Date: Fri, 3 Apr 1998 11:42:19 -0500 (EST) From: Maxwell Spangler <maxwell@clark.net> To: aic7xxx@FreeBSD.ORG Subject: HELP: Linux SCSI crashes (AHA2940) Message-ID: <Pine.GSO.3.96.980403112511.14111E-100000@shell.clark.net>
next in thread | raw e-mail | index | archive | help
My Linux system is crashing regularly and I'm looking for advice as to what might be causing my problems. I've been getting random lockups for the last few weeks: I'll be in X, doing something and then the system will lock up hard. No keys work, no disk activity, and a hard reset is the only thing that will overcome it. Last night I attempted to troubleshoot the problem: I login on F4 as 'maxwell' and run top to show system activity, but more importantly I login on F1 as 'root' and run a few simple commands like find / -name maxwell -print to excercise the disk and memory systems. Then, always during disk activity, I'll get some sort of PANIC type messages on the console, with what appears to be stack trace information, lots of debugging information. Last night I thought I caught a few SCSI error messages, but I can't remember what they said, (SORRY!). After that, I get "Aiee. Attempt to enter scsi driver reentrantly" or something along those lines. Again, sorry for the lack of specific wording. Is there any way to catch this information? I thought about connecting a 2nd PC up to COM2 as a console device and then capturing anything that comes through, but that seems a bit laborious. So, what can I test to determine what my problems are? This system consists of: RedHat 5.0 (aic7xxx driver included with that, no upgrades since, and the system has worked fine for many weeks) I have been rebuilding kernels, so I'm not ont he original RH5 based kernel. AHA2940 (original, non-ultra, non ultra-wide) +---Quantum Atlas 1G SCSI2 #0 +---IBM Spitfire 1G SCSI2 #1 +---Quantum Fireball 1G SCSI2 #2 +---Quantum LPS 240M SCSI2 #? +---Pioneer CDROM It's important to note that just a few weeks ago, I added the 240M drive. This caused me to change cables, move drives around, etc. I took that drive out of the SCSI chain last night--it was not terminated nor the last drive--and the problem still occured. Tonight I'll be trying the original cable with only the three original drives. Also in the machine: Intel Zappa motherboard with Pentium 90 CPU, CPU fan working. Kingston KNE40T PCI 10M ethernet Diamond Stealth 64 VRAM Soundblaster AWE32 Com1/Com2/parallel, serial mouse Last night I left the machine running Windows 95 (it's dual-boot) and in Windows 95, it was running Adaptec's SCSI Bench program to work the disks. It successfully managed about 10+ hours of simultaneously reading 64K blocks from the drives. I figured if there was a clear problem with the AHA2940, or the cable, or the drives, this might find it after such a long period of testing... Any help, tips, or ideas are greatly appreciated! ----------------------------------------------------------------------------- Maxwell Spangler, Software Developer, Greenbelt, Maryland, USA!SouthparkKCKSAS! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe aic7xxx" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.3.96.980403112511.14111E-100000>