Toni Mueller wrote at Wed, 03 Feb 2010 18:17:51 +0100: > Package: qemu-kvm > Version: 0.11.0+dfsg-1~bpo50+1 > Severity: important > > Hello, > > my virtual machine(s) become unresponsive every few days and need to be > killed then. While being unresponsive, the machine consumes 100% CPU, > fortunately only on one of my cores. Usually, some benign action > triggers the hang. Here's the latest snippet where it happened: > > $ su > Password: > debian-sid 18:58:32 /home/toni/deb/funkload > # dpkg -i funkload_1.11.0-1_all.deb > (Reading database ... 57857 files and directories currently installed.) > Preparing to replace funkload 1.10.0-3 (using funkload_1.11.0-1_all.deb) ... > Unpacking replacement funkload ... > Setting up funkload (1.11.0-1) ... > Processing triggers for python-support ... > Processing triggers for man-db ... > > At this point, nothing happened, disk I/O on the host was very low, only > one of the cores was pegged at 100%.
Toni, please verify if this bug is present in the current version of qemu-kvm in unstable (0.12.3+dfsg-4). If you don't like using unstable, try out qemu-kvm from my page (http://www.corpit.ru/debian/tls/kvm/ - grab the qemu-kvm.deb for your arch) >From what you provided above it sounds like some sort of i/o problem, but without much more details (exact kvm command line, kinds of disk devices you use, and some sorts of traces it is impossible to say what's going on. In any way, I can't reproduce anything like that here, and I haven't seen anyone complained like this before -- that's why a lot more details are needed. But please test the current version before trying to debug this one. Thanks! /mjt -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
