From owner-freebsd-current@FreeBSD.ORG Sun Jul 22 21:27:50 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 70AAB16A41B for ; Sun, 22 Jul 2007 21:27:50 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.layeredtech.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 4536E13C46A for ; Sun, 22 Jul 2007 21:27:50 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l6MLRlQ2007932; Sun, 22 Jul 2007 16:27:48 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46A3CBCE.3000300@freebsd.org> Date: Sun, 22 Jul 2007 16:27:42 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.5 (Macintosh/20070716) MIME-Version: 1.0 To: freebsd-current@freebsd.org, gurney_j@resnet.uoregon.edu References: <469E3B7C.4020402@centtech.com> <20070722001514.GB1221@funkthat.com> In-Reply-To: <20070722001514.GB1221@funkthat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: Subject: Re: processes stuck in kqread X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 22 Jul 2007 21:27:50 -0000 John-Mark Gurney wrote: > kevin kramer wrote this message on Wed, Jul 18, 2007 at 11:10 -0500: >> I have been having issues for some time on -CURRENT. My latest update >> was 7/11. I have identified these processes as being stuck in kqread >> while waiting for them to continue. >> >> Processes that I know are a problem: >> sshd - I can watch a login and the process goes from Select to kqread, >> takes about 3-4 minutes to get a login prompt >> top - takes about 2-3 minutes to come up >> tar - never really even starts to read data in and stays in kqread >> >> Processes not affected: >> scp >> gstat >> cvsup >> make >> >> This machine is running AMD64 GENERIC kernel on a Dual-Core Xeons. >> dmesg.txt and top-trace.txt here, http://users.centtech.com/~kramer/snapshot >> >> What do I do to debug this? > > running the process under ktrace would help identify what it is waiting > for... Though if it usually hangs for a minute or two in kqread, and > then continues, it's probably a DNS lookup issue... > You mean like the one he put here: http://users.centtech.com/~kramer/snapshot/top-trace.txt :) A quick snippet: 84667 top 0.006110 RET sendto 56/0x38 84667 top 0.006128 CALL kevent(0x6,0x800c11910,0x1,0x7fffffffdad0,0x1,0x7fffffffdb10) 84667 top 5.008057 GIO fd 6 wrote 32 bytes 0x0000 0500 0000 0000 0000 ffff 0100 0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 |................................| 84667 top 5.008073 GIO fd 6 read 0 bytes "" 84667 top 5.008078 RET kevent 0 84667 top 5.008086 CALL gettimeofday(0x7fffffffdb20,0) 84667 top 5.008091 RET gettimeofday 0 84667 top 5.008096 CALL sendto(0x5,0x800c11ac4,0x38,0,0x800c11808,0x10) 84667 top 5.008139 GIO fd 5 wrote 56 bytes 0x0000 4696 2a64 0000 0000 0000 0002 0001 86a0 0000 0002 0000 0003 0000 0000 0000 0000 0000 0000 0000 0000 |F.*d....................................| 0x0028 0001 86a7 0000 0002 0000 0006 0000 0000 |................| 84667 top 5.008155 RET sendto 56/0x38 84667 top 5.008163 CALL kevent(0x6,0x800c11910,0,0x7fffffffdad0,0x1,0x7fffffffdb10) 84667 top 15.007883 GIO fd 6 wrote 0 bytes "" Eric