we have tried some poc's to achieve this sometime back 2011/2/17 Tino Vazquez <[email protected]>
> Hi YE Lei, > > The first issue that I see here is the migration, since it is > impossible to migrate a VM from a host that is down. > > What is possible is the reestablishment of the service in placeB, > where the VM images should be previously placed (to avoid transferring > them, although this is possible as well), so only the VM state (in the > form of snapshots of memory, and deltas of what has changed in the > disk images since the VM was up) needs to be transferred. This > snapshots should reside in a shared filesystem (to avoid losing access > when the host is down), or synced in placeB using DRDB for instance. > since place A and place B are quiet distant and we did not want to share the same file system, we used rsync to keep the disks in check. if the file system is a failure at place A then the latest image would used at placeB else the images are synced before starting. OpenNebula at place A can detect that the hosts are down and start the > reestablishment of the service talking with the OpenNebula at place B. > This setup can be achieved with some minor adjustments to the > out-of-the-box OpenNebula. > > Can you give any pointers on this, as i have achieved it using a script above oned to detect failures and request place B to start the vm's after syncing with the current vm. I would like to know how oned can be configured to talk to placeB ( i suspect you are talking about hooks?). > For extra protection, the OpenNebula at placeA could be running two > instances in different servers, to avoid the fail in the disaster > recovery in case the OpenNebula front-end goes down. > > isn't this being discussed in another thread about oned redundancy?. which is presently not present in one. This is an interesting use case, we will be happy to provide > assistance in the setup of this scenario. > > Best regards, > > -Tino > > -- > Constantino Vázquez Blanco, MSc > OpenNebula Major Contributor / Cloud Researcher > www.OpenNebula.org | @tinova79 > > > > On Thu, Feb 17, 2011 at 10:22 AM, YE LEI <[email protected]> > wrote: > > For example, we have a product (in VM) running in the servers on placeA, > > and in case disaster happens on placeA, all servers on placeA are > > damaged, could opennebula migrate this product to the severs on > > placeB(thousands miles away from placeA) which are managed by opennebula > > in the cloud? > > > > Best Regards, > > YE Lei > > > > -----Original Message----- > > From: Ignacio M. Llorente [mailto:[email protected]] > > Sent: Thursday, February 17, 2011 4:55 PM > > To: YE LEI > > Cc: [email protected] > > Subject: Re: [one-users] Does Opennebula 2.0.1 support disaster > > recovery? > > > > Please, could you elaborate on the use case for this? Which > > functionality would you require? > > > > Thanks > > > > On Thu, Feb 17, 2011 at 7:58 AM, YE LEI <[email protected]> > > wrote: > >> Hello, > >> > >> As mentioned in the mail title, does Opennebula 2.0.1 support disaster > >> recovery cross-regional? > >> > >> Best Regards, > >> > >> YE Lei > >> > >> _______________________________________________ > >> Users mailing list > >> [email protected] > >> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org > >> > >> > > > > > > > > -- > > Ignacio M. Llorente, PhD, MBA > > Project co-Lead and Director > > OpenNebula - The Open Source Toolkit for Cloud Computing > > www.OpenNebula.org | [email protected] > > _______________________________________________ > > Users mailing list > > [email protected] > > http://lists.opennebula.org/listinfo.cgi/users-opennebula.org > > > _______________________________________________ > Users mailing list > [email protected] > http://lists.opennebula.org/listinfo.cgi/users-opennebula.org > -- K Manikanta Swamy +919059014442
_______________________________________________ Users mailing list [email protected] http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
