CompuServe Thread

#A3000 Help, Please

18 messages in this thread
#81417From: Steven D. KapplinDec 14, 1992 9:18 PM
I have a new A3000 to which I have added a removable media drive on the SCSI buss. I note that unless there is a diskette in the removable drive that the system won't boot from the internal hard drive. Is this "normal" behavior for the A3000? Is it possible to get the system to boot from the internal drive without having a diskette in the removable drive? If so, how is it done? I would appreciate any help.
#81447From: Darrin RaposaDec 15, 1992 5:56 AM
Actually, If you wait long enough, the system will boot. It takes about 30 seconds for the A3000's controller to stop searching the removable drive for a boot block. This, unfortuanately is the way the A3000's internal SCSI works. I have the same problem but have gotten around it by installing a second SCSI controller in my machine (ICD AdScsi 2000), and attaching the removable to that. The other option is to simply leave the syquest turned off if you aren't going to use it.
#81489From: Steven D. KapplinDec 15, 1992 5:16 PM
Darrin, Well, in fact if the removable is turned off (BTW it's a Konica, not a Syquest, 10 meg floppy drive) the system won't boot at all. If the Konica is on, but no disk inserted, then the A3000 controller keeps searching but will never boot until a floppy is inserted into the Konica. If this is a "feature" of the controller, then I may need to do what you have, namely install a different controller card. Thanks for the suggestions.
#81560From: Mike SchillerDec 16, 1992 4:11 AM
Steven: I currently have a SyQuest, and CD-ROM on my 3000, and have in the past had a couple of different Konica drives on my machine (only 1 at a time), and I've never had the problem you mention. As was mentioned to you in an earlier message, the A3000's SCSI port will try a device for 30 seconds before giving up. If I started my machine with neither a Konica floppy, nor syquest or CDrom in the machine it's wait 1.5 minutes before booting! While 30 seconds doesn't sound like a long time, if you're waiting for the machine to boot it can seem like a LONG time! have you tried to boot without a disk in the Konica lately? If not, try it, and then go away for a while (or wait with a timer), and come back a couple of minutes later, and you should see your system booted & waiting for you. If not, perhaps you have a problem either with your SCSI, or perhaps with your Konica. BTW, I didn't sell you your Konica did I? // -Mike Schiller \X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
#81589From: Steven D. KapplinDec 16, 1992 12:43 PM
Mike, Contrary to my previous observation, the system DOES boot after about 30 seconds if the Konica has no disk inserted. However, it will not boot if the Konica is powered off at boot time. The system simply hangs. I have the Konica connected to the external SCSI port on the A3000 using a standard Mac-type SCSI cable. I have found that the internal drive must have the termination resistpaks installed in order to work correctly, even though the Konica is also terminated. Without termination on the internal drive, the system wouldn't do writes correctly, often hanging, and in one case the drive developed key errors which required reformatting. So far, I have the internal drive terminated and the Konica terminated and it all seems to work correctly except for the boot delay when a diskette isn't in the Konica or the hang if the Konica is powered off. If you have further suggestions, please leave them via EMAIL because I am leaving the country for the next two weeks. Happy Holidays.
#81704From: BILL LEACHDec 17, 1992 9:02 AM
Steven; Replying to your message to Mike…. The failure to boot with Konica powered off sure sounds like either: the A3000 terminator power diode is installed backward or the MacSCSI cable is bad. While I personally have not had a bad MacSCSI cable (and I have about a dozen), I have talked with quite a few people that have had. Other possible problems: the Konica might not have a terminator power diode at all, and no auctioneering circuit for terminator power (this is likely if the Konica is not designed to share the SCSI bus). —————- The boot delay is just something that you will have to learn to live with unless you re-order the scsi addressing and set the "last drive" bit on the internal drive. —————- Again, the motherboard is what should not be terminated when using both internal and external drives. —————- Have a good trip! -bill
#81450From: BILL LEACHDec 15, 1992 6:49 AM
Steven; You may have the "battmen" setting at 2 seconds rather than the short setting. This will cause the Ami to wait quite a long time for the drive to respond. If you also have the multiple LUN option on the time increases by a factor of 8. -bill
#81490From: Steven D. KapplinDec 15, 1992 5:19 PM
Bill, The wait time is set at 128 ms. In fact, I had to set it to 2 sec to give the Konica a chance to come up to speed on a cold boot, or the A3000 would just hang. The wait time probably needs to be increased even more since the Konica may not come up to speed fast enough even with a 2 sec wait. However, that's not the real problem. I would like to have the A3000 boot even if the Konica has NO disk inserted. Right now unless a disk is inserted in the removable drive the system won't boot at all. The controller simply keeps searching for that second drive. It won't boot if the removable is turned off, either.
#81539From: Don Curtis/SYSOPDec 16, 1992 1:07 AM
Steve, If the Konica is turned off the SCSI bus should ignore it if: 1. Termination is correct 2. The Konica is not the device that's supposed to supply TPWR Check to see if it's got a TWPR jumper and if so, if it's installed. If it is, see what happens if you remove it. If the bus termination is correct and the Konica isn't supplying TPWR, then it's defective if it hangs the bus when it's turned off. Don
#81588From: Steven D. KapplinDec 16, 1992 12:38 PM
Don, To the best of my knowledge and what's revealed in the meager docs on the Konica, it doesn't supply tpwr nor have a jumper to enable/disable it. It does, however, have a termination resistpak. What I found was: a. If the resistpaks on the A3000 drive are removed and the Konica is connected to the SCSI connector on the back of the A3000, then the A3000 develops odd write problems which hang the internal drive. b. If the resistpaks are installed on the internal drive and the resistpak is installed on the Konica, then the system works perfectly. I have found that apparently both the internal and Konica drives must be terminated for the internal drive to work correctly with the Konica connected to the A3000's external SCSI connector (using a standard Mac-type SCSI cable.) However, if the Konica is shut down (power off), then the A3000 won't boot at all. If the Konica is on, but no diskette inserted, then the A3000 will boot after about 30 seconds. If the Konica has a disk inserted at boot time, then the A3000 boots normally. I have set the SCSI timeout to 2 secs to allow the Konica to come to speed before the A3000 starts the boot sequence. I appreciate your help. If you have further suggestions for me, please leave them via EMAIL because I am leaving the country for the next two weeks. Happy Holidays.
