From owner-freebsd-smp Thu Nov 22 6:16:58 2001 Delivered-To: freebsd-smp@freebsd.org Received: from prinz.fujimori.cache.waseda.ac.jp (prinz.fujimori.cache.waseda.ac.jp [133.9.152.239]) by hub.freebsd.org (Postfix) with ESMTP id E8CE637B417 for ; Thu, 22 Nov 2001 06:16:41 -0800 (PST) Received: from prinz.fujimori.cache.waseda.ac.jp (localhost [127.0.0.1]) by prinz.fujimori.cache.waseda.ac.jp (8.11.6/8.11.6) with ESMTP id fAMEGVw11024; Thu, 22 Nov 2001 23:16:31 +0900 (JST) (envelope-from fujimori@prinz.fujimori.cache.waseda.ac.jp) Message-Id: <200111221416.fAMEGVw11024@prinz.fujimori.cache.waseda.ac.jp> To: Ian Dowse Cc: freebsd-smp@freebsd.org Subject: Re: nfsd/ypserv die on smp In-reply-to: Your message of "Thu, 22 Nov 2001 13:24:07 GMT." <200111221324.aa00865@salmon.maths.tcd.ie> Date: Thu, 22 Nov 2001 23:16:31 +0900 From: Yoriaki FUJIMORI Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I once stopped the machine and started ktrace on it. The following is a part of ktrace.out by kdump: pid 3998 is the master nfsd, and the 3999 is its child; 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd CSW stop kernel 3999 nfsd CSW resume kernel 3999 nfsd PSIG SIGPROF SIG_DFL 3998 nfsd CSW resume kernel 3998 nfsd PSIG SIGCHLD caught handler=0x8048bf8 mask=0x0 code=0x0 3998 nfsd RET accept RESTART 3998 nfsd CALL wait4(0xffffffff,0,0x1,0) 4000 nfsd CSW resume kernel As you see in tha above, at the sizth line from the bottom, 3999 is dead. Before that point, a thousnd of similar lines of CSW appear. # I guess the scheduling between two cpus are not working. Yoriaki Fujimori To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message