Hello Andrija, Thank you so much for your quick reply. And the issue has been solved. The root cause was DB error. There were a storage_pool and its related data duplicated on DB. after deleting storage pool id, kvm hosts have recovered their status 'up' from 'alert'. I also restarted libvirt and cloudstack agent as you recommended. Thanks,
MJ On 2019/11/05 16:17:18, Andrija Panic <andrija.pa...@gmail.com> wrote: > Hi > > Before the steps below, I suggest (since your whole rack is affected...) to > verify via telnet "telnet <mgmt-server-IP-address> 8250" -that your mgmt > server is reachable - confirm that there is no connectivity issue on IP > level (firewall and such). > > Then, try the following: > - stop cloudstack agent > - restart libvirt (make sure it's restarted, I've seen cases when libvirt > would not restart due to being stuck etc - confirm that the process has the > new PID) > - start cloudstack agent > > > This will make sure your libvirt has NO pools in it (existing VMs are still > happily running), and CloudStack agents will connect to the management > server and, hopefully, everything should be fine. > > Regards, > Andrija >
