It seems 0 hive records was extracted during cube build. Look into the first step of the build job and verify the hive query is correct. To see the hive query, click the parameter icon (likes a key) under the job step.
On Mon, Apr 4, 2016 at 3:19 PM, Jie Tao <[email protected]> wrote: > this may be caused by Hbase. See the message from Shaofeng: > > -------------------------------------------------------------------- > > v1.5.1 is on the way; it will include a couple of bug fixes; The next > release date hasn't been decided. You will receive the vote email when it > is kick off. > > 2016-03-31 14:54 GMT+08:00 Jie Tao<[email protected]>: > > Thanks for the message. You are right, it may be Hbase problem. I have > hbase-0.98.17-hadoop2. It may not be the Hadoop problem because I tested > with Hadoop 2.6 and got the same error: > > Java.lang.IllegalArgumentException: No enum constant > org.apache.hadoop.mapreduce > .JobCounter.VCORES_MILLIS_REDUCES > > When is the next release? I will wait. Now I work with v1.3. It is fine > with my Hbase and Hadoop. > > Best regards, > > Jie > > > > Am 30.03.2016 um 16:36 schrieb ShaoFeng Shi: > > Hi Jie, >> >> I compared the hadoop version in 1.5 and 1.3, they are almost the same, >> except hbase-hadoop2.version, which is 0.98.8 on v1.3 but 0.98.4 on >> v1.5.0). Maybe it has some impactions; if you couldn't wait for next >> release, you can clone Kylin's v1.5.0 tag, change the pom.xml, make a new >> binary package, and then deploy it in your cluster to see whether it solve >> this issue. >> >> Thank you for reporting issue to Kylin! >> >> >> >> 2016-03-29 22:41 GMT+08:00 Jie Tao<[email protected]>: >> >> I see errors in the log: >> >>> java.lang.IllegalArgumentException: No enum constant >>> org.apache.hadoop.mapreduce >>> .JobCounter.VCORES_MILLIS_REDUCES >>> >>> The problem may be my Hadoop version 2.7, according to this JIRA: >>> https://issues.apache.org/jira/browse/KYLIN-1183. >>> >>> But: v1.2 and v1.3 run on Hadoop 2.7 without any problem. Maybe you can >>> check v1.5 to find what was changed. Your suggested Hadoop version is >>> 2.4-2.7. So I work with Hadoop 2.7. >>> >>> Best regards, >>> >>> Jie >>> >>> >>> >>> 2016-03-29 17:06 GMT+08:00 Jie Tao<[email protected]>: >>> >>> Hi, >>> >>> I installed v1.5 and built the sample cube. All was fine. But when I >>>>> query >>>>> with >>>>> >>>>> select part_dt, sum(price) as total_selled, count(distinct seller_id) >>>>> as >>>>> sellers from kylin_sales group by part_dt order by part_dt >>>>> >>>>> I got No Result / Results (0). Kylin.log shows: >>>>> >>>>> User: ADMIN >>>>> Success: true >>>>> Duration: 0.172 >>>>> Project: learn_kylin >>>>> Realization Names: [kylin_sales_cube] >>>>> Cuboid Ids: [64] >>>>> Total scan count: 0 >>>>> Result row count: 0 >>>>> Accept Partial: true >>>>> Is Partial Result: false >>>>> Hit Exception Cache: false >>>>> Storage cache used: false >>>>> Message: null >>>>> >>>>> >>>>> I created a cube with my own table and also no results. I installed >>>>> v1.2 >>>>> earlier but I deleted all tables (including kylin_metadata) in Hbase. I >>>>> think there should be no problem now with version conflict. But I did >>>>> not >>>>> delte Hive table. >>>>> >>>>> What is wrong? >>>>> >>>>> Best regards, >>>>> >>>>> Jie >>>>> >>>>> >>>>> >>>>> > > Am 03.04.2016 um 15:54 schrieb kevinchen: > >> And why cube size is zero, there are data in the hbase table. >> >> -- >> View this message in context: >> http://apache-kylin.74782.x6.nabble.com/Empty-result-return-in-the-Insight-tp4034p4039.html >> Sent from the Apache Kylin mailing list archive at Nabble.com. >> > >
