cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Animesh Chaturvedi <>
Subject RE: [ASF4.2.1] Release Notes
Date Fri, 15 Nov 2013 18:12:00 GMT

-----Original Message-----
From: Chip Childers [] 
Sent: Friday, November 15, 2013 6:28 AM
Subject: Re: [ASF4.2.1] Release Notes

IMO, we should kill the CHAGES file and just get the release notes document under control.
 I'm fine if "Changes" is in bad shape for this release personally, as long as the release
notes are accurate.

Animesh> Yes we followed the same approach for 4.2 as I was pointed out to me that there
was a prior discussion on keeping things in release notes. I had added link to JIRA filter
for known issues and fixed issues [1].  

Another thought to remind folks about in this thread:

Changes to the cloudstack.git repo's 4.2 branch that we want to be in the 4.2.1 release will
cause a re-spin and re-vote.

Changes to the documentation repo have nothing to do with the release vote, except that we
(as a community) seem to agree that our docs should be at least updated and pushed to the
website *before* announcing 4.2.1.

Make sense?

Animesh> Yes thanks for calling it out clearly. I will update the Release Management page
on wiki for future reference.


On Fri, Nov 15, 2013 at 8:19 AM, Abhinandan Prateek <>
> Ok I will go that way till someone says that listing 175 tickets in 
> CHANGES file will needlessly clutter it.
> Can we focus the list to blockers and criticals at least ?
> -abhi
> On 15/11/13 6:34 pm, "Daan Hoogland" <> wrote:
>>Why not include the output of the query instead of the query? I think 
>>this is what Sebastien means. A list of the important ones can still 
>>be prepended in more readable form afaic.
>>On Fri, Nov 15, 2013 at 1:32 PM, Abhinandan Prateek 
>><> wrote:
>>> For listing down the fixed issues, since there are ~175 of these. I 
>>> will list down some important fixes.
>>> Followed by the query to give a exhaustive list, is that acceptable ?
>>> For known issues will look at the 4.3/4.2 open tickets list down the 
>>> important ones.
>>> This will go in the CHANGES in source repo and RN in code repo.
>>> -abhi
>>> On 15/11/13 5:54 pm, "Abhinandan Prateek"
>>> wrote:
>>>>To address the concern of RN we will not conclude the vote on RC (i.e.
>>>>make a release)
>>>>till the RN in general and upgrade instructions in particular are 
>>>>also of acceptable quality.
>>>>As for other inconsistencies will work towards ironing those out.
>>>>On 15/11/13 3:30 pm, "Sebastien Goasguen" <> wrote:
>>>>>On Nov 15, 2013, at 4:43 AM, Abhinandan Prateek 
>>>>><> wrote:
>>>>>> As a RM I had agreed to Sebatian's suggestion of fixing the docs

>>>>>>specially  the upgrade section of it.
>>>>>> And off course doing a GA after the docs are fixed is on the cards.
>>>>>> As for the list of fixed and known issues I was told that a 
>>>>>>filter is good  enough but it should be pretty easy to get the 
>>>>>>listing in the docs itself.
>>>>>> If someone has specific preferences it is easy to fix that.
>>>>>> So it boils down to get opinion from folks on the following:
>>>>>> 1. RC build, this does not contain docs. I have seen no complains

>>>>>> or issues here.
>>>>>That's fine, but releasing something without the upgrade 
>>>>>instructions committed is bad.
>>>>>Even if the release of such upgrade instructions happen after the 
>>>>>release of the code.
>>>>>> 2. Putting a full listing of bug fixes in RN Vs a filter. Even I

>>>>>> will think full listing is good or a query (instead of a URL?)
>>>>>I am in favor of consistency. Prior to 4.2 we listed all BUGS 
>>>>>We should keep doing that.
>>>>>> 3. Upgrade instructions are known to be bad and we will have to 
>>>>>>wait at  least till Wednesday to get these right.
>>>>>>     We have some volunteers already working on those and their 
>>>>>>effort is  highly appreciated.
>>>>>Right, and since there is no rush, why not wait a bit till we can 
>>>>>all look this with cool heads, double check the RN, bugs listing, 
>>>>>upgrade instructions etc...
>>>>>> -abhi
>>>>>> On 15/11/13 2:50 pm, "Daan Hoogland" <>
>>>>>>> So the -1 is because of the lack of rn and upgrade path docs,

