Date: Mon, 17 Jun 2002 16:35:13 -0700 (PDT) From: Nick Colakovic <nickc01@corp.firstindustrial.com> To: freebsd-gnats-submit@FreeBSD.org Subject: kern/39447: 4.5R &4.6R Kernels fail to boot w/ AHA2940U2W attached to an IFT-3102 controller Message-ID: <200206172335.g5HNZDDP077698@www.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 39447
>Category: kern
>Synopsis: 4.5R &4.6R Kernels fail to boot w/ AHA2940U2W attached to an IFT-3102 controller
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Mon Jun 17 16:40:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator: Nick Colakovic
>Release: 4.6-RELEASE
>Organization:
First Industrial Realty Trust, Inc.
>Environment:
FreeBSD stvincent.firstindustrial.com 4.4-RELEASE-p11 FreeBSD 4.4-RELEASE-p11 #2: Wed May 8 10:00:45 CDT 2002 nickc@stvincent.firstindustrial.com:/usr/obj/usr/src/sys/SSCLEAN.R44 i386
>Description:
When attempting to boot multiple host systems attached via AHA2940UW's (or the OEM equivalent) to a IFT-3102 controller the kernel locks up in the "Waiting 15 seconds for SCSI devices to settle" phase of the kernel boot process. This happens with the 4.5R and 4.6R boot floppies, but not with 4.4R. In addition an upgrade was attempted via a source upgrade from 4.4R to 4.5R. Despite playing with the kernel config to the system would fail at the same point in the kernel boot process. Infrotrend (IFT) 3102 is a SCSI-SCSI raid controller that present's its logical RAID disks as conventional SCSI drives to a conventional SCSI HBA. One indication that the kernel hasn't crashed is that the busy LED on the IFT-3102 stays on after the kernel's "Waiting 15 seconds for SCSI devices to settle" message appears. In all test cases no kernel error messages are displayed indicating any problem.
>How-To-Repeat:
System #1 is:
Intel LG440GX+ MB (Bios 13.0)
2 X P2-450Mhz
1GB ECC PC100 SDRAM
Onboard Adaptec AHA-78xx Ultra/UW controller (UW Channel A attached to IFT-3102, Firmware 2.20S1B1)
System #2 is:
Intel 810E
P3-550E
512MB PC100 SDRAM
Adaptec AHA2940U2W Controller (Firmware 2.20)
Both systems attached to seperate host channels on the IFT-3102. Both systems boot fine under 4.1R to 4.4R. Using the 4.5R and 4.6R boot floppies either system will crash the the "Waiting 15 seconds for SCSI devices to settle" phase of the kernel boot process. The systems will respond that the kernel has no debugger (as these are the generic boot floppy kernels) when the Control-Alt-Escape sequence is pressed. Despite this the kernel will not move beyond this point (even after several hours). The only solution for a reboot at this point is a hard reset. This same sequence of events was found when attempting to upgrade from 4.4R->4.5R via source.
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200206172335.g5HNZDDP077698>