I loaded the data with the timestamp field unsuccessful

[kex]

I loaded the data with the timestamp field unsuccessful,and timestamp field is null.

my sql:
carbon.sql("create TABLE IF NOT EXISTS test1 (date timestamp,id string) STORED BY 'carbondata' TBLPROPERTIES ('DICTIONARY_INCLUDE'='date','DATEFORMAT'='date:yyyy/MM/dd')")

carbon.sql("LOAD DATA inpath 'hdfs://myha/user/carbon/testdata/test4.csv' INTO TABLE test1 options('FILEHEADER'='date,id')")

my data test4.csv:
2017/3/23,2
2017/1/11,1
2017/9/17,3

when i select:
+----+---+
|date| id|
+----+---+
|null|  1|
|null|  2|
|null|  3|
+----+---+

i print time format is correct:
println(CarbonProperties.getInstance().getProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT))
yyyy/MM/dd

What could be the reason for it?
thx.

[David CaiQiang]

try yyyy/M/dd

Best Regards
David CaiQiang

[Liang Chen]

Hi

If the issue has be fixed?
BTW, you don't need add date column to DICTIONARY_INCLUDE, it do index for date/timestamp columns.

Regards
Liang

kex wrote

I loaded the data with the timestamp field unsuccessful,and timestamp field is null.

my sql:
carbon.sql("create TABLE IF NOT EXISTS test1 (date timestamp,id string) STORED BY 'carbondata' TBLPROPERTIES ('DICTIONARY_INCLUDE'='date','DATEFORMAT'='date:yyyy/MM/dd')")

carbon.sql("LOAD DATA inpath 'hdfs://myha/user/carbon/testdata/test4.csv' INTO TABLE test1 options('FILEHEADER'='date,id')")

my data test4.csv:
2017/3/23,2
2017/1/11,1
2017/9/17,3

when i select:
+----+---+
|date| id|
+----+---+
|null|  1|
|null|  2|
|null|  3|
+----+---+

i print time format is correct:
println(CarbonProperties.getInstance().getProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT))
yyyy/MM/dd

What could be the reason for it?
thx.

[kex]

This problem has been solved. I run the same sql to another cluster  is normal, may be the cluster of carbon.properties have problems.