From mboxrd@z Thu Jan 1 00:00:00 1970 From: linas@austin.ibm.com Date: Thu, 28 Aug 2003 12:02:44 -0500 To: Giuliano Pochini Cc: Benjamin Herrenschmidt , linuxppc-dev@lists.linuxppc.org Subject: Re: Random crashes Message-ID: <20030828120243.A50750@forte.austin.ibm.com> References: <20030824092456.3c36fa5b.pochini@shiny.it> <1061714789.753.12.camel@gaston> <20030824174408.520f2cb3.pochini@shiny.it> <1061740217.31688.33.camel@gaston> <20030827220627.1e437823.pochini@shiny.it> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20030827220627.1e437823.pochini@shiny.it>; from pochini@shiny.it on Wed, Aug 27, 2003 at 10:06:27PM +0200 Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: On Wed, Aug 27, 2003 at 10:06:27PM +0200, Giuliano Pochini wrote: > > > > Random... lockups, oopses, sig11... but nothing useful because is happens > > > in random places. Ok, so the answer is no. Maybe the hw is faulty. It's > > > the only kernel I used on this mac. Can you monitor cpu temp somehow? I had this once when a cpu fan would barely spin. Slightly off-topic: I've always wanted to have a memory-cache checker kerneld that ran continuously in the background. More off-topic. I've always wanted to have an on-line, background fsck checker running continuously. I've had problems where a journalling FS would think the FS was fine, log journal clean, but in fact, the fs had slowly accumulated errors over many months due to ?? faulty electronics ??. Biting the bullet and spending half a day on fsck would reveal the problem; but that's an unhappy way to do things. I guess i've got this jones for systems that run perfectly reliably perfectly securely all the time ... --linas ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/