Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:57573 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 15052 invoked from network); 30 Jan 2012 15:02:19 -0000 Received: from unknown (HELO lists.php.net) (127.0.0.1) by localhost with SMTP; 30 Jan 2012 15:02:19 -0000 Authentication-Results: pb1.pair.com smtp.mail=lior.k@zend.com; spf=pass; sender-id=pass Authentication-Results: pb1.pair.com header.from=lior.k@zend.com; sender-id=pass Received-SPF: pass (pb1.pair.com: domain zend.com designates 212.199.177.89 as permitted sender) X-PHP-List-Original-Sender: lior.k@zend.com X-Host-Fingerprint: 212.199.177.89 il-mr1.zend.com Received: from [212.199.177.89] ([212.199.177.89:33417] helo=il-mr1.zend.com) by pb1.pair.com (ecelerity 2.1.1.9-wez r(12769M)) with ESMTP id 5E/46-53934-9F0B62F4 for ; Mon, 30 Jan 2012 10:02:18 -0500 Received: from il-gw1.zend.com (unknown [10.1.1.22]) by il-mr1.zend.com (Postfix) with ESMTP id E9B5C607ED for ; Mon, 30 Jan 2012 17:00:41 +0200 (IST) Received: from [10.1.3.57] (10.1.3.57) by il-ex2.zend.net (10.1.1.22) with Microsoft SMTP Server id 14.1.255.0; Mon, 30 Jan 2012 17:01:34 +0200 To: PHP Internals Content-Type: text/plain; charset="UTF-8" Date: Mon, 30 Jan 2012 17:02:16 +0200 Message-ID: <1327935736.11128.15.camel@pc-liork> MIME-Version: 1.0 X-Mailer: Evolution 2.28.3 Content-Transfer-Encoding: 7bit X-Originating-IP: [10.1.3.57] Subject: Re: One more critical problem in 5.4 From: lior.k@zend.com (Lior Kaplan) > On Wed, 25 Jan 2012 09:13:43 -0800, Stas Malyshev wrote: > > > Hi! > > > >> We've just found one more critical problem in 5.4. > >> > >> Apache/PHP crashes in case of error on startup, when > >> display_startup_errors=1. It's probably related to new output API. > >> > >> I afraid it may affect many php users. > > > > OK, thanks. Looks like we'd be needing another RC then. Let's fix this > > one and the __FILE__ crash in that case. > > Fix committed. The fix does work (thanks), but introduces another problem with scripts without output in FastCGI. The problem is that the headers aren't sent when there's no output from the script. If the script does produce an output then the headers are sent. should be reproducible with this simple script: