Date: Tue, 20 Oct 1998 04:11:18 -0400 (EDT) From: "Matthew N. Dodd" <winter@jurai.net> To: "Kenneth D. Merry" <ken@plutotech.com> Cc: scsi@FreeBSD.ORG Subject: Re: Strange error Message-ID: <Pine.BSF.4.02.9810200354580.17054-100000@sasami.jurai.net> In-Reply-To: <199810191840.MAA24201@panzer.plutotech.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Well, I finally got annoyed enough to stay up and work on this 'strange error'. After liberating the drive from the library (no small feat) I set it up sans-case (just the mechanics and electronics sitting on a shelf) and attempted to introduce it to Win95. Win95, being its normal reliable self, froze up and attempted to make good time towards a blue screen. It appears that Win95 will retry a 'START UNIT' command followed by a 'TEST UNIT READY forever, until success. As Peter da Silva is wont to say "Bugger Bill Gates with a woodchuck." I was not pleased. My problem now looks like a hardware problem and if any of you have ever opened up a HP Magneto Optical unit you know the depths of my despair. Listening to the drive attempt to spin up a cart gave me the bright idea that the drive may have some sort of start capacitor, which, being flaky or smokeless, is causing these problems. Undaunted by the sheer complexity and small ribbon strips in the drive, I disassembled most of the major components and re-assembled them, checking contacts etc. I did not find my fabled start capacitor. (Screwy direct drive stepper.) My efforts were rewarded with the same noise, plus a rather unhealthy 'click'. Not good. I happen to recall my old, dead C1716T (the drive in question is a C1716C) and figure "what the heck, I'll keep going until I hit the pacific or I let the smoke out." I swap the good controller assy. onto the motor frame and -nothing-. Knowing the state of the unfortunate donor drive I grit my teeth and pull the ejector frame sled and perform the swap. *Presto* 30 minutes later and a library nearly full of happy MO carts I can finally see the disks, mount data and the whole nine yards. There are a few outstanding CAM problems I need to track down for Kenneth but Yippie! my cool cat scaring surplus store reject works. I wonder if I can convince the other unit to show some signs of life now. Kenneth, thanks for all the help and suggestions and thanks to everyone else who put up with my complete waste of bandwidth during my attempts to get this blasted piece of hardware working. The new CAM stuff rules alot; I'm really happy with it. Far less hoop-jumping than the last time I had to get MO hardware running under FreeBSD. On Mon, 19 Oct 1998, Kenneth D. Merry wrote: > Matthew N. Dodd wrote... > > On Mon, 19 Oct 1998, Kenneth D. Merry wrote: > > > Matthew N. Dodd wrote... > > > > On Fri, 16 Oct 1998, Kenneth D. Merry wrote: > > > > Oct 14 17:56:23 eisa-test /kernel: (da2:bt0:0:2:0): READ CAPACITY. CDB: 25 0 0 0 0 0 0 0 0 0 > > > > Oct 14 17:56:23 eisa-test /kernel: (da2:bt0:0:2:0): NOT READY asc:4,2 > > > > Oct 14 17:56:23 eisa-test /kernel: (da2:bt0:0:2:0): Logical unit not ready, initializing cmd. required > > > > Oct 14 17:56:23 eisa-test /kernel: (da2:bt0:0:2:0): fatal error, failed to attach to device(da2:bt0:0:2:0): removing device entry > > > > > > > > When I boot with media in the drive. > > > > > > > > Most annoying this behavior. I'm not sure why it thinks it needs to > > > > remove the device entry. > > > > > > Did the patch I sent fix that problem? > > > > I believe so. I get a panic every time when the media is in the drive > > though. :) dasendorderedtag or something like that. I've got problems > > with my serial console right now so I can't capture the error. > > That's not good. Send me a stack trace when you get your serial console > working. > > > > Well, I've taken a look through the spec (yes, the whole thing), and I > > > didn't really see anything that would indicate what the problem is. > > > > That was my thought as well. They went a good ways towards documenting > > the whole of the SCSI spec in that doc didn't they. :) > > Yeah, some vendors have a tendency to do that. It comes in handy > sometimes, since you don't have to look at both their docs and the scsi > spec. There is a lot of vendor-specific stuff that's documented in there, > though. > > > > So, I've got a couple of ideas: > > > > > > - Hook it up to an NT box (or some other platform that HP supports) > > > and see whether it works. I know this is distasteful, and you > > > might have to install some HP drivers to make it work, but it > > > would tell us whether there is a hardware problem or not. > > > > I've got a Win95 box so I'll see if I can get it to work. > > Yeah, sounds like a good idea. > > > > - If it works under NT, send me the changer and I'll try to get it > > > to work. We've got SCSI analyzers here and can send bus traces > > > to HP if necessary. > > > > This is a rather large unit. :) UPS probably won't take it as its around > > 100 lbs. I've got 2 of them so I've really got no problems shipping one > > to you otherwise. > > Well, I dunno about UPS, but I've seen FedEx take large packages before. > (like Pluto VideoSpace boxes, Ampex DST drives, etc.) > > > Let me play with it a bit more and pull one of them apart and see if I can > > get it working standalone outside the changer. It may be that the drive > > is just really dirty and needs cleaning. (sensors obscured etc.) Though > > it is rather odd that both units exhibit the same errors. > > > > Where are gyou guys located anyway? > > Boulder, CO. > > Ken > -- | Matthew N. Dodd | 78 280Z | 75 164E | 84 245DL | FreeBSD/NetBSD/Sprite/VMS | | winter@jurai.net | This Space For Rent | ix86,sparc,m68k,pmax,vax | | http://www.jurai.net/~winter | Are you k-rad elite enough for my webpage? | To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.02.9810200354580.17054-100000>