack 4.0 release plan
>
> >
> > The only exception I would make to the "everything in 4.0" rule is for
> documentation. I don't see any value in branching the docs, and I want to
> make sure that we can keep updating the docs at a fast pace even as we're
> deliberately slowing down the rate of change on the code. For this reason, I
> would have all the docs work continue in master, so there would be no
> additional gatekeeping there, just the usual review. Is everyone OK with
> that?
> >
>
> I've read this several times, and not sure that I understand.....Are you saying:
> that docs will continue to go into both 4.0 and master and will skip the
> gatekeeper?
> or that docs (e.g. the docs that will exist for 4.0) will happen only in master
> going forward?
>
> The latter concerns me, as we will release a source tarball whose docs won't
> match our 4.0 release docs. Presumably master will rapidly move on towards
> 4.1 docs after release, and we'll have no demonstrable way to hack on 4.0
> docs or for downstreams to consume 4.0 docs or for us to produce
> translations of the exact 4.0 docs after they move on. In short my concern is:
> how do we produce a 'this is the source for docs that match this code
> release', and ensure that that point in time still exists 6 months down the
> road?
How about we keep working on docs in master, and then dump the latest snapshot into the release branch before we do the final build?
Ewan.
Re: [ASFCS40] CloudStack 4.0 release plan
Posted by David Nalley <da...@gnsa.us>.
>
> The only exception I would make to the "everything in 4.0" rule is for documentation. I don't see any value in branching the docs, and I want to make sure that we can keep updating the docs at a fast pace even as we're deliberately slowing down the rate of change on the code. For this reason, I would have all the docs work continue in master, so there would be no additional gatekeeping there, just the usual review. Is everyone OK with that?
>
I've read this several times, and not sure that I understand.....Are you saying:
that docs will continue to go into both 4.0 and master and will skip
the gatekeeper?
or that docs (e.g. the docs that will exist for 4.0) will happen only
in master going forward?
The latter concerns me, as we will release a source tarball whose docs
won't match our 4.0 release docs. Presumably master will rapidly move
on towards 4.1 docs after release, and we'll have no demonstrable way
to hack on 4.0 docs or for downstreams to consume 4.0 docs or for us
to produce translations of the exact 4.0 docs after they move on. In
short my concern is: how do we produce a 'this is the source for docs
that match this code release', and ensure that that point in time
still exists 6 months down the road?
--David
RE: [ASFCS40] CloudStack 4.0 release plan
Posted by Ewan Mellor <Ew...@eu.citrix.com>.
> -----Original Message-----
>
> [Snip]
>
> +1, from monday on we'll have a 4.0 branch.
>
> That means however that you have to submit bugs fixes into master and
> 4.0.x, right?
Yes, that's right. All changes should go into master, and anything that we want for 4.0 should go into the release branch too. That's going to mean pretty much everything at this point.
If there is going to be any value in having a release branch, we should have someone gatekeeping between master and 4.0, so rather than throwing everything into 4.0, we have an additional review step. We should be aiming to slow down the rate of churn on 4.0 so that we can fix a build to give people time to do some testing. This person needs to be a committer, and needs to know the codebase in and out. Alex, I think that this person is you. What do you think?
The only exception I would make to the "everything in 4.0" rule is for documentation. I don't see any value in branching the docs, and I want to make sure that we can keep updating the docs at a fast pace even as we're deliberately slowing down the rate of change on the code. For this reason, I would have all the docs work continue in master, so there would be no additional gatekeeping there, just the usual review. Is everyone OK with that?
Thanks,
Ewan.
Re: [ASFCS40] CloudStack 4.0 release plan
Posted by John Kinsella <jl...@stratosec.co>.
On Aug 10, 2012, at 1:19 AM, Wido den Hollander wrote:
> That means however that you have to submit bugs fixes into master and 4.0.x, right?
yes
Re: [ASFCS40] CloudStack 4.0 release plan
Posted by Wido den Hollander <wi...@widodh.nl>.
On 08/10/2012 02:27 AM, Ewan Mellor wrote:
>> -----Original Message-----
>> From: Chip Childers [mailto:chip.childers@sungard.com]
>>
>> [Snip]
>>
>> Thinking about the milestone along those lines, I agree that cutting a
>> release branch over the weekend is a good thing. This is specifically
>> because of the pending review board submissions that we should start
>> getting integrated into master as a parallel activity to 4.0 work.
>
> OK, so I think from this and David's earlier email and general consensus that we want to get into a time-based schedule, then we're still on to start the feature freeze at the end of Friday, and have a 4.0 release branch ready for Monday morning. Everyone agreed?
+1, from monday on we'll have a 4.0 branch.
That means however that you have to submit bugs fixes into master and
4.0.x, right?
>
>> [Snip]
>>
>> - Regardless of our desire to be time-bound for our releases, our
>> first release has a higher-order issue (licensing) that will block us
>> if anything significant is outstanding.
>> - Given the importance of the licensing aspects, I believe that we
>> have to think of the "First release candidate build" milestone as
>> being predicated on the community believing / agreeing that we have
>> achieved a satisfactory level of compliance to pass an IPMC vote.
>
> Yes, makes sense to me.
>
>> [Snip]
>>
>> Let's think about "convenience builds" along two lines (because there
>> are different issues to deal with):
>>
>> Binary distribution of the core project - I think this is basically
>> sorting through the optional vs. required components. Any binary
>> distro we produce would be only include the required build targets,
>> not the optional ones. I'm not sure there is much debate open here.
>> We're headed down the right path. :-)
>
> Yep.
>
>> System VMs - I've reviewed the discussion notes from OSCON again, just
>> to refresh my memory. I might not be understanding the verbal
>> conversation correctly, but it doesn't seem to match what I see on the
>> apache.org/legal site. Specifically, David mentioned that there was
>> verbal consensus that we could possibly "prepare a system VM as we
>> distribute it now as a convenience binary". The item that I'm
>> concerned about is the "Distribution" section of the Third-Party
>> Licensing Policy [1] that says "YOU MUST NOT distribute a prohibited
>> work from an apache.org server.". Doesn't the system VM qualify as a
>> prohibited work?
>
> I agree that System VM binaries are forbidden by written policy. It is possible to ask for exceptions to this policy -- I have done so for libvirt-java, and Sam Ruby says that he'll approve that next week if no-one disagrees in the meantime. If we felt that we needed to make a similar request regarding System VMs, we can do that.
>
> Getting alternative infrastructure is a possibility. I'm sure Citrix could manage that, though I worry about the perception that would create, if the software is effectively unusable without Citrix servers.
>
> Cheers,
>
> Ewan.
>
>