#81605From: Steven D. KapplinDec 16, 1992 3:52 PM
Don, Since reading your message, I went back and tried removing the terminator resistpak on the Konica. The result is that the A3000 will now boot if the Konica is turned off, will boot after about 30 seconds if the Konica is on, but no diskette inserted, and will boot immediately if the Konica is on and a diskette is inserted. So this seems to have been the solution, although I cannot understand why it works when the Konica is unterminated while the internal drive IS terminated. I did find a reference in the A3000 doc on page 4-23 which says "… you will need to remove the terminator packs on all but the last external SCSI device in the chain, and all but the last internal SCSI device in the chain." This suggests that indeed the internal drive should remain terminated (it's the only internal drive), but that the single external drive should BE terminated. However, for the A3000 to boot with the Konica powered off, the Konica had to be unterminated. I can only surmise that the Konica might have TPWR without the terminator pack, or takes termination from the internal drive, or supplies active termination. The latter doesn't seem likely since the Konica has a terminator resistpak. Anyway, it now seems to work as it should, other than it takes so long for the A3000 to boot when the Konica is on but has no diskette inserted. Otherwise, read/writes seem to be OK and the system apparently is now working correctly. Thanks for the suggestions.
#81625From: Christopher TolmieDec 16, 1992 6:50 PM
Steven, "It's that old SCSI black magic, its got you in its spell…" {sung to the tune Old Black Magic} Chris
#81662From: Steven D. KapplinDec 16, 1992 9:37 PM
Chris, Sooo true! When I think I've gotten over it, it's on me like white on rice! Just when you think it's safe to get back in the water …..
#81679From: Mike SchillerDec 16, 1992 10:52 PM
Steven: If you're interested I have several copies of the Konica docs (I have several I'm selling as I mentioned earlier). If you're interested I can send you a copy of the docs. // -Mike Schiller \X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
#81572From: BILL LEACHDec 16, 1992 9:02 AM
Steven; I guess I'm comming into the middle of this so forgive me for the obvious. Have you tried setting the Konica to an address higher than all other drives in the system (ie: 6, and move drive 6 down), set the last drive bit in the rdb for the drive just below the Konica? Then try cold booting with and without a disk in the Konica (and maybe off as well). You could also try setting for 128 mSec under those conditions to see if the system will boot. SCSI mounter could then be used to make the recognized. As an aside, if you set LUN's on, the delay time can increase significantly (a factor of 8). -bill
#81590From: Steven D. KapplinDec 16, 1992 12:48 PM
Bill, No, I haven't tried your suggestions. I currently have left the internal drive at address 6 and the Konica is at address 2 (although it works similarly at addresses of 0, 1, and 3. I guess, I could try reversing the address orders. I had checked through my docs to find out what the boot order is to determine which direction the controller goes. Presently, each drive is set to the SCSI addresses indicated, but the LUN for both drives is 0. I haven't found anything in the docs about whether the LUNs should be set differently for each drive, even though they are both on the same controller. The boot priority for the internal drive is set at 1, while the priority for the Konica is set at 0 (at least on the diskette, though that should make no difference if a diskette isn't inserted.) Please enlighten me further, given what I've told you. Please leave further suggestions via EMAIL because I am leaving the country for two weeks. Happy Holidays.
#81573From: BILL LEACHDec 16, 1992 9:31 AM
Steven; I just though of something else. The system should not be bothered by the drive being off. The fact that the drive being off affects the Ami indicates that the terminators are on that drive and powered from the drive. If that is the situation then having the drive off would likely cause a boot failure. If the drive is not the last drive in the chain (terminators not installed) or terminator power is being supplied by the Ami then if having the drive off stops the boot up then the drive is probably bad (interface chips). One other possiblilty is that the Konica drive uses a subset of the scsi spec that was allowed for connecting only a single drive to a controller. Drives designed to this spec cannot be connected to multiple drive buses. -bill
#81591From: Steven D. KapplinDec 16, 1992 12:53 PM
Bill, This is a possibility since the Konica does implement a SCSI subset, but the docs don't seem to suggest that the Konica supplies TPWR (a suggestion made by Don Curtis.) If this is the case, then perhaps it explains the problem, which, obviously, is unsolvable if it's a peculiarity of the Konica. Right now I have found that unless both the internal drive and the Konica aren't terminated, then the internal drive develops write errors, symptomatic of what happens if a buss isn't properly terminated. The Konica is connected to the A3000 via the external SCSI connector using a Mac-type SCSI cable. Perhaps the nature of the cabling requires the termination of both drives in order to reduce noise and standing waves. I tried removing the resistpaks on the internal drive and leaving the Konica terminated. Although the system worked OK on reads, it didn't work correctly on writes. However, in neither case did this affect the boot problem when the Konica was powered off.