cbonami commented on issue #1193: Backup/Restore Mechanism for BookKeeper URL: https://github.com/apache/bookkeeper/issues/1193#issuecomment-368594923 Ok, maybe this would save the day: In every region we add a couple of bookies that are *not* in an autoscaling group, but are autorecovering EC2-instances (including their EBS-volumes !!! we snapshot the EBS-volumes on regular basis); so when the autoscalinggroup melts down (somebody screwed up the ZK-folder, connection problems, whatever, ...) we still have all the data on the nodes outside the autoscaling group. If the latter go down, AWS restores them after a couple of minutes, and we recover our data. Of course, we'd snapshot the EBS volumes of those nodes on a very regular basis. So then we have the flexibility of the AutoscalingGroups, but also the guaranteed restorability of the data given to us by the 3 additional nodes outside the AutoscalingGroup. Does that sound as a way out here (as a countermeasure, in absence of the official solution that is yet to come)?
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
