Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Duplicate
-
None
-
None
-
None
Description
Hi All
I'm trying to build a cube from a table not in a Database Default
but it seems that Kylin only use a 'table name' without 'database name' when executing a hive script.
Below is my 'kylin_metadata_qa' and error message from job execution.
I guess Kylin get 'TABLE_NAME' from this meta info.
I know any misunderstanding can be mine.
Any help would be appreciated!
Thanks in advance.
-
-
-
-
- Cube Meta in hbase ( see key 'table' which was loaded from 'sample.test_fact' )
```
hbase(main):039:0* scan 'kylin_metadata_qa', {LIMIT => 10, STARTROW => '/cube_desc/cube_test.json' , ENDROW => '/cube_desc/cube_test.json'}ROW COLUMN+CELL
/cube_desc/cube_test.json column=f:c, timestamp=1419223117348, value={\x0A "uuid" : "e6afd141-6a88-4325-bfd6-b2688b976822",\x0A "name" : "cube_test",\x0A "description" : "",\
x0A "dimensions" : [Unknown macro: {x0A "id" }],\x0A "measures" : [ {\x0A "id" : 1,\x0A "name" : "COUNT",\x0A "function" :
Unknown macro: {x0A "expression" },\x0A "dependent_m
easure_ref" : null\x0A } ],\x0A "rowkey" :Unknown macro: {x0A "rowkey_columns" },\x0A "signature" : "mHAbTS09j0sIKU1omL
{\x0A "partition_date_column" : null,\x0A "partition_date_start" : 0,\x0A "cube_partition_type" : "APPEND"\x0A }
+4xg==",\x0A "capacity" : "MEDIUM",\x0A "last_modified" : 1419223117344,\x0A "fact_table" : "TEST_FACT",\x0A "null_string" : null,\x0A "filter_con
dition" : null,\x0A "cube_partition_desc" :,\x0A "hbase_mapping" : {\x0A "column_family" : [
Unknown macro: {x0A "name" }]\x0A },\x0A "notify_list" : [ ]\x0A}
/cube_desc/cube_test.json column=f:t, timestamp=1419223117348, value=\x00\x00\x01JpJ.
1 row(s) in 0.0180 seconds
```
- Cube Meta in hbase ( see key 'table' which was loaded from 'sample.test_fact' )
-
-
-
-
-
-
-
- Hive execution error log ( see table name )
```
[QuartzScheduler_Worker-8]:[2014-12-22 14:52:17,829][INFO][com.kylinolap.job.flow.JobFlowNode.execute(JobFlowNode.java:88)] - cmd:mkdir -p /tmp/kylin/logs;set -o pipefail; hive -e "DROP TABLE IF EXISTS kylin_intermediate_cube_test_FULL_BUILD_c0812af1_76fa_410b_b782_cda4b0abd855;
CREATE EXTERNAL TABLE IF NOT EXISTS kylin_intermediate_cube_test_FULL_BUILD_c0812af1_76fa_410b_b782_cda4b0abd855
(
SELLER_ID bigint
)
ROW FORMAT DELIMITED FIELDS TERMINATED BY '\177'
STORED AS SEQUENCEFILE
LOCATION '/tmp/kylin-c0812af1-76fa-410b-b782-cda4b0abd855/kylin_intermediate_cube_test_FULL_BUILD_c0812af1_76fa_410b_b782_cda4b0abd855';
SET mapreduce.job.split.metainfo.maxsize=-1;
SET mapreduce.job.max.split.locations=2000;
SET dfs.replication=2;
SET hive.exec.compress.output=true;
SET hive.auto.convert.join.noconditionaltask = true;
SET hive.auto.convert.join.noconditionaltask.size = 300000000;
INSERT OVERWRITE TABLE kylin_intermediate_cube_test_FULL_BUILD_c0812af1_76fa_410b_b782_cda4b0abd855
SELECT
TEST_FACT.SELLER_ID
FROM TEST_FACT
;
" 2>&1 | tee /tmp/kylin/logs/c0812af1-76fa-410b-b782-cda4b0abd855_0.log
[QuartzScheduler_Worker-8]:[2014-12-22 14:52:17,829][INFO][com.kylinolap.job.flow.JobFlowNode.execute(JobFlowNode.java:89)] - output:
Logging initialized using configuration in file:/etc/hive/conf.dist/hive-log4j.properties
OK
Time taken: 1.995 seconds
OK
Time taken: 0.524 seconds
FAILED: SemanticException [Error 10001]: Line 5:5 Table not found 'TEST_FACT'
```
- Hive execution error log ( see table name )
-
-
-
---------------- Imported from GitHub ----------------
Url: https://github.com/KylinOLAP/Kylin/issues/248
Created by: edyeo
Labels: bug,
Assignee: qhzhou
Created at: Mon Dec 22 16:48:24 CST 2014
State: open