Description of problem: I installed fc4 x86_64 on a brand new dual core dual cpu opteron system. Ran "yum -y update" and witnessed the system crash Version-Release number of selected component (if applicable): kernel-2.6.11-1.1369_FC4, yum-2.3.2-7, rpm-4.4.1-21 How reproducible: every time Steps to Reproduce: 1. Install fc4 2. run 'yum -y update' 3. kaboom!!! Actual results: kernel panic Expected results: system upgrade Additional info: System Specs: 2x Dual Core AMD Opteron 275 2.2GHz Supermicro H8DAR-T with IPMI (AOC-1UIPMI-B) 4GB PC3200 Reg ECC 4x250GB Seagate SATA (ST3250823AS 3Ware 9000 Series SATA Raid Controller Raid 5 across all disks
Created attachment 120526 [details] Kernel Panic
Built vanilla 2.6.14 using kernel-2.6.14-x86_64-smp.config of kernel-2.6.13-1.1526_FC4.src.rpm. Machine is working fine. Looks like problem has been fixed upstream.
After stress testing the system with kernel 2.6.14, we have had no issues. This vanilla was built using the smp config from kernel-2.6.13-1.1526_FC4.src.rpm on x86_64 platform. I tested kernel-smp-2.6.14-1.1633_FC4 from the testing branch and all looks good. The system runs stable and under heavy load. However there is a call trace that happens during shutdown when the system is unmounting nfs directories. This only shows on the kernel-smp-2.6.14-1.1633_FC4 kernel, and not the vanilla. Badness in interruptible_sleep_on_timeout at kernel/sched.c:3403 (Not tainted) Call Trace:<ffffffff80341dc0>{interruptible_sleep_on_timeout+131} <ffffffff80130981>{default_wake_function+0} <ffffffff881d0377>{:lockd:lockd_down+207} <ffffffff881e6845>{:nfs:nfs_kill_super+81} <ffffffff80186779>{deactivate_super+95} <ffffffff8019cbb5>{sys_umount+735} <ffffffff801106ba>{syscall_trace_enter+217} <ffffffff801106f7>{syscall_trace_leave+55} <ffffffff8010da80>{tracesys+113} <ffffffff8010dae0>{tracesys+209}
Created attachment 120629 [details] Call Trace Looks funky in the posting. Attaching txt file to save formatting.