Mailing-List: contact users-help@trafficserver.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@trafficserver.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Date: Tue, 15 Apr 2014 15:54:41 -0500
From: "Alan M. Carroll" <amc@network-geographics.com>
Organization: Network Geographics, Inc.
Message-ID: <558096448.20140415155441@network-geographics.com>
To: Phil Sorber <users@trafficserver.apache.org>
Subject: Re: [VOTE] Release Apache Traffic Server 4.2.1 (RC0)
In-Reply-To: 
 <CABF6JR1QJ=0w8_0X+rGjV6HFqs6Ffdi9kL46ZjmRSFpLR5bjYQ@mail.gmail.com>
References: 
 <CABF6JR1QJ=0w8_0X+rGjV6HFqs6Ffdi9kL46ZjmRSFpLR5bjYQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

-1

I think there's still a bug, demonstrated by some testing Phil has done. We have a theory on the cause and Phil is setting up the testing again with a fix. Our view is that the previous patch introduced a race condition involving the ram cache and this is the cause of the crash. The fix is to not do the 4.2.0 fixup on objects from the ram cache, as they have already had it done when read from disk. This is not directly a problem because the 4.2.0 is idempotent, but it can cause a race but not doing the fixup avoids the race condition.