couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jan Lehnardt <...@apache.org>
Subject Re: [VOTE] Apache CouchDB 1.1.0 release, round 2.
Date Mon, 30 May 2011 09:22:42 GMT

On 30 May 2011, at 11:11, Benoit Chesneau wrote:

> On Mon, May 30, 2011 at 11:05 AM, Jan Lehnardt <jan@apache.org> wrote:
>> 
>> On 30 May 2011, at 10:50, Benoit Chesneau wrote:
>> 
>>> On Mon, May 30, 2011 at 2:23 AM, Dave Cottlehuber <dave@muse.net.nz> wrote:
>>>> On 29 May 2011 05:26, Robert Newson <robert.newson@gmail.com> wrote:
>>>>> Hello,
>>>>> 
>>>>> I would like call a vote for the Apache CouchDB 1.1.0 release, round
2.
>>>> 
>>>> Thanks Bob & all - that is a *lot* of bugs squashed!!
>>>> 
>>>> md5, sha, etaps all good.
>>>> Mac OS X 10.6.7
>>>> NB I needed to do `brew link icu4c` to get couchdb ./configure to work.
>>>> ffox passes except for replication {"error": "not_found"}; logs at
>>>> http://friendpaste.com/2IS0zh1PjEAmJrgtqWp66u
>>>> 
>>>> Swapping R14B01 in for R14B03 -> replication tests pass again as
>>>> discussed on IRC.
>>>> 
>>> 
>>> using R14B93, and still have the replication error. Tests with safari
>>> just block on replication and never end. I'm -1 until that's fixed.
>>> Will open/update a ticket with logs.
>> 
>> In case it is unclear, it is R14B03 that introduces this error, 01
>> and 02 are fine.
>> 
>> Cheers
>> Jan
>> --
>> 
>> 
> Does it means 1.1 is supporting <= R14B01 only ? In that case we
> should be clear about it in README imo. We actually we say:
> 
>  * Erlang OTP (>=R12B5)       (http://erlang.org/)
> 
> 
> Also in this case I will recheck with this version and others. (would
> be good if we had a bot doing that)

I don't think we came to any conclusions yet. I agree that we should
do so and make the correct change in the README and ./configure version
checks.

Ideally, we should be compatible with latest releases. Now, 02 introduced
a few instabilities that I believe we have worked around, and 03 fixes
them on the Erlang side. So while not advisable, we currently support <=02
and 03 is a rather fresh release.

I think it'd be worth halting this vote in favour of fixing replication.js
with 03, but I'll let Robert make that call.

Cheers
Jan
-- 













Mime
View raw message