From owner-freebsd-hardware Tue Apr 22 00:17:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA20107 for hardware-outgoing; Tue, 22 Apr 1997 00:17:19 -0700 (PDT) Received: from milehigh.denver.net (jdc@milehigh.denver.net [204.144.180.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA20088; Tue, 22 Apr 1997 00:17:15 -0700 (PDT) Received: (from jdc@localhost) by milehigh.denver.net (8.6.12/8.6.12) id BAA06416; Tue, 22 Apr 1997 01:18:18 -0600 Date: Tue, 22 Apr 1997 01:18:18 -0600 (MDT) From: John-David Childs To: hardware@freebsd.org cc: questions@freebsd.org Subject: Yet another Adaptec 2940UW problem :( Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I recently purchased a Pentium-Pro 200 (I found back in September that Jordan made mention of "PPRO 200Mhz and new Intel 82440fx problems"...which is to a T the system I have), 64MB parity RAM, Adaptec 2940UW SCSI controller, 2 IBM Ultrastar 4.3GB Ultrawide SCSI drives, a Jaz drive, IDE CD-ROM (why for god's sake I did that instead of a SCSI CD-ROM I can't figure out ;( etc. for use as a server...and spent all day trying to get the darn thing working with 2.1.7 FreeBSD! If only I'd read the archives *before* spending $5000 on a server :( The biggest problem I had seems to be a common one with the Adaptec 2940... reference kern/3312 and kern/1296 for adaptec 2940 "timeout while {devicein, deviceout, idle}". Searching through the mailing-list archives, I found recommendations to 1) disable sync, 2) disable disconnect, 3) slow the MB/sec transfer rate, 4) disable wide negotiation, 5)turn of "ultra" negotiation, and of course 6)check cabling,termination. I also formatted/verified the drives via the Adaptec BIOS SCSI-Select Utility. I also tried adjusting the Jaz drive's MB/sec, and disabling ultra/wide transfers as appropriate for that device while leaving the other settings for the IBM UW drives alone....no luck. I've done all of the above. The server isn't even in "real" use yet and panicked five times today during relatively light transfers. Any suggestions on a replacement SCSI controller which works better?? Hellp!!! I hate looking like a fool and having my employers whine about "they should have gone with..." -- John-David Childs (JC612) http://www.denver.net System Administrator jdc@denver.net & Network Engineer Think, Listen, Look, then ACT! "A verbal contract isn't worth the paper it's written on" - Louis B Mayer