Hello! 1. Yes, you should do both. 2. I'm not sure, I guess you will have to supply different Zk clusters. 3. It should probably work, but testing is needed.
Regards, -- Ilya Kasnacheev вт, 1 окт. 2019 г. в 17:00, Seshan, Manoj N. (TR Tech, Content & Ops) < [email protected]>: > Thanks. A couple of follow-up questions: > > > > 1. There doesn’t seem to be a way to specify the Port Range, without > specifying IP addresses? Could you help more specifically? > 2. How would we support multiple clusters with the Zookeeper Discovery > method? > 3. Could we use different MultiCast Groups, or a different Multicast > Port to setup the multiple clusters? > > > > Rgds > > > > *Manoj Seshan - Senior Architect* > > Platform Content Technology, Bangalore > > [image: cid:[email protected]] > > *Voice:* +91-98806 72987 +91-80-67492572 > > > > *From:* Ilya Kasnacheev <[email protected]> > *Sent:* Monday, September 30, 2019 4:16 PM > *To:* [email protected] > *Subject:* Re: How exactly does one start multiple Ignite Clusters on a > given YARN (Hadoop) cluster? > > > > Hello! > > > > You can start two clusters if they have different Discovery settings, > e.g., one cluster is active on ports 47500..47520, other on 48500..47520. > > > > You should configure clients accordingly. You can also change connector > configurations for thin clients' connecting. > > > > Regards, > > -- > > Ilya Kasnacheev > > > > > > пн, 30 сент. 2019 г. в 13:31, Seshan, Manoj N. (TR Tech, Content & Ops) < > [email protected]>: > > Please help, on how we can achieve this with the Ignite Yarn client > > > > 1. How exactly do we specify Cluster1 & Cluster2? Would we use the > “igniteInstanceName” property & Spring Configuration > > <property name="igniteInstanceName" value="Cluster1"/> > > > > 2) How would the Ignite Client specify which Cluster (Cluster1 or > Cluster2) to work with? > > > > Rgds > > > > > > *Manoj Seshan - Senior Architect* > > Platform Content Technology, Bangalore > > [image: cid:[email protected]] > > *Voice:* +91-98806 72987 +91-80-67492572 > > > >
