When I've seen this (and I've seen it a lot) I usually see that it thinks it's missed a ping and so it puts the host in an avoid set. I think the "health check" feature in CS needs a lot of work to be much more reliable. Sometimes if I let the host sit long enough, it will magically (there's no rhyme or reason to it) start working again. My only sure fix has been pulling the host completely out of Cloudstack and then re-adding it.
On Jul 23, 2013, at 11:18 PM, Indra Pramana <[email protected]> wrote: > Dear all, > > Good day to you. > > From time to time, CloudStack will put a hypervisor host on avoid set and > not willing to deploy VM on that host. > > === > 2013-07-24 13:12:34,639 DEBUG [allocator.impl.FirstFitAllocator] > (Job-Executor-2:job-672 FirstFitRoutingAllocator) Host name: hv-kvm-02, > hostId: 28 is in avoid set, skipping this and trying other available hosts > 2013-07-24 13:12:34,639 DEBUG [allocator.impl.FirstFitAllocator] > (Job-Executor-2:job-672 FirstFitRoutingAllocator) Host name: hv-kvm-01, > hostId: 32 is in avoid set, skipping this and trying other available hosts > === > > 1. What are the possible reasons on why CloudStack would put a hypervisor > host on avoid set? > 2. How to manually force CloudStack to remove the hypervisor host from its > avoid set? > > Looking forward to your reply, thank you. > > Cheers.
