> On Jan 6, 2017, at 2:23 AM, the outsider <[email protected]> wrote: > > How do you guys solve server redundancy? > I mean, what is the best way to make sure that all zones on a particular > piece of hardware can be started on a second piece of hardware when the first > one fails or needs maintenance? > > I found a script (zrep) to transport the zone data from server 1 to server 2, > but I can’t figure out how to transport the zone config from 1 to 2.
I prefer to solve the failover problem at the application level - eg; if I have a zone running a database such as Postgres or MySQL, I'd rather set those up using their own replication facilities than worry about how quickly I can shoot multi-TB ZFS images across the network in a hurried attempt to restore the stricken zone. A floating service IP would follow whichever zone is then the active one for that service; or other appropriate means of network redirection. Obviously, being able to do this is very particular to what you are running on a given zone, and perhaps you will need to back up ZFS datasets and restore them using tools like zrep or zetaback. As for the zone configuration themselves, you can get a list of zones with 'vmadm list', and do a 'vmadm show' on each uuid, pipe that to a file, and back those up along with their filesystems. /dale ------------------------------------------- smartos-discuss Archives: https://www.listbox.com/member/archive/184463/=now RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00 Modify Your Subscription: https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb Powered by Listbox: http://www.listbox.com
