Or a different question. What's the reason that you have to create two cubes that are so similar?
On Mon, Jul 11, 2016 at 10:40 PM, ShaoFeng Shi <shaofeng...@gmail.com> wrote: > Kylin 1.5 has no change on the routing logic I think. > Creating multiple cubes with the same dim/measure/join condition in the > same project will make kylin not knowing which cube is the one you want to > query. separating them into different projects is a good approach. > > Regards， > > Shaofeng Shi > > shaofeng...@gmail.com > > From Outlook Mobile > > > > > On Mon, Jul 11, 2016 at 3:22 PM +0800, "杨海乐" <yangha...@letv.com> wrote: > > > > > > > > > > > Hi all， > We are going to popularize kylin1.1.1 in our department and > encounter > some problems： > In my opinion, kylin1.1.1 has some route rules .When I set up two > cubes > Acube（including dimension A，B，C），Bcube（including dimension A，B，D） in an > same > project and query dimension A，B ，Kylin will return the data from Acube or > Bcube 。We can't know which cube the result is from.In order to resolve the > probelm,we intend to limit the num of cross cubes in a project . For > example,a project don't contain cross cubes.we put Acube and Bcube into > different project. I want to know there is better way ? In new version > 1.5,is the problem still existing ? > > -- > View this message in context: > http://apache-kylin.74782.x6.nabble.com/some-questions-about-kylin-route-rule-tp5277.html > Sent from the Apache Kylin mailing list archive at Nabble.com. > > > > > >
