Date: Thu, 3 May 2001 11:26:05 -0500 (EST) From: Mike Squires <mikes@ct980320-b.blmngtn1.in.home.com> To: The Hermit Hacker <scrappy@hub.org> Cc: freebsd-scsi@freebsd.org Subject: Re: Can a SCSI drive be *too* busy? Message-ID: <200105031626.f43GQ5L62815@ct980320-b.blmngtn1.in.home.com> In-Reply-To: <Pine.BSF.4.33.0105021446080.411-100000@mobile.hub.org> "from The Hermit Hacker at May 2, 2001 02:47:06 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
> I'm trying to debug a hanging problem on one of our servers ... two drives My information is based on experience with older hardware. Are the drives running too hot? You should be able to touch the drive board without discomfort. Drives running too hot will act strangely. Are the drives up to date on firmware? Early versions of the Seagate Barracuda and Cheetah had this kind of problem; the 9GB Barracuda wasn't stable until firmware 020. The update had to be factory-installed at the time. I also had this problem with Quantum Atlas II drives; in that case the firmware was flashable and I was able to download the firmware and flash my own drives after talking with vendor tech support. Is the controller up to date on firmware - this is especially a problem if the drives are newer than the controller. Try another controller - FBSD will boot of most of them without making any changes to the disks. I've heard of weird problems with onboard SCSI, possibly due to the lack of firmware updates (which have to come from the board vendor usually, and which may not be available at the same level as separate controllers). This is likely to be discussed on the comp.periph.scsi (sp?) newsgroup. I would also do the usual - replacing cabling, checking power supply connections and power supply, etc. I've had weird problems with a server which had a marginal (in terms of quality) power supply which put spikes onto the 5V bus when the fan started to fail. Try changing termination - if using drive termination, try an active terminator plugged into the cabling. I worked with a number of Archive 458XNP DAT changers which only worked reliably when using an NEC 6X CD-ROM as the terminating device (with an Adaptec 2940UW controller), and didn't work reliably (to the extent that they ever did work reliably) with either their own termination or a passive terminator plugged into the cable. MLS 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?200105031626.f43GQ5L62815>