Hi, I've been working with Linux clustering for several years, mostly in Azure. However I've got a bit of a challenge right now. I'm trying to set up a "geo-cluster" and would like to direct client machines to one geo or the other based on DNS. I've read about the dnsupdate resource provider here: https://documentation.suse.com/sle-ha/15-SP1/html/SLE-HA-all/cha-ha-geo-ip-relocation.html which looks somewhat promising.
There really isn't much documentation on actually using this though. It will put the IP address in for a domain name when the resource provider runs, and will remove it when the RP stops. It's unclear how this would manage two IP addresses. Has anyone here set this up successfully? I'm wondering if you need two of these (one with the load balancer IP address on each side) with location and colocation constraints. Is it possible to have location constraints that would make one instance of dnsupdate run when the accessed resources are running in that location of the network? [https://documentation.suse.com/sle-ha/15-SP1/html/SLE-HA-all/static/images/logo.png]<https://documentation.suse.com/sle-ha/15-SP1/html/SLE-HA-all/cha-ha-geo-ip-relocation.html> Setting Up IP Relocation via DNS Update | Geo Clustering Guide | SUSE Linux Enterprise High Availability Extension 15 SP1<https://documentation.suse.com/sle-ha/15-SP1/html/SLE-HA-all/cha-ha-geo-ip-relocation.html> In case one site of your Geo cluster is down and a ticket failover appears, you usually need to adjust the network routing accordingly (or you need to have configured a network failover for each ticket). Depending on the kind of service that is bound to a ticket, there is an alternative solution to … documentation.suse.com Thanks for your help! Ross Sponholtz rsponho...@hotmail.com r...@microsoft.com
_______________________________________________ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
