I installed kylin on EMR in standalone mode. Ran a bunch of SQLs and they run fine.
Then I add a query cluster. So I have kylin.server.mode in one cluster set to "all" and one cluster in "query" mode. I set kylin.server.cluster-servers=jobnode1:7070,querynode1:7070 on both query and job cluster. Restart kylin on both Then I get this: No dictionary found by /dict/DEFAULT.KYLIN_ACCOUNT/ACCOUNT_COUNTRY/1196ba3d-22e6-0337-cfe8-aaceb02d81dd.dict, invalid cube state; cube segmentsales_cube1[20000201000000_20120401000000], col BUYER_ACCOUNT:DEFAULT.KYLIN_ACCOUNT.ACCOUNT_COUNTRY while executing SQL: "SELECT ACCOUNT_COUNTRY, SUM(PRICE) , COUNT(*) FROM KYLIN_SALES JOIN KYLIN_ACCOUNT ON KYLIN_SALES.BUYER_ID = KYLIN_ACCOUNT.ACCOUNT_ID GROUP BY ACCOUNT_COUNTRY LIMIT 200" is there some other setting that needs to be done which is not mentioned on http://kylin.apache.org/docs/install/kylin_cluster.html -- Sent from: http://apache-kylin.74782.x6.nabble.com/
