From owner-freebsd-stable Mon Oct 1 12:51:46 2001 Delivered-To: freebsd-stable@freebsd.org Received: from weenix.guru.org (weenix.guru.org [24.163.46.66]) by hub.freebsd.org (Postfix) with ESMTP id BDAF237B40F for ; Mon, 1 Oct 2001 12:51:43 -0700 (PDT) Received: by weenix.guru.org (Postfix, from userid 3002) id 87F0025; Mon, 1 Oct 2001 15:51:42 -0400 (EDT) From: Keith Mitchell To: dillon@earth.backplane.com (Matt Dillon) Cc: freebsd-stable@freebsd.org, kmitch@guru.org Subject: Re: fs corruption (ATA / 4.4-REL) User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (FreeBSD/4.2-STABLE (i386)) Message-Id: <20011001195142.87F0025@weenix.guru.org> Date: Mon, 1 Oct 2001 15:51:42 -0400 (EDT) Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG > It's worth updating to the latest -stable kernel. At least one other > person has reported filesystem corruption due to kstack underflows which > went away when I bumped UPAGES from 2 to 3. I don't think the stack > fixes for the IDE drivers have been MFCd to -stable yet so heavy IDE use > could be hitting up against the kstack problem. After reading more, it doesn't look like my MB is likely the problem. I have been doing massive builds in ports when the problem hits. The install seems to go ok without corruption. I'm in the process of building a RELENG_4 image to try (I'm going to build a whole CD just to make sure). It's probably going to be a day or two before my new CD is ready for me to try. Is there any way from an installed 4.4.0 system to tell if I'm running into this kstack problem definatively? -- Keith Mitchell Email: kmitch@guru.org PGP key available upon request To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message