We're running CentOS 6.4 with kernel 2.6.32.358.el6.x86_64 as a kvm guest. When rebooting from the guest, the OS often goes into a bootloop between the bios (seabios) and grub. Right after the system probes EDD the system stops without logging. Sometimes I get a call trace, but unfortunately there isn't any useful info in there (system crashes, so I can't scroll in terminal). I've tried updating SeaBIOS to the latest version, but that doesn't solve the problem. I also tried to update qemu-kvm to the latest version, but that also doesn't seem to help. What does help, is compiling another (newer) kernel. For testing purposes I've compiled a 3.5 kernel and that's the only workaround we have. Seeing that we have a lot more other CentOS systems, manually compiling the kernel for each system obviously isn't a solution.
↧
