From owner-freebsd-amd64@FreeBSD.ORG Wed Jun 1 06:56:03 2005 Return-Path: X-Original-To: freebsd-amd64@freebsd.org Delivered-To: freebsd-amd64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D22C716A41C for ; Wed, 1 Jun 2005 06:56:03 +0000 (GMT) (envelope-from michiel@boland.org) Received: from neerbosch.nijmegen.internl.net (neerbosch.nijmegen.internl.net [217.149.193.38]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5814443D1D for ; Wed, 1 Jun 2005 06:56:02 +0000 (GMT) (envelope-from michiel@boland.org) Received: from brakkenstein.nijmegen.internl.net by neerbosch.nijmegen.internl.net via brakkenstein.nijmegen.internl.net [217.149.193.41] with ESMTP for id j516u1Wj005686 (8.13.2/1.4); Wed, 1 Jun 2005 08:56:01 +0200 (MET DST) Received: from localhost by brakkenstein.nijmegen.internl.net via mboland@localhost with ESMTP for id j516u1qM011908 (8.13.2/2.02); Wed, 1 Jun 2005 08:56:01 +0200 (MEST) X-Authentication-Warning: brakkenstein.nijmegen.internl.net: mboland owned process doing -bs Date: Wed, 1 Jun 2005 08:55:59 +0200 (MEST) From: Michiel Boland To: freebsd-amd64@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Subject: ELF interpreter not found during heavy lifting X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 01 Jun 2005 06:56:03 -0000 Hi. I was doing some silly stress-tests like the following while [ 1 = 1 ] do time sh ./convert-all.sh done convert-all.sh does this: for a in * do convert -scale 90% $a ../out/$a & done wait In other words, it starts a number of concurrent ImageMagick 'convert' processes. At the moment I am trying to run 720 converts simultaneously. If I start the test after a fresh reboot, I get tons of errors like the following: ELF interpreter /libexec/ld-elf.so.1 not found Abort trap but if I interrupt the test, and re-start it, things go ahead as normal. Unfortuntately after 16 or so runs of convert-all.sh, the ELF errors reappear. Who has any idea why this is? Hardware is a sun fire v20z with 2 opteron 250 CPUs and 4G mem. I am using 5.4-STABLE from last week with a GENERIC kernel, except for the addition of SMP. Cheers Michiel