Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 7 Jun 2002 03:46:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 7 Jun 2002 03:46:34 -0400 Received: from ip213-185-39-113.laajakaista.mtv3.fi ([213.185.39.113]:23729 "HELO dag.newtech.fi") by vger.kernel.org with SMTP id ; Fri, 7 Jun 2002 03:46:34 -0400 Message-ID: <20020607074629.27617.qmail@dag.newtech.fi> X-Mailer: exmh version 2.5 07/13/2001 with nmh-0.27 To: Richard Gooch cc: Ruth Ivimey-Cook , Dag Nygren , linux-kernel@vger.kernel.org, dag@newtech.fi Subject: Re: Devfs strangeness in 2.4.18 In-Reply-To: Message from Richard Gooch of "Thu, 06 Jun 2002 17:04:15 MDT." <200206062304.g56N4Fg05480@vindaloo.ras.ucalgary.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 07 Jun 2002 10:46:29 +0300 From: Dag Nygren Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > Ruth Ivimey-Cook writes: > > On Thu, 6 Jun 2002, Dag Nygren wrote: > > > > >The problems are tha the sg? links doesn't correspond to the real > > >devices shown by /proc/scsi/scsi (Which matches the real situation) > > >sg0 matches the first disk, OK > > >sg1 matches the Medium changer, OK > > >sg2 matches nothing...... There is no target 2 on host1 !!! > > >sg3 matches the DLT tape drive > > >sg4 matches the DAT tape drive > > > > > >The other problem is the st? links. > > >st0 is linked out into nothing ... > > > > > >Seems like 3 host adapters is too much for devfs...... > > >Do I need an upgrade ? > > > > In my experience, devfs doesn't create /dev/sg or /dev/st softlinks. > > Indeed. It's devfsd that creates it. > > > The only links it creates are from /dev/discs/... to /dev/ide/... or > > /dev/scsi/... as appropriate. > > > > I would look into the mandrake boot sequence in detail. > > Also check that you don't have bogus entries in your dev-state > area. Mandrake had some configuration problems a few months back. Yess!!! That's what did it, removing the sg? and st? entries from /lib/dev-state did the trick. There were some oldies ghosting there. Thanks a lot Richard, I didn't even know that there was a dev-state directory to look for ;-). Is there any comprehensive documentation on devfsd and devfs anywhere on the net? Could be good to read a bit more about this. BRGDS Dag - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/