From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0; DigExt) Description of problem: Upon init 6 and "reboot" commands on a console or in Xwindows, the system hangs at the console on "rebooting system" instead of resetting the power. All services are reporting to be shut down before hanging, and it takes a very long time to see any services being shut down. System is a Dell PowerEdge 300 with 2 processors and 128 MB of RAM. Occurs on SMP kernel only. How reproducible: Always Steps to Reproduce: 1. Load Fairfax beta 2 on a PowerEdge 300 2. Give an "init 6" or "reboot" command 3. System shuts down all services but hangs at "rebooting system" instead of resetting power. Actual Results: System will not reset power Expected Results: System power should reset. Additional info:
Does passing "reboot=b" to the lilo/grub commandline help? If so, could you please run the http://people.redhat.com/arjanv/dmidecode.c program and send the output to hardwarebugs-list with the subject "needs reboot=b"
This defect is considered SHOULD-FIX for Fairfax.
putting reboot=b or reboot=bios on the kernel line in grub.conf does not fix this problem.
The "reboot=b" fixes this problem. Sent output from dmidecode.c to hardwarebugs-list
Fixed in next build
Is this fix for all processor speeds, or just the 800 MHz? We need it for all...
Uhm I assume the "800" in the output is variable then.... ok will fix thanks for the warning.
The Power Edge 300 needs the reboot=smp or reboot=smb,bios passed in order to make it reboot consistently.
What kernel version ?
Josh, it automatically should pick up 'reboot=bios'. Does it always need 'reboot=bios,smp' instead? (looking at 2.4.7-2.19 now) dmesg should show it detecting the PE300 and reporting it's going to always do BIOS reboots.
If it doesn't pick it up, or gets it wrong I'd like dmidecode output so I can fix it asap
Changed to smp,bios just to be safe