Mailing-List: contact user-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@couchdb.apache.org
Received-SPF: pass (athena.apache.org: domain of adam.kocoloski@gmail.com
 designates 209.85.221.191 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=sender:message-id:from:to:in-reply-to:content-type
         :content-transfer-encoding:mime-version:subject:date:references
         :x-mailer;
        b=hu8Hr8lcaKKNqamGTA8PI0J5xbda6ef5EDJOUjJKQfBsWKLlczt4nitQehqIipSE0g
         F6a+r9gD1+IYOJCMFeUei3FFHrRCC/LroLJPcoTgbNOEH5JM2SNTIX/IktI0c901Nn11
         eSbl3kjoN/g/gsV1bUAJBFZ00PnMaLDbk61Rg=
Sender: Adam Kocoloski <adam.kocoloski@gmail.com>
Message-Id: <6E03EE6F-09F5-4522-8ED7-F0D31FEE5E11@apache.org>
From: Adam Kocoloski <kocolosk@apache.org>
To: user@couchdb.apache.org
In-Reply-To: <4A534ED3.8040307@vpro.nl>
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v935.3)
Subject: Re: Apache in front of CouchDB: failing tests
Date: Tue, 7 Jul 2009 09:44:42 -0400
References: <4A530D2D.8010808@vpro.nl>
 <FC19F144-DBAC-4B36-951D-D92B3AE1650F@dionne-associates.com>
 <4A534ED3.8040307@vpro.nl>

On Jul 7, 2009, at 9:34 AM, Nils Breunese wrote:

> Robert Dionne wrote:
>
>> On Jul 7, 2009, at 4:54 AM, Nils Breunese wrote:
> >
>>> stats: # Assertion 'open_databases > 0 && max >= open_databases,  
>>> name'
>>>        failed: should keep the same number of open databases when
>>>        reaching the max_dbs_open limit
>> This was a long standing bug with initializing the stats collector  
>> and  is now fixed in trunk
>
> So it's only the stats collector that is not functioning properly?  
> Can we just use CouchDB 0.9.0 if we do not really care about the  
> stats collector?
>
> This test is not failing on our testing CouchDB (also 0.9.0) by the  
> way. What is the explanation for that?
>
> Nils.

Hi Nils, the bug is also fixed in the 0.9.1 release candidate being  
voted on now.  It's a bug in the stats collector and the test suite.   
The stats collector was not being reset properly on a call to  
_restart, and the test suite was measuring the wrong field.  The  
latter bug is probably why you only saw intermittent failures or none  
at all in 0.9.0.

In short, the suite was measuring the maximum rate of DBs being  
opened, when it should have measured the number that were open.  So  
the results of the test really depended on the environment and the  
hardware.  Hope it helps,

Adam