From owner-freebsd-scsi Fri May 30 10:06:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA26102 for freebsd-scsi-outgoing; Fri, 30 May 1997 10:06:12 -0700 (PDT) Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id KAA26092 for ; Fri, 30 May 1997 10:06:07 -0700 (PDT) Received: from x14.mi.uni-koeln.de (annexr2-40.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA19495 (5.67b/IDA-1.5 for ); Fri, 30 May 1997 19:04:30 +0200 Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id TAA04892; Fri, 30 May 1997 19:04:30 +0200 (CEST) X-Face: " Date: Fri, 30 May 1997 19:04:29 +0200 From: Stefan Esser To: Robert Schien Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: How to turn off DAT compression References: <19970530131444.53612@x14.mi.uni-koeln.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.74 In-Reply-To: ; from Robert Schien on Fri, May 30, 1997 at 06:30:18PM +0200 Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On May 30, Robert Schien wrote: > Yes, I tried a cleaning cartridge. Unfortunately, the DAT > drive ejected it 10 seconds after loading :-( Hmmm, that may be normal behaviour ... Did the information supplied with the cartridge indicate the cleaning would take longer ? > > > May 29 13:50:30 robkaos /kernel: st1: MEDIUM ERROR asc:3,2 Excessive write errors > > > > Well, this is not a driver message, but a message sent > > by your drive !!! > > And the DAT drive ? :) > Two years! Depending on the number of backups performed, the head may have been worn out and may need to be replaced ... Early DAT drives had head life times in the low thousands of hours. If you do daily backups of 2GB total, the drive will be busy for some 2 hours a day. 1000 operating hours are reached after 1.5 years, if the drive is only used for the backups ... You may find some information on claimed head life time in your drives docs ... > I don't think that the MEDIUM ERROR message is 'real'. Under normal Well, how could it NOT be real ? You've got to understand the way this message originated: The driver issued a write request, the drive didn't succeed executing it, and returned an error status. The generic SCSI driver than sent an INQUIRY command to find out about the reason for the failure, and the drive was very specific in pointing out a MEDIUM ERROR. > conditions (i.e. transfering files which are not easily compressible like > audio files) there isn't the slightest problem with the drive. > ONLY when transfering files which can be shrinked to 1/10th of their > original size by gzip the DAT drive gets crazy. My theory is that > there is some problem with the data transfer rate: when reading or > writing compressible data the transfer rate increases from a few > hundred kb/s to MB/s and my slow PPro-200 isn't capable to transfer > the data :-) No, there should never be a problem because of too low a data rate. The drive will stop streaming, and will wait for more data to arrive before it begins writing again. If your drive got an 1MB buffer, and that buffer contains uncompressed data, then writing the buffer will take some 1.5 seconds (assuming the best case compression factor of 4). The drive may start/stop in 1.5 second periods, but that should not cause it to wind the tape forward and backward. If it does, then I suspect mechanical problems in the drive. You may try without data compression, anyway, but I would not trust any backup created under this conditions! Did you check, whether tapes written on that drive can be read back on another one ? Gruss, STefan