>>>>>>>right, I  think I proposed earlier this thread to release after

>>>>>>>the doc  hackathon privided that. I wasn't really explicit about

>>>>>>>it I think as  no one commented on this strategy. Would that be

>>>>>>>acceptable to you all  (especially David and Sebastien)?
>>>>>>> I agree btw that docs must be available, but I don't think these

>>>>>>>have  to be as stable as the release. We should allow for 
>>>>>>>improving the docs  on a release if needed. The net result of

>>>>>>>what I am proposing is that  there will be a release and a docs

>>>>>>>rc. This is what the splitting of  of the docs was about in my

>>>>>>>view,. Makes sense?
>>>>>>> If not, we should not try to make CCC Europe with 4.2.1. I think

>>>>>>>this  is what the hurry is about
>>>>>>> Daan
>>>>>>> On Fri, Nov 15, 2013 at 9:14 AM, Sebastien Goasguen 
>>>>>>> wrote:
>>>>>>>> I might be behind on the discussions here, but I will veto
>>>>>>>>RC that  does not have list of bugs fixed and proper upgrade

>>>>>>>>path documented  (minimum of a fix from 4.2.0 upgrade docs).

>>>>>>>>Separate repo of the docs or  not.
>>>>>>>> Right now I see that the bugs fix list points to a jira filter.
>>>>>>>> not consistent with the way it was done in prior releases

>>>>>>>> listing) and in 4.2 (which pointed to the RN). We need consistency.
>>>>>>>> happens if someone changes this jira filter ?
>>>>>>>> I also would like to see the results of the test matrix for

>>>>>>>>4.2.1  running within  This  
>>>>>>>> runs against

>>>>>>>>master  and has been failing for a while.
>>>>>>>> PS: I did test it and did the usual smoke test
>>>>>>>> so -1 (binding) at this time
>>>>>>>> -sebastien
>>>>>>>> On Nov 14, 2013, at 4:20 PM, Chip Childers 
>>>>>>>> wrote:
>>>>>>>>> Except that the separation only helps if it allows RC
>>>>>>>>>+ voting  during doc finalization.  If we announce before
>>>>>>>>>it hurts us.
>>>>>>>>> I'm against another announcement that goes out with the
>>>>>>>>>in poor  shape.
>>>>>>>>> On Thu, Nov 14, 2013 at 3:44 PM, Animesh Chaturvedi 
>>>>>>>>> <> wrote:
>>>>>>>>>> Unless there are objection to the RC, I would prefer
to have 
>>>>>>>>>>it  released and make the announcement sooner and
showcase in 
>>>>>>>>>>collab  conference. As Chip mentions docs were broken
>>>>>>>>>>separately anyway.
>>>>>>>>>> Animesh
>>>>>>>>>> On 14/11/13 8:12 pm, "Sebastien Goasguen" <>
>>>>>>>>>>> Anyway we can wait next week to release.
>>>>>>>>>>> quite a few of us will be together in Amsterdam,
we can 
>>>>>>>>>>>dedicate a  hackathon session to 4.2.1 , make
sure RN are 
>>>>>>>>>>>good, upgrade path  etcŠthen testŠ.
>>>>>>>>>>> I'd recommend keeping the vote open until then.
>>>>>>>>>>> -sebastien
>>>>>>>>>>> On Nov 14, 2013, at 5:57 AM, Radhika Puthiyetath

>>>>>>>>>>> <> wrote:
>>>>>>>>>>>> Hi,
>>>>>>>>>>>> The master has the most up-to-date RN for
>>>>>>>>>>>> From: Abhinandan Prateek
>>>>>>>>>>>> Sent: Thursday, November 14, 2013 2:22 PM
>>>>>>>>>>>> To: CloudStack Dev
>>>>>>>>>>>> Cc: Radhika Puthiyetath
>>>>>>>>>>>> Subject: Re: [ASF4.2.1] Release Notes
>>>>>>>>>>>> It seems the upgrade section of release notes
will require 
>>>>>>>>>>>>a review,  probably followed by a revamp (?).
>>>>>>>>>>>> Can we have some volunteers who are familiar
with various 
>>>>>>>>>>>>upgrade  paths comment on it ?
>>>>>>>>>>>> Me and Radhika will try to consolidate those
>>>>>>>>>>>>snippets and  fix the RN for 4.2.1.
>>>>>>>>>>>> -abhi
>>>>>>>>>>>> RN for 4.2.1 =
>>>>>>>>>>>> f
>>>>>>>>>>>> =re
>>>>>>>>>>>> 4
>>>>>>>>>>>> .2
View raw message