From owner-freebsd-questions@FreeBSD.ORG Thu Jul 22 22:32:39 2010 Return-Path: Delivered-To: FreeBSD-Questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7089A106566B for ; Thu, 22 Jul 2010 22:32:39 +0000 (UTC) (envelope-from nlandys@gmail.com) Received: from mail-qw0-f54.google.com (mail-qw0-f54.google.com [209.85.216.54]) by mx1.freebsd.org (Postfix) with ESMTP id 29F588FC18 for ; Thu, 22 Jul 2010 22:32:38 +0000 (UTC) Received: by qwg5 with SMTP id 5so210516qwg.13 for ; Thu, 22 Jul 2010 15:32:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:date:message-id :subject:from:to:content-type; bh=WL6cw2GOlEvF312qPbAGzIRRa4tc2phZ95CyRSPMwlw=; b=XPrkSv+x9FPsAcHAp2A7gBq4yOW0kg7SGVEBSb9ENMCn2bFJmz4KsyecLlkJSh40ED IOBnLso5Ay/y4pkaSDe9Wjecb2G6FnVDCXnlWFFlzRuqOSngjddqGwg5ADFBdnh/GGjk K4kig3ZojJc1DeOx9fs9G/wNQj86t/EZ888fo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=lDfbT4KJHrp9KBWzI9FxJNMuMRDGTHIhi9VoBm4hHmO8jvjHLraayroAnCQiliyGe5 nfjtsRRhVUXtgf2LEqEvRfSR4GnwiLFoXQ9luhYTFUiw8S7ShWGSZMRl9LOUFozocElW IScQILukKypxx3c6TjhobvrSEyfQN+z4RGM60= MIME-Version: 1.0 Received: by 10.224.65.200 with SMTP id k8mr1803302qai.47.1279837956563; Thu, 22 Jul 2010 15:32:36 -0700 (PDT) Received: by 10.229.231.193 with HTTP; Thu, 22 Jul 2010 15:32:36 -0700 (PDT) Date: Thu, 22 Jul 2010 15:32:36 -0700 Message-ID: From: Nerius Landys To: FreeBSD-Questions@freebsd.org Content-Type: text/plain; charset=ISO-8859-1 Cc: Subject: ULE scheduler and the WCPU column in top X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 22 Jul 2010 22:32:39 -0000 I have this interesting behavior in the top utility on _both_ my 7.1 and 8.0 FreeBSD servers (updated to latest patches). The interesting behavior happens only when my kernel is compiled with the ULE scheduler. It does not happen when the kernel uses the old BSD scheduler. Here is a link of a screenshot of top on one of my servers: http://daffy.nerius.com/temp/top.png The output of top usually looks just like this, at most times of the day. The row that worries me is the process "ioUrTded.i3" run by the user urt1, which is reported to be using 1.17% WCPU in the screenshot. It's the sixth row down. The thing that makes no sense is that this process is in fact using more CPU than any other process on my system, and I know this as a fact. The processes that are most active are all video game servers, and the game server run by the urt1 user is the most populated with the most going on, by far. With my kernel compiled to use the old BSD scheduler, the process run by urt1 is _always_ the most active as reported by top; the WCPU shows between 30 and 40 percent on this process normally (with the BSD scheduler). The WCPU percentage on the process owned by urt1 never reaches very high - it always stays abnormally low as reported by top (with the ULE scheduler). The process itself is running just fine and the game server is very busy. Any ideas? Is this a known issue when running the ULE scheduler? Any negative impacts that might occur?