Date: Tue, 27 Aug 1996 22:23:59 +0200 (MET DST) From: Philippe Regnauld <regnauld@tetard.glou.eu.org> To: smp@csn.net (Steve Passe) Cc: hackers@freebsd.org (hackers) Subject: Re: hackers-digest V1 #1415 Message-ID: <199608272023.WAA00389@tetard.glou.eu.org> In-Reply-To: <199608271915.NAA28523@clem.systemsix.com> from Steve Passe at "Aug 27, 96 01:15:12 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Steve Passe écrit / writes: > Hi, > > > They are :-) Comments and critics welcome. > reading this I saw many things that I disagree with, however thinking > about them I realized that its not so much a matter of "right way vs. > wrong way", but just different outlooks. Well, I'd be interested in your points of view -- some of the things in there are plain common sense or luck with working configurations I've setup -- so any technically valid improvements are accepted (No, dunking the mainboard in freon to cool it is *not* valid) :-) > > mind, not fault tolerance. To start with, try and avoid all forms of > > "small tower" designs, as they have a tendency to double as ovens -- > my approach is to ALWAYS use mini-towers, but NEVER put your disks in them. > the general design of most cabinets is such that you don't get good airflow > around the disk area, no matter what you do. put the disks in external > cabinets! (not so easy with IDE disks!) Yes, and moreover using external cabinets with low-quality cables increases risks (bus resets, device timeouts, etc...). > > o always terminate both ends of the bus with the right terminator -- > > this means SCSI-I (big centronics-type connector) terminators for > > SCSI-I busses (also known as passive terminators), and SCSI-II (small > > connector) for busses containing SCSI-II devices (active > > terminators); > you can get SCSI-I terminators in both passive and active versions. this > is also true of SCSI-II and SCSI-III. > I would discuss the issue of connectors: > SCSI-I -> 50 pin centronics > SCSI-II -> 50 pin half pitch > SCSI-III -> 68 pin half pitch Right on that one. > > o internal termination by the drives/devices themselves (i.e.: jumper > > activated) should be avoided -- use clip-on terminators that fit on > > the ribbon cable like any device; > why so, modern drives provide active termination, one less connector to > cause problems... Because I've seen flaky such terminators, including reports here in -hackers: whether the jumper was in or not, there would still be problems (no probe, timeouts, hanging at boot). Also, it happens that you forget to terminate the new replacement disk when you swap out the first disk on the cable. Without mentioning when you insert a terminated disk in the middle of a chain :-) So you just get one of those long 7 connector SCSI ribbons and snap a terminator on one end. As I said, it's mostly handy tips, I'm not forcing anyone to go and buy such a terminator if they don't have one. -- Phil -- -[ Philippe Regnauld / regnauld@eu.org / +55.4N +11.3E @ Sol3 / +45 31241690 ]- -[ "To kärve or nøt to kärve, that is the qvestion..." -- My sister ]-
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199608272023.WAA00389>