Feature #2856
redundant oned server
Status: | Closed | Start date: | 04/28/2014 | |
---|---|---|---|---|
Priority: | Normal | Due date: | ||
Assignee: | - | % Done: | 0% | |
Category: | - | |||
Target version: | - | |||
Resolution: | Pull request: |
Description
Hello Dev's,
it would be great if it is possible to create a redundant oned service. He is the brain of the environment.
Here is an old email from the mailing-list with the same question.
http://lists.opennebula.org/pipermail/users-opennebula.org/2011-February/014045.html
I guess the maintenance of a large environment would be easier because one server of two (or more) can be updated/ rebooted / replaced without any problems.
Or is there already a solution?
History
#1 Updated by Ruben S. Montero about 7 years ago
- Status changed from Pending to Closed
The information to configure oned in HA mode is here
http://docs.opennebula.org/4.6/advanced_administration/high_availability/oneha.html
#2 Updated by Paul Peterson about 7 years ago
Yes, I know this documentation already. But that isn't a good solution because I can't control my cloud while the 2nd oned is booting.
#3 Updated by Ruben S. Montero about 7 years ago
OpenNebula boot time should not be longer than a few seconds, you need to add the time needed to find out the service is down but that's also true for an active-active configuration. So the difference between two approaches should be those few seconds. IMHO the extra-logic needed to implement the active-active approach not worth the effort.