robb1e wrote: > odd, I've been running in MANAGED all the time and I'm also seeing these > issues.
My hunch for my local CPU-burning (MANAGED better than SYSTEM but still bad ~30%) is that the failing communication with the eucalyptus-nc provokes the eucalyptus-cloud to power-dump repetitive redundant no-one-will-look-at annoying error descriptions to the log files. And those non-surprisingly not only cost IO but also CPU time. Nodes that don't work should possibly be forgotten about after some initial attempts, at least the time until trying again should increase. Maybe it could be the nodes that introduce themselves to the server, instead? I can grant access to my laptop should that be helping with debugging. But I can just wait for 1.6. Cheers, Steffen -- high CPU usage from cloud controller https://bugs.launchpad.net/bugs/403969 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
