From owner-freebsd-stable@FreeBSD.ORG  Wed Jun 27 11:27:56 2007
Return-Path: <owner-freebsd-stable@FreeBSD.ORG>
X-Original-To: stable@freebsd.org
Delivered-To: freebsd-stable@FreeBSD.ORG
Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52])
	by hub.freebsd.org (Postfix) with ESMTP id CA60B16A400
	for <stable@freebsd.org>; Wed, 27 Jun 2007 11:27:56 +0000 (UTC)
	(envelope-from citrin@citrin.ru)
Received: from mail.classis.ru (classis.ru [213.248.60.120])
	by mx1.freebsd.org (Postfix) with ESMTP id 8A79F13C45B
	for <stable@freebsd.org>; Wed, 27 Jun 2007 11:27:56 +0000 (UTC)
	(envelope-from citrin@citrin.ru)
Received: from citrin.rambler.stack.net (unknown [81.19.65.20])
	(using TLSv1 with cipher AES256-SHA (256/256 bits))
	(No client certificate requested)
	(Authenticated sender: citrin.citrin.ru)
	by mail.classis.ru (Postfix) with ESMTP id A20C61227B42
	for <stable@freebsd.org>; Wed, 27 Jun 2007 15:16:27 +0400 (MSD)
Date: Wed, 27 Jun 2007 15:16:26 +0400
From: Anton Yuzhaninov <citrin@citrin.ru>
X-Mailer: The Bat! (v3.62.14) Professional
Organization: Rambler
X-Priority: 3 (Normal)
Message-ID: <837624448.20070627151626@citrin.ru>
To: stable@freebsd.org
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Cc: 
Subject: vfs/bio livelock on 6.2
X-BeenThere: freebsd-stable@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Production branch of FreeBSD source code <freebsd-stable.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>, 
	<mailto:freebsd-stable-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-stable>
List-Post: <mailto:freebsd-stable@freebsd.org>
List-Help: <mailto:freebsd-stable-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>,
	<mailto:freebsd-stable-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 27 Jun 2007 11:27:56 -0000

Hello,

After repair table for big (~3 Gb) fragmented (varchars) table system
was entered the livelock.
(repair of fragmented table case huge number of small pread/pwrite).

In livelock all processes respond with delay 60-120 seconds (it seems
than almost all syscalls delayed).

mysql don't stop in more than hour after kill

top show almost all processes in flswai state.

On console was seen messages like:
g_vfs_done():da2[WRITE(offset=xxx, length=xxx)]error=6

gstat show ~30k writes/s and 800 Mb/s for da2

on da2 located mysql data files.

System was rebooted after ~16 hours in this state.

Problems seems to be identical to this:
http://docs.FreeBSD.org/cgi/mid.cgi?20010907202924.A6811

Environment:
6.2-RELEASE-p3 i386, SMP Kernel
Adaptec AIC7902 Ultra320 SCSI adapter
da2 - IBM DDYS-T18350M S93E

Any suggestions?

-- 
 Anton Yuzhaninov.