From owner-freebsd-scsi Mon Apr 24 13:56:40 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from verdi.nethelp.no (verdi.nethelp.no [158.36.41.162]) by hub.freebsd.org (Postfix) with SMTP id 7430837BBE5 for ; Mon, 24 Apr 2000 13:56:32 -0700 (PDT) (envelope-from sthaug@nethelp.no) Received: (qmail 75154 invoked by uid 1001); 24 Apr 2000 20:56:29 +0000 (GMT) To: mjacob@feral.com Cc: ken@kdm.org, freebsd-scsi@FreeBSD.ORG Subject: Re: Safe mounting of one file system on multiple hosts? From: sthaug@nethelp.no In-Reply-To: Your message of "Mon, 24 Apr 2000 13:48:22 -0700 (PDT)" References: X-Mailer: Mew version 1.05+ on Emacs 19.34.2 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Mon, 24 Apr 2000 22:56:29 +0200 Message-ID: <75152.956609789@verdi.nethelp.no> Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Define 'problems'. Like this: (da4:ahc2:0:0:2): SCB 0x0 - timed out while idle, LASTPHASE == 0x1, SEQADDR == 0xa (da4:ahc2:0:0:2): Queuing a BDR SCB (da4:ahc2:0:0:2): Bus Device Reset Message Sent (da4:ahc2:0:0:2): no longer in timeout, status = 34b ahc2: Bus Device Reset on A:0. 10 SCBs aborted (da4:ahc2:0:0:2): SCB 0x2b - timed out while idle, LASTPHASE == 0x1, SEQADDR == 0x8 (da4:ahc2:0:0:2): Queuing a BDR SCB (da4:ahc2:0:0:2): Bus Device Reset Message Sent (da4:ahc2:0:0:2): no longer in timeout, status = 34b ahc2: Bus Device Reset on A:0. 26 SCBs aborted (da4:ahc2:0:0:2): SCB 0x19 - timed out in datain phase, SEQADDR == 0x5e (da4:ahc2:0:0:2): Other SCB Timeout (da4:ahc2:0:0:2): SCB 0x17 - timed out in datain phase, SEQADDR == 0x5e (da4:ahc2:0:0:2): Other SCB Timeout Unexpected busfree. LASTPHASE == 0x40 SEQADDR == 0x5e ahc2:A:0: no active SCB for reconnecting target - issuing BUS DEVICE RESET SAVED_TCL == 0x1, ARG_1 == 0x15, SEQ_FLAGS == 0x0 ahc2: Bus Device Reset on A:0. 14 SCBs aborted As I mentioned, if I turn off tagged queuing on the host channel with the two LUNs, everything seems to work just fine - but somewhat slower for certain operations, of course. Steinar Haug, Nethelp consulting, sthaug@nethelp.no > > On Mon, 24 Apr 2000 sthaug@nethelp.no wrote: > > > > I would recommend that you test out the controller carefully under high > > > load conditions if you have more than one array configured on it. > > > > > > There have been a number of reports (see the list archives) about problems > > > with CMD RAID arrays with more than one LUN (i.e. more than one array) > > > configured. > > > > I have two LUNs on one of the host channels, and I can easily provoke > > problems here (running a few copies of bonnie, for instance) with > > tagged queueing enabled. If I turn off tagged queueing on the host > > channel with multiple LUNs (as suggested by Oliver Blasnik > > ), there are no problems. > > > > Steinar Haug, Nethelp consulting, sthaug@nethelp.no > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-scsi" in the body of the message > > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message