From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows 98) Description of problem: Kernel locks on boot w/ EPoX DV3A Motherboard w/ Dual PIII 667s. Gets to PTY: Unix98 ptys ... or I get bad EIP value kernel panic: attempted to kill the idle task! Then locks. This happens regardless of which MPS is being used (1.1 or 1.4) also happens whether APM is on or off. Note that the 2.4.7-10Enterprise kernel will boot most of the time in MPS 1.1 mode. Version-Release number of selected component (if applicable): Kernel 2.4.18-3-SMP How reproducible: Always Steps to Reproduce: 1. Boot the SMP Kernel in Red Hat 7.3 on a motherboard of same type 2. 3. Expected Results: The kernel should boot. Additional info: Video Card: Riva TNT2 Ultra CPUs: two Intel P3 667s Motherboard: EPoX EP-D3VA Raid Controler: HighPoint 370 ATA100 Raid Controler Hard Drive: Maxtor 7200 RPM ATA100 20 Gig Carefully researched and tried every related suggestion on www.linuxdoc.org to no avail.
if it's the same EIP every time, could you try to see which function that is? (The kernel will print that with the addresses)
have you try the kernel 2.4-18-5SMP?
I have a similar issue with 3 other Red Hat servers. (v 7.1 and 7.3) When using the 2.4.18-? kernels the system locks on boot everytime. I can go back on the version (2.4.9-34 or earlier) and the boot is fine. One server is an Intel 440gx server platform which needs the SMP kernel to work. The other two are MSI and gigabyte with AMD processors and I'm not using the SMP kernel. It looks like the one common thread is that they all have some type of raid controller. My systems all run off of SCSI RAID, I believe AMI controllers. What kind of information can I get for you off of these systems? This bug sounds like the same issue I'm having.
Sorry I haven't updated this bug. Apparently, all Red Hat kernels will lock if this device is attached to the USB port: SanDisk ImageMate USB (SDDR 31) Anytime this device is plugged in, the kernel will lock, otherwise it boots fine. All versions of the SMP kernels I've used act the same. I now boot the UP kernel to use the ImageMate.
Is it possible to connect serial console to the box which breaks? That EIP value is important (and the rest of the trace).
Serial console? You'll have to explain that to me. A kernel panic does not always occur, mostly the kernel stops at "UNIX 98 PTYS Configured". The SMP kernel in Red Hat 8.0 has only done that, actually. I haven't and EIP value to send.
I don't have anything like that hooked up to mine. All (3) of the computers are servers off the Internet. One of the servers that has a problem also stops at the "UNIX 98 PTYS Configured", I'm not sure on the other two since I had to get them up right after the boot. I'd need to schedule down time for those. I don't have USB devices, but there may be a controller that I'm not aware of.
How is 2.4.20-20.7 doing?