Anton, Anton Altaparmakov wrote: > Dear Steve, > > Thank you very much for the quick reply! > > On 11 Mar 2010, at 15:08, Steve Mckinty wrote: >> You can't do this with AVS within a single cluster. > > Ah! > >> You can setup two single-node clusters in a Geo Cluster configuration >> and configure AVS replication between the two "sites". > > Right, that sounds like what we want. Well, do bear in mind that Geo Edition is a disaster recovery solution that combines two clusters which are assumed to be significantly far apart. By design the switchover is not automatic. It is automated, but manually triggered, per normal DR practice. That may not be what you need.
> Could you by any chance point me in > the right direction of how I do that, please? (A link to an example > documentation would be wonderful!) And just to clarify, in this > configuration I will still be able to fail over from one cluster > to the other? For the non-ZFS case, try: http://docs.sfbay.sun.com/app/docs/doc/820-6218 > > We don't yet support ZFS in this configuration for production use, > it is still undergoing QA testing, but it works Ok in my lab. > > Great, thanks! Now I just have to figure out how to do it... > Any guidance would be much appreciated! I have so far managed to > install the Sun Cluster Geographic edition on both machines. For > now I installed the quorum server on one of the machines, too Since SC Geo links two independent clusters you don't need quorum between them. For ZFS use the main thing to remember is that you'll need to delete the default disk-specific cluster device groups (format 'dsk/dN') and create a named raw device group with the same name as the zpool which contains all the devices in the pool. Then follow the guidelines for setting up HAStoragePlus, in particular setting the ZpoolsSearchDir property to /dev/did/dsk and using the same devices for the AVS replication. *Please* bear in mind that this is *not* yet supported, no guarantees that it will work correctly. By all means use it to prototype/test but not in production. You will need the latest (3.2 11/09) release as well. > though in service we would put the quorum server on a third site > (as we do with our heartbeat v2 setup at the moment). But I haven't > gotten anywhere with configuring them as all commands seems to need > to be pointed at shared storage for storing their state and the > whole point of our set up is that there is no shared storage anywhere > in the system. I'm curious about why shared storage isn't possible, since even a simple two-port SCSI disk can be used for that. Can you expand a little? It's a model that we have had requests for in the past. Hartmut's suggestion about using iSCSI is also a possible route, by using iSCSI targets on each host exported to the other host. Looks like shared storage, but isn't. I don't know what the support status of that is. Are you planning on using this in production, or just for internal testing? You could also go to the opensolaris.org site and look in the HA community. If you search for posts by Marcelo Leal you'll find his AVS agent which he developed to use with AVS *inside* a single 2-node cluster. It's a user supplied agent for Open HA cluster, so obviously isn't usable in a configuration where formal support would be required. It might be useful if you're just doing internal test/experimentation. Regards Steve > > Thanks a lot in advance for any further help you can give me! > > Best regards, > > Anton > >> Steve >> >> >> Anton Altaparmakov wrote: >>> Hi, >>> I have been trying to setup Solaris Storage AVS with Sun Cluster in the >>> hope of having a ZFS file system replicated synchronously (via TCP/IP only) >>> between two machines so that it is mounted on one machine read-write and if >>> that machine fails it is mounted read-write on the other machine. >>> I have been reading all sorts of documentation and man pages and >>> experimenting but everything I have tried immediately asks for >>> configuration of shared storage which we don't have as the two machines are >>> only connected by TCP/IP. >>> We have such a system running at the moment using Linux, iSCSI plus >>> software raid for the replication and XFS as the file system and heartbeat >>> v2 for the failover and that works well. We then have an NFS server which >>> exports the XFS file system and the NFS server is migrated together with >>> the service ip address and the XFS file system between the two nodes in the >>> heartbeat cluster but I have now spent ages trying to figure out what to do >>> with Sun Cluster and AVS to achieve the same and I am completely failing to >>> do it. )-: Would someone, pretty please with sugar on top, point me at the >>> documentation I am failing to find or alternatively giving me some pointers >>> as to which commands it is I should be using? >>> Thank you very much in advance! >>> Best regards, >>> Anton >
