Red Hat Bugzilla – Bug 17161
Adding network card caused SCSI to into endless loop
Last modified: 2008-05-01 11:37:58 EDT
Distribution: RH 6.2 Kernel: 2.2.16 #10 SMP Thu Jul 27 18:41:17 CDT 2000 i586 unknown Machine: HP Vectra series, two x Pentium 90, 32MB RAM, 4.5Gb SCSI (/dev/sda) Modules: pcnet32 (eth0), tulip (eth1), tmscsim (SCSI) I got the machine working and everything was fine until I tried to add a second ethernet adapter. I put in a Netgear FA310TX card and the system seemed to work - but just about any activity on the second ethernet card (eth1) will cause the SCSI controller to go into an endless series of bus resets. The only way to recover is a hard reset (which often corrupts the disk). The ethernet card seems to work OK all the time, even if the SCSI controller isn't. If there is no activity at all on the ethernet card, everything is fine. The only thing that changed when I installed the ethernet card is eth1 and SCSI now share an interrupt and I do not know how to change that. More information: /sbin/lspci: 00:00.0 Host bridge: Intel Corporation 82434LX [Mercury/Neptune] (rev 11) 00:01.0 VGA compatible unclassified device: S3 Inc. 86c864 [Vision 864 DRAM] vers 0 00:02.0 Non-VGA unclassified device: Intel Corporation 82375EB (rev 04) 00:03.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 02) 00:04.0 SCSI storage controller: Advanced Micro Devices [AMD] 53c974 [PCscsi] (rev 02) 00:05.0 IDE interface: CMD Technology Inc PCI0640 (rev 02) 00:0e.0 Ethernet controller: Lite-On Communications Inc LNE100TX (rev 21) /proc/interrupts: CPU0 CPU1 0: 57228 61345 IO-APIC-edge timer 1: 365 764 IO-APIC-edge keyboard 2: 0 0 XT-PIC cascade 4: 17 8 IO-APIC-edge serial 8: 0 1 IO-APIC-edge rtc 9: 2732 2660 IO-APIC-edge PCnet/PCI 79C970 10: 138473 105376 IO-APIC-edge tmscsim, eth1 <----- 13: 1 0 XT-PIC fpu NMI: 1 ERR: 0 I don't know if this is a problem with the SCSI module (tmscsim), the ethernet module (tulip) or my machine. Thanks for the help.
Board problem. The BIOS is not permitted to share interrupts and assign the PCI line in question PCI shared. (I assume you were the same person who posted the PCI bus fix a bit later into bugzilla to work around your board)