Date: Tue, 18 Jul 1995 12:15:01 -0700 (PDT) From: "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com> To: paul@FreeBSD.org Cc: FreeBSD-current@FreeBSD.org Subject: Re: scsi problem solved Message-ID: <199507181915.MAA05141@gndrsh.aac.dev.com> In-Reply-To: <199507181804.TAA00314@server.netcraft.co.uk> from "Paul Richards" at Jul 18, 95 07:04:18 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > In reply to Rodney W. Grimes who said > > > > Humm, this really suprizes me as the usually symptom of a device that > > does not understand sync when probed by the 1542 bios is a scsi bus > > lockup during POST in the 1542 bios. You rarely get to where you > > can boot the system, and I have never seen it actually transfer data > > when this is the cause of the problem. > > > > Really really strange, I am going to have to remeber this one!! Can > > you enlightenme with some details about which model of the 1542 you > > have (B/C/CF) and just what model dat drive this is (dmesg output > > would probably do for both since I think we now print the 1542 board > > id info, but maybe that is only for boot -v. > > aha0: AHA-1542CF BIOS v2.01-VE.0, enabling mailbox, enabling residuals > aha0: reading board settings, dma=5 int=11 (bus speed defaulted) > aha0 at 0x330-0x333 irq 11 drq 5 on isa > (aha0:0:0): "SEAGATE ST15150N 0019" type 0 fixed SCSI 2 > sd0(aha0:0:0): Direct-Access 4095MB (8388315 512 byte sectors) > (aha0:5:0): "HP C1533A 9406" type 1 removable SCSI 2 > st0(aha0:5:0): Sequential-Access density code 0x24, variable blocks, write-enabled > (aha0:6:0): "TOSHIBA CD-ROM XM-3601TA 0265" type 5 removable SCSI 2 > cd0(aha0:6:0): CD-ROM cd present.[307527 x 2048 byte records] Hummm okay, that is really strange, every thing there is claiming scsi 2 compliancy yet it has problems with sync :-(. This is really starting to smell of cable and termination (I know, we have been down that road :-(). > I ran into more problems when I added the cdrom though and I'm now running > with adpatec bios default settings, which is no scsi-ii and no sync on > anything. We'll see if it survives a bit longer this time.... Ahh.. longer chain.. okay.. just for my enjoyment (:-)), what is switch 1 set to on the aha1542CF? I remeber you saying all your devices are external. Well, I have heard that sometimes the software termination enable just does not work quite right on some 1542CF's. Turn switch 1 on which hardwires the terminators on the board to the on possition and ignores the sofware setting. I also suspect that the 3601TA was added to the end of this chain, and am pretty sure that it uses passive 220/330 ohm termination, try switching the order of things so that your ST15150N which should have active terminations is the last drive in the chain. Also from the aha2940 cheat sheet make sure your total scsi cable length is 3 meters (9.8 feet) or under (this is the fast scsi-ii recomendation, they say 6 meters for scsi-i, but we want to be as safe as possible here in trying to find the problem). Also that is a 1542CF, which is known to have some very fast edges on the scsi bus drivers and really wants to have scsi-ii double shielded 110 ohm cables to work reliably. Anything less is asking for trouble with it :-(. You can take the exact same working setup that is totally within spec running on a 1542B and plug it into a 1542CF and have it give you royal fits until you upgrade the cables even if everything is running at 3MB/sec scsi-i async speeds :-(). First time I saw this problem was at WC when we upgraded the controller for the cdrom burner from an ancient 1542 that died to a 1542CF, it proceded to destroy 5 or 6 attempts at one off mastering before I finally replaced the external 1 meter (really a 3 footer) old scsi-i cable with a really nice scsi-ii double shielded job, then it purred like a kitten. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199507181915.MAA05